Guia de Manejo de Nac en Adultos 65 Pags

doi:10.1136/thorax.56.suppl_4.iv1 2001;56;1-64 Thorax

Acquired Pneumonia in AdultsBTS Guidelines for the Management of Community

http://thorax.bmj.com/cgi/content/full/56/suppl_4/iv1Updated information and services can be found at:

These include:

References

http://thorax.bmj.com/cgi/content/full/56/suppl_4/iv1#otherarticles23 online articles that cite this article can be accessed at:

http://thorax.bmj.com/cgi/content/full/56/suppl_4/iv1#BIBLThis article cites 354 articles, 156 of which can be accessed free at:

serviceEmail alerting

top right corner of the article Receive free email alerts when new articles cite this article - sign up in the box at the

Notes

http://journals.bmj.com/cgi/reprintformTo order reprints of this article go to:

http://journals.bmj.com/subscriptions/ go to: ThoraxTo subscribe to

on 18 April 2008 thorax.bmj.comDownloaded from

http://thorax.bmj.com/cgi/content/full/56/suppl_4/iv1

http://thorax.bmj.com/cgi/content/full/56/suppl_4/iv1#BIBL

http://thorax.bmj.com/cgi/content/full/56/suppl_4/iv1#otherarticles

http://journals.bmj.com/cgi/reprintform

http://journals.bmj.com/subscriptions/

http://thorax.bmj.com

Synopsis

Scope of these guidelines (sections 1.1–1.7)+ These guidelines refer to the management of

adults with community acquired pneumonia(CAP) of all ages in the community or inhospital. They have been developed to applyto the UK healthcare system and popula-tion, but they should equally be applicableto any other countries which operate similarhealthcare services.

+ They are not aimed at patients with knownpredisposing conditions such as cancer orimmunosuppression admitted with pneu-monia to specialist units such as oncology,haematology, palliative care, infectious dis-eases units, or AIDS units.

+ They do NOT apply to the much largergroup of adults with non-pneumonic lowerrespiratory tract infection, including ill-nesses labelled as acute bronchitis, acuteexacerbations of chronic obstructive pulmo-nary disease (COPD), or “chest infections”.

+ Details of methods, the level of evidence,and grading of recommendations are givenin the text (sections 1.8–1.15) and appendi-ces and are summarised briefly for easyreference in table 1.

Synopsis of main summary pointsAETIOLOGY AND EPIDEMIOLOGY (SECTION 3)+ Only a small range of pathogens causes

CAP, with Streptococcus pneumoniae beingthe most frequent (tables 2–6, fig 3) [Ib].

+ The frequency of pathogens can vary inspecific patient groups. Mycoplasma and

legionella infections are less frequent in theelderly (box 1, fig 5) [Ib].

+ The low frequency of legionella, staphyloco-ccal, Chlamydia psittaci and Coxiella burnetiiinfection in patients with CAP, together withthe likely high frequency of the relevant epi-demiological risk factors in the generalpopulation (for example, recent travel orcontact with someone with an influenza typedisease) suggests that routine enquiry aboutsuch factors is likely to be misleading [IV].

+ Only in those with severe illness, where thefrequency of legionella and staphylococcalinfection is higher, may enquiry aboutforeign travel and influenza symptoms be ofpredictive value [IV].

+ Knowledge of increased mycoplasma activ-ity in the community during an epidemicperiod may help guide the clinician to theincreased likelihood of mycoplasma infec-tion (fig 4) [IV].

CLINICAL AND RADIOLOGICAL FEATURES

(SECTIONS 4, 5.1–5.3)+ The likely aetiological agent causing CAP

cannot be accurately predicted from clinicalor radiological features [II].

+ The term “atypical” pneumonia should beabandoned as it incorrectly implies thatthere is a characteristic clinical presentationfor patients with infection caused by “atypi-cal” pathogens [II].

+ Elderly patients with CAP more frequentlypresent with non-specific symptoms and areless likely to have a fever than youngerpatients [II].

+ Radiological resolution often lags behindclinical improvement from CAP, particularlyfollowing legionella and bacteraemic pneu-mococcal infection [III].

+ Radiographic changes caused by atypicalpathogens clear more quickly than thoseassociated with pneumonia caused by bacte-rial infection [III].

+ Radiological resolution is slower in theelderly and in cases where there is multilobeinvolvement [Ib].

MANAGEMENT

+ Figure 1 provides an algorithm for the man-agement of adult patients with CAP in thecommunity, and fig 2 provides an algorithmfor the management of adult patients withCAP in hospital (see over).

Table 1 Brief description of the generic levels of evidenceand guideline statement grades used*

Evidencelevel Definition

Guidelinestatementgrade

Ia A good recent systematic review of studiesdesigned to answer the question of interest

A+

Ib One or more rigorous studies designed toanswer the question, but not formallycombined

A–

II One or more prospective clinical studieswhich illuminate, but do not rigorouslyanswer, the question

B+

III One or more retrospective clinical studieswhich illuminate, but do not rigorouslyanswer, the question

B–

IVa Formal combination of expert views CIVb Other information D

*A fuller description is given in section 1 and in appendices 1–4.

Thorax 2001;56 (suppl IV) iv1

www.thoraxjnl.com



Has the patient got pneumonia? (section 1.7)

These guidelines do NOT apply

Consider alternative diagnosis and

management e.g. exacerbation of COPD

Non-severe pneumonia

Manage in hospital

NO

These guidelines apply

Assess severity (sections 6.4–6.6, figs 7, 8)

YES

• Antibiotic management. Initial empirical choice (tables 8, 9)• General management (section 7)• Investigations (section 5, table 7, fig 6)• Monitoring and medical review plan (sections 6.7, 7.3)

Severe pneumonia

Manage in hospital?Needs high dependency bed

Mild non-severe pneumonia

?Suitable for outpatientmanagement

Improvement

Review antibiotic durationand route (boxes 4, 5, table 10)

Discharge and follow up planning (section 5.3)Prevention/vaccination advice (section 10)

Failure to improve (box 6)Complications (table 12)

Specific pathogen identified

Review antibiotic choice (table 11)

Figure 2 Synopsis of the management of adult patients seen in hospital with suspected CAP.

Figure 1 Synopsis of the management of adult patients seen in the community with suspected CAP.

Has the patient most probably got pneumonia? (section 1.7)

These guidelines do NOT apply

Consider alternative diagnosis

and management

e.g. exacerbation of COPD

Refer to hospital without delayConsider starting antibiotics if patient severely ill or delay in hospital transfer (section 8.9)

• Start preferred antibiotic (tables 8, 9)• General management advice (section 7.1)• Arrange clinical review (section 7.2)

Failure to improve (box 6) Review need forinvestigations or

hospital referral (section 7.2)

NO

These guidelines apply

Can the patient be managed at home?

Assess severity (sections 6.4–6.6, figs 7, 8)Assess social issues

Review follow up plans after recovery or hospital discharge (section 5.3)Prevention/vaccination advice (section 10)

YES

NOYES

iv2 Thorax 2001;56 (suppl IV)

www.thoraxjnl.com



Synopsis of main recommendations

Investigations (section 5)GENERAL INVESTIGATIONS FOR PATIENTS MANAGED IN

THE COMMUNITY (SECTION 5.5)+ General investigations, including a chest

radiograph, are not necessary for the majority ofpatients with suspected community acquiredpneumonia (CAP) who are managed in thecommunity [C].

+ Out of hours and emergency general practitionerassessment centres should consider using pulseoximeters to allow for simple assessment ofoxygenation [D].

MICROBIOLOGICAL INVESTIGATIONS FOR PATIENTS

MANAGED IN THE COMMUNITY (SECTION 5.8)+ Microbiological investigations are not recom-

mended routinely [D].+ Examination of sputum should be considered for

patients who do not respond to empirical antibiotictherapy [D].

+ Examination of sputum for Mycobacterium tubercu-losis should be considered for patients with apersistent productive cough, especially if malaise,weight loss or night sweats, or risk factors for tuber-culosis (e.g. ethnic origin, social deprivation, theelderly) are present [D].

+ Serological investigations may be considered duringoutbreaks (e.g. Legionnaires’ disease) or epidemicmycoplasma years, or when there is a particularclinical or epidemiological reason [D].

GENERAL INVESTIGATIONS FOR PATIENTS ADMITTED TO

HOSPITAL (SECTION 5.6)+ All patients should have the following investiga-

tions performed on admission: chest radiograph[C]; full blood count [B–]; urea, electrolytes andliver function tests [C]; C reactive protein (CRP)when locally available [B–]; oxygenation assess-ment [C].

MICROBIOLOGICAL INVESTIGATIONS FOR PATIENTS

ADMITTED TO HOSPITAL (SECTIONS 5.7–5.9)+ It is not necessary or appropriate to perform a full

range of microbiological investigations on everypatient with CAP. The investigations performedshould be guided by the severity of pneumonia, epi-demiological risk factors, and the response to treat-ment (table 7) [D].

+ Blood culture is recommended for all patients, pref-erably before antibiotic treatment is commenced[D].

+ Sputum samples should be sent for culture andsensitivity tests from patients with non-severe CAPwho are able to expectorate purulent samples andhave not received prior antibiotic treatment. Speci-mens should be transported rapidly to the labora-tory [D].

+ Sputum cultures should also be performed forpatients with severe CAP or those who fail toimprove [D].

+ Laboratories should oVer a reliable Gram stain forpatients with severe CAP or complications, as onoccasions this can give immediate indication oflikely pathogens. Routine performance or reportingof sputum Gram stain on all patients is unnecessarybut can aid the laboratory interpretations of cultureresults [D].

+ Laboratories performing sputum Gram stainsshould adhere to strict and locally agreed criteriafor interpretation and reporting of results[B+].

+ Paired serological tests should be performed for allpatients with severe CAP, those who are unrespon-sive to â-lactam antibiotics, and for selectedpatients with particular epidemiological risk fac-tors or in whom a specific microbiological diagno-sis is important for public health measures (fig 6)[D].

+ Serological tests should be extended to all patientsadmitted to hospital with CAP during outbreaksand when needed for the purposes of surveillance.The criteria for performing serological tests inthese circumstances should be agreed locallybetween clinicians, laboratories, and public healthoYcers [D].

+ Pneumococcal antigen tests should be usedfor patients with severe CAP, if available locally[D].

+ Investigations for legionella infection arerecommended for all patients with severeCAP, for other patients with specific risk factors,and for all patients with CAP during outbreaks[D].

+ Rapid testing and reporting for legionella urineantigen should be available in at least one laboratoryper region [D].

+ Legionella culture should be specifically requestedby clinicians on laboratory request forms frompatients with severe CAP, or where legionellainfection is suspected on epidemiological grounds[D].

+ Legionella cultures should be routinelyperformed on invasive respiratory samples (e.g.obtained by bronchoscopy) from patients withCAP [D].

+ Serological assays with complement fixationtests (CFTs) are widely available and shouldremain the mainstay of diagnosis for atypicaland common respiratory viral pathogens[C].

+ Chlamydial antigen detection tests should be avail-able for invasive respiratory samples from patientswith severe CAP or where there is a strong suspicionof psittacosis [D].


www.thoraxjnl.com



+ The CFT remains the most suitable and practicalserological assay for routine diagnosis of respiratorymycoplasmal and chlamydial infections [B–]. Thereis no currently available serological test that canreliably detect infections due to Chlamydiapneumoniae.

Severity assessment (section 6)GENERAL RECOMMENDATIONS

+ Severity assessment is recommended as the key toplanning appropriate management both in thecommunity and in hospital [D].

+ Certain adverse prognostic features (detailedbelow) have been associated with an increasedrisk of death and should be assessed in all patients[A–].

+ None of the available predictive models or thealgorithms provided in these guidelines allow theunequivocal categorisation of patients into definiterisk groups and they should be regarded as an aid toclinical judgement, which is essential in assessingappropriate management [D]

+ Regular reassessment of severity during the courseof the illness is mandatory if management is to beadjusted appropriately [D].

ADVERSE PROGNOSTIC FEATURES

“Pre-existing” adverse prognostic features+ Age 50 years and over [Ib].+ Presence of coexisting disease [Ib].

“Core” clinical adverse prognostic features (CURB)+ Confusion: new mental confusion (defined as an

Abbreviated Mental Test score of 8 or less, see box2) [Ib].

+ Urea: raised >7 mmol/l (for patients being seen inhospital) [Ib].

+ Respiratory rate: raised >30/min [Ib].+ Blood pressure: low blood pressure (systolic blood

pressure <90 mm Hg and/or diastolic blood pres-sure <60 mm Hg) [Ib].

“Additional” clinical adverse prognostic features+ Hypoxaemia (SaO2 <92% or PaO2 <8 kPa) regard-

less of FiO2 [Ib]. Oxygen saturation measurementsmay be available to some general practitioners in thecommunity who have oximeters.

+ Bilateral or multilobe involvement on the chestradiograph [Ib].

IDENTIFYING THOSE PATIENTS WHO CAN USUALLY BE

SAFELY TREATED AT HOME (FIG 7)+ Patients who display no adverse prognostic features

are at low risk of death and do not normally requirehospitalisation for clinical reasons [D].

+ Patients who display two or more “core” adverseprognostic features are at high risk of death andshould be referred urgently to hospital [D].

+ For all other patients the decision to treat at homeor refer to hospital is a matter of clinical judgement[D].

+ When deciding on home treatment, the patient’ssocial circumstances and wishes must be taken intoaccount in all instances [D].

IDENTIFYING THOSE WITH SEVERE CAP IN HOSPITAL

(FIG 8)+ Patients who have two or more “core” adverse

prognostic features are at high risk of death andshould be managed as having severe pneumonia[A–].

+ Patients who display one “core” adverse prognosticfeature are at increased risk of death. The decisionto treat such patients as having severe ornon-severe pneumonia is a matter of clinicaljudgement, preferably from an experienced clini-cian. This decision can be assisted by considering“pre-existing” and “additional” adverse prognosticfeatures [D].

+ Patients who display no adverse prognostic featurescan be managed as having non-severe pneumoniaand may be suitable for outpatient treatment orearly hospital discharge [B+].

REVIEWING SEVERITY STATUS AFTER INITIAL ASSESSMENT

(SECTION 6.7)+ Regular assessment of disease severity is recom-

mended for all patients following hospital admis-sion. The “post take” round by a senior doctor andthe medical team provides one early opportunity forthis review [D].

+ All patients who display one or more “core” adverseprognostic features on admission should be re-viewed medically at least 12 hourly until shown to beimproving [D].

General management of CAP (section 7)IN THE COMMUNITY (SECTIONS 7.1–7.2)+ The need for hospital referral should be assessed

using the recommended severity criteria andclinical judgement [C].

+ Those with features of severe infection should beadmitted urgently to hospital [C].

+ Patients with suspected CAP should be advisednot to smoke, to rest, and to drink plenty of fluids[D].

+ Pleuritic pain should be relieved using simple anal-gesia such as paracetamol [D].

+ Nutritional supplements should be considered inprolonged illness [C].

+ Pulse oximetry, with appropriate training, shouldbecome increasingly available to general practition-ers for assessment of severity and oxygen require-ment for patients with CAP and other acuterespiratory illnesses [D].

+ Review of patients in the community with CAP isrecommended after 48 hours or earlier if clinicallyindicated. “Core” and “additional” adverse prog-nostic features should be assessed as part of theclinical review [D].


www.thoraxjnl.com



+ Those who fail to improve after 48 hours oftreatment should be considered for hospital admis-sion or chest radiography [D].

IN HOSPITAL (SECTION 7.3)+ All patients should receive appropriate oxygen

therapy with monitoring of oxygen saturations andFiO2 with the aim to maintain PaO2 >8 kPa andSaO2 >92%. High concentrations of oxygen cansafely be given in uncomplicated pneumonia[D].

+ Oxygen therapy in patients with pre-existingchronic obstructive pulmonary diseasecomplicated by ventilatory failure should beguided by repeated arterial blood gas measure-ments [C].

+ Patients should be assessed for volume depletionand may require intravenous fluids [C].

+ Nutritional support should be given in prolongedillness [C].

+ Temperature, respiratory rate, pulse, bloodpressure, mental status, oxygen saturation, andinspired oxygen concentration should bemonitored and recorded initially at least twicedaily and more frequently in those with severepneumonia or requiring regular oxygen therapy[C].

+ The CRP level should be remeasured [B–] and thechest radiograph repeated [C] in patients who arenot progressing satisfactorily.

ON THE INTENSIVE CARE UNIT (SECTION 7.4)+ Detailed recommendations are outside the scope of

these guidelines.+ Patients with CAP admitted to an intensive care

unit (ICU) should be managed by specialists withappropriate training in intensive care and respira-tory medicine [D].

+ Bronchoscopy can be valuable to remove retainedsecretions, obtain samples for culture for othermicrobiological investigations, and to excludeendobronchial abnormality [C].

FOLLOW UP PLANNING: WHEN TO REPEAT THE

CHEST RADIOGRAPH AND WHAT ACTION TO TAKE IF THE

RADIOGRAPH HAS NOT RETURNED TO NORMAL

(SECTIONS 5.2, 5.3, 7.5)+ The chest radiograph need not be repeated prior to

hospital discharge in those who have made asatisfactory clinical recovery [D].

+ At discharge or at follow up, patients should beoVered access to information about CAP such as apatient information leaflet [D].

+ Clinical review should be arranged for all patients ataround 6 weeks, either with their general prac-titioner or in a hospital clinic [D].

+ It is the responsibility of the hospital team toarrange the follow up plan with the patient and thegeneral practitioner [D].

+ A chest radiograph should be arranged at that timefor those patients who have persistent symptoms orphysical signs or who are at higher risk ofunderlying malignancy (especially smokers andthose over 50 years) [C].

+ In a patient who is improving clinically andfor whom there are no concerning clinical features,it will usually not be necessary to performfurther investigations just because radiologicalimprovement lags behind clinical recovery[B+].

+ Further investigations which may include bron-choscopy should be considered in patients withpersisting signs, symptoms, and radiological abnor-malities about 6 weeks after completing treatment[C].

Antibiotic management (section 8)EMPIRICAL ANTIBIOTIC CHOICE IN THE COMMUNITY

(TABLE 8)+ Amoxicillin remains the preferred agent but

at a higher dose than previously recommended[D].

+ A macrolide (erythromycin or clarithromycin) isoVered as an alternative choice and for thosepatients who are hypersensitive to penicillins[D].

+ For those patients referred to hospital withsuspected CAP, general practitioners may con-sider administering antibiotics immediately wherethe illness is considered to be life threatening orwhere there are likely to be delays (over 2 hours) inadmission [D].

EMPIRICAL ANTIBIOTIC CHOICE FOR ADULTS

HOSPITALISED WITH NON-SEVERE CAP (TABLE 9)+ Most patients can be adequately treated with oral

antibiotics [C].+ Combined oral therapy with amoxicillin and a mac-

rolide (erythromycin or clarithromycin) is preferredfor patients who require hospital admission forclinical reasons [D].

+ Oral monotherapy should be considered in the fol-lowing circumstances:

+ Amoxicillin monotherapy: (i) those previouslyuntreated in the community or (ii) thoseadmitted to hospital for non-clinical reasons whowould otherwise be treated in the community(e.g. the elderly or socially isolated) [D].

+ Monotherapy with a macrolide may be suitablefor patients who have failed to respond to anadequate course of amoxicillin prior toadmission. Deciding on the adequacy of priortherapy is difficult and is a matter of individualclinical judgement. It is therefore recommendedthat combination antibiotic therapy is thepreferred choice in this situation and that thedecision to adopt monotherapy is reviewed onthe “post take” round within the first 24 hours ofadmission [D].

+ When oral treatment is contraindicated, recom-mended parenteral choices include intravenousampicillin or benzylpenicillin, together with eryth-romycin or clarithromycin [D].


www.thoraxjnl.com



+ New fluoroquinolones are not recommended asfirst line agents or for community use for pneumo-nia, but may provide a useful alternative in selectedhospitalised patients with CAP [C].

+ A fluoroquinolone active against S pneumoniaeis an alternative regimen for those intolerantof penicillins or macrolides or where there arelocal concerns over Clostridium difficile associateddiarrhoea. However, experience with suchnewer fluoroquinolones in the treatment ofCAP and their interaction and side eVect profileis at present limited and further reportedexperience is required [B–]. Levofloxacin is theonly recommended agent currently licensed in theUK.

EMPIRICAL ANTIBIOTIC CHOICE FOR ADULTS

HOSPITALISED WITH SEVERE CAP (TABLE 9)+ Patients with severe pneumonia should be treated

immediately after diagnosis with parenteral antibi-otics [B–].

+ An intravenous combination of a broad spectrumâ-lactamase stable antibiotic such as co-amoxiclavor a second generation (e.g. cefuroxime) orthird generation (e.g. cefotaxime or ceftriaxone)cephalosporin together with a macrolide (e.g.clarithromycin or erythromycin) is preferred[C].

+ For those who are intolerant of â-lactam ormacrolide therapy or where there are localconcerns over C diYcile associated diarrhoea, afluoroquinolone with enhanced activity against Spneumoniae together with intravenous benzyl-penicillin is oVered as an alternative [D].Levofloxacin is currently the only such fluoro-quinolone licensed in the UK.

ROUTE OF ANTIBIOTIC ADMINISTRATION (BOXES 4 AND 5)+ The oral route is recommended in those with non-

severe pneumonia admitted to hospital pro-vided there are no contraindications to oral therapy[B+].

+ Patients treated initially with parenteral anti-biotics should be transferred to an oral regimenas soon as clinical improvement occurs and thetemperature has been normal for 24 hours, provid-ing there is no contraindication to the oral route[B+].

+ The choice of route of administration should bereviewed initially on the “post take” round and thendaily [D].

+ Ward pharmacists could play an important role infacilitating this review by highlighting prescriptioncharts where parenteral antibiotic treatment contin-ues [D].

DURATION OF ANTIBIOTIC ADMINISTRATION (TABLE 10)+ For patients managed in the community and

most of those admitted to hospital with non-severeand uncomplicated pneumonia, treatment withappropriate antibiotics for 7 days is recommended[C].

+ For patients with severe microbiologically unde-fined pneumonia, 10 days of treatment is proposed.This should be extended to 14–21 days wherelegionella, staphylococcal, or Gram negative en-teric bacilli pneumonia are suspected or confirmed[C].

FAILURE TO IMPROVE (TABLE 12, BOX 6)+ For patients who fail to improve as expected, there

should be a careful review by an experienced clini-cian of the clinical history, examination, prescrip-tion chart, and results of all available investigationresults [D].

+ Further investigations, including a repeat chestradiograph, CRP and white cell count, and furtherspecimens for microbiological testing should beconsidered in the light of any new information afterthe clinical review [D].

+ When a change in empirical antibiotic treatmentis considered necessary, a macrolide could besubstituted for or added to the treatment for thosewith non-severe pneumonia treated with amoxicil-lin monotherapy in the community or in hospital[C].

+ For those with non-severe pneumonia in hospital oncombination therapy, changing to a fluoroqui-nolone with eVective pneumococcal cover is anoption [C].

+ The addition of rifampicin may be considered forthose with severe pneumonia not responding tocombination antibiotic treatment [C].

ANTIBIOTIC THERAPY WHEN A SPECIFIC PATHOGEN HAS

BEEN IDENTIFIED (TABLE 11)+ If a specific pathogen has been identified, the

guidelines recommend specific antibiotic options[C].

Prevention: vaccination strategies (section 10)INFLUENZA VACCINATION (SECTION 10.2)+ Influenza vaccination is recommended for those at

“high risk” of mortality from influenza or compli-cating pneumonia [C].

+ These “high risk” groups include those withchronic lung, heart, renal and liver disease,diabetes mellitus, immunosuppression due todisease or treatment, and those aged over 65 years[C].

+ Influenza vaccine is contraindicated for those withhypersensitivity to hens’ eggs [C].

PNEUMOCOCCAL VACCINATION (SECTION 10.3)+ Pneumococcal vaccination is recommended by the

Departments of Health for all those aged 2 years orolder in whom pneumococcal infection is likely tobe more common or serious, although there is noevidence that it is eVective in preventing CAP insuch “at risk” groups [A+].

+ Pneumococcal vaccine should not be given duringacute infection and is not recommended duringpregnancy. Re-immunisation within 3 years iscontraindicated [C].

+ Pneumococcal and influenza vaccines can be giventogether at diVerent sites [A–].


www.thoraxjnl.com



1 Introduction and methods

1.1 IntroductionCommunity acquired pneumonia (CAP) iscommon. It is associated with significant mor-bidity, mortality and utilisation of healthservice resources (see section 2). Changes overthe last decade suggest that it is appropriate toreview the UK guidelines for the managementof CAP published in 1993.

1.2 Available management guidelines forCAPRecognising the clinical importance of CAP,numerous countries have developed nationalguidelines in the last decade, including theUK,1 USA,2 3 Canada,4 France,5 and Italy,6

among others. Those from the British ThoracicSociety (BTS)1 and the American ThoracicSociety3 received much publicity and diVeredwith regard to likely aetiology and antibioticmanagement.

More recently the Infectious Disease Societyof America2 and the Canadian InfectiousDiseases Society working with the CanadianThoracic Society4 have updated their recom-mendations for North America, and the Cent-ers for Disease Control and Prevention of theUSA have also published their own recommen-dations, specifically regarding the treatmentand control of drug resistant Streptococcuspneumoniae.7 The European Respiratory Soci-ety8 has also provided a review of the manage-ment of lower respiratory tract infections,including pneumonia, to cover Europe. (Sincefinalising the contents of these guidelines, theAmerican Thoracic Society has also publishedits own revised guidelines for North America.9)

However, it may be unwise to extrapolatenational guidelines from other countries to theUK for a number of reasons:

(1) They should depend on the results ofgood CAP studies performed within thatcountry. There are substantial diVerences instudy results from diVerent parts of the world(see section 3).6 10

(2) Diagnosis and management of CAP willbe influenced by the organisation of the healthsystem. In common with some other countries,the UK has a network of National HealthService general practitioners who are the firstport of call for most people with an acuteillness and without cost to the patient. Theywill often diagnose and manage a patient with-out any investigations or secondary care advice.General practitioners usually only refer pa-tients to hospital either just for a chestradiograph or because of clinical concern andthe likely need for admission.

(3) By contrast, other health services such asin the USA depend on primary care physiciansor specialists, some operating on a fee for serv-ice basis and usually operating from wellequipped clinics or emergency rooms with easyand immediate access to radiology and otherinvestigations. In such circumstances mostpatients with suspected pneumonia will be

diagnosed by chest radiography and investi-gated before deciding on optimum manage-ment either as an outpatient or by admission tohospital.

(4) There is a diVerence in antimicrobialagents licensed and available in diVerent coun-tries. For instance, levofloxacin is the only fluo-roquinolone with some enhanced pneumococ-cal activity licensed and available in the UK atthe time of preparing these guidelines, whereasother countries also have available neweragents such as moxifloxacin.

1.3 What problems have become apparentwith the 1993 BTS CAP guidelines?Firstly, these were consensus guidelines1 whichrelied on UK CAP studies performed in theearly 1980s,11–15 all of which had design limita-tions which question the application of theirresults to current UK practice.

Secondly, they have also been criticised fornot clearly diVerentiating between CAP andnon-pneumonic lower respiratory tract infec-tions such as acute exacerbations of chronicobstructive pulmonary disease and thus en-couraging overtreatment of some patients, par-ticularly the elderly. The severity criteriaproposed by the 1993 BTS guidelines resultedin many patients both with and without pneu-monia being identified as severely ill. Thesefactors have been blamed for the dramaticincrease in the use of intravenous broadspectrum antibiotics, combination antibiotictherapy, antibiotic costs, and side eVects,particularly Clostridium diYcile associateddiarrhoea.16–18

New guidelines must try to address theseareas of misunderstanding. Unfortunately,there have been no comprehensive, prospectivehospital studies of CAP published in the UKsince the early 1980s, although there have beena number of focused hospital and communitystudies to address specific areas.13 19–23 Whileproviding useful information about particularfacets of CAP, such studies do not allow a bal-anced overview of its current aetiology in theUK.

By contrast, CAP continued to be activelystudied in other countries during the 1990s.24–29

These data suggest that the importance ofS pneumoniae infection is waning as thefrequency of newer pathogens such as Chlamy-dia pneumoniae and “non-penicillin responsive”pathogens such as the aerobic Gram negativeenteric bacilli increase, including the Enterobac-teriaceae and Pseudomonas aeruginosa.

1.4 What changes have happened in thearea of CAP in the last two decades?Over the last decade there is perceived to havebeen a change in the pattern of adult CAP inthe UK. This has resulted in a pressure to alterthe empirical management of CAP to newbroader spectrum antibiotics and also combi-nation antibiotic therapy.


www.thoraxjnl.com



These perceived changes include:(1) Increasing emphasis on new pathogens,

particularly C pneumoniae and Legionella spe-cies and on older pathogens such as Moraxellacatarrhalis and Gram negative enteric bacilli,but the clinical relevance to UK practice needsto be clarified (see section 3).

(2) The increasing age of the population,many with co-morbid illness, and the markedincrease in the use of residential and nursinghomes over the last few years is perceived toprovide an expanding “at risk” population forrespiratory pathogens such as Gram negativeenteric bacilli30 who therefore require broadspectrum antibiotic therapy (see section 3).

(3) The admission of patients for non-medical reasons (such as inadequate socialsupport for the elderly) and the common use ofantibiotics for minor respiratory illness in thecommunity31 may encourage the hospital doc-tor to use broad spectrum antibiotics becauseof the perceived problem of either failed com-munity therapy or antibiotic resistant bacteriain hospitalised patients who have not re-sponded to initial antibiotics.

(4) A change in delivery of acute medicalservices in most hospitals, with integration ofadult and geriatric medicine and the use ofassessment and admissions wards and loss ofcontinuity of care throughout the admission,has encouraged the use of a “belt and braces”management strategy for all types of patientswith CAP in order to cover all pathogens and“be safe”, with an emphasis on parenteral andcombination antibiotic therapy. This occursespecially in the elderly who are perceived to beat increased risk of complicated bacterial infec-tions, but also at risk of antibiotic related mor-bidity including C diYcile associated diarrhoea(see sections 3, 6 and 8).

(5) Patients with CAP will usually beassessed and admitted by relatively inexperi-enced medical trainees which may contributeto overtreatment “to be safe”.32 It is nowrecommended that the “consultant post take”round within 24 hours of admission is a part ofgood medical practice33 [IVb] and this valuableresource needs to be integrated into guidelinepractice for the management of patients hospi-talised with pneumonia (section 7).

(6) Reports of increasing antibiotic resist-ance of common respiratory pathogens haveproduced recent worries about antibiotic re-sistance in the UK from Government expertcommittees.34 35 Such concerns have alreadyinfluenced antibiotic prescribing in severalEuropean countries and in North America.36

The issue of penicillin resistant pneumococci isparticularly relevant. It is important to reviewwhether such changes in antibiotic resistanceare now clinically important in UK practice37

and in countries with similar characteristics(see section 8).

(7) The marketing of newer antibiotics (par-ticularly macrolides and fluoroquinolones) isexposing doctors to increasing pressure to usethem “to cover all likely pathogens” because ofconcerns about changes in pathogens, anti-biotic resistance, and the “at risk” population.Some of these compounds may appear to have

attractive properties for the management ofCAP, but guidance is needed on the strength ofthe available comparative data of these newerantibiotics (see section 8). In these guidelineswe only consider antibiotics licensed and avail-able in the UK at the time of preparation of thedocument.

(8) Newer serological, molecular biological,and antigen detection techniques are nowavailable for the diagnosis of viral, atypical, andbacterial pathogens—for example, urine anti-gen detection which has improved diagnosis oflegionella infections—and guidance is neededon when to request such tests (see section 5).

(9) There has been a dramatic increase in theuse of both influenza vaccine (1.3 million dosesin 1980/81 and 7 million in 1997/8) and pneu-mococcal vaccine (5000 doses in 1989 and750 000 in 1997) in the community which mayhave influenced the aetiology of communityacquired pneumonia (data provided by Dr JaneLeese, Department of Health). Advice isneeded on the value of preventative strategies(see section 9 on guidelines on vaccination).

Summary+ Up to date guidelines are needed to review

the current data and to assess the impact, ifany, of these changes on new managementguidelines of CAP in the UK.

1.5 What is the target end user audience?We want these guidelines to be of value to:+ hospital based medical and other staV

involved with managing adult patients withcommunity acquired pneumonia;

+ general practitioners;+ those teaching the subject at both under-

graduate and postgraduate level.The guidelines have been developed to apply

to the UK healthcare system and population,but they should also be of value to other coun-tries which operate similar healthcare serviceswith appropriate modification to take intoaccount diVerences in licensing and availabilityof antimicrobial agents.38

1.6 What patient populations are includedand excluded?Our guidelines address the management ofunselected adults with CAP who are managedby their general practitioner or admitted tohospital as an emergency.

They are not aimed at the much largergroup of adults with non-pneumonic lowerrespiratory tract infection, including ill-nesses labelled as acute bronchitis, acuteexacerbations of chronic obstructive pul-monary disease (COPD), or “chest infec-tions”.

Although there are similarities in the princi-ples of management between pneumonic lowerrespiratory tract infection (CAP) and non-pneumonic lower respiratory tract infection,there are diVerences in the aetiology, severityassessment, management, and outcome. Rec-ommendations for the antibiotic managementof acute exacerbations of COPD are included


www.thoraxjnl.com



in the published BTS guidelines on themanagement of chronic obstructive pulmonarydisease.39

We do not consider the management ofpneumonia in:(1) Patients where the pneumonia is anexpected terminal event or who are known tohave lung cancer, pulmonary tuberculosis,cystic fibrosis, primary immune deficiency orsecondary immune deficiency related to HIVinfection, or drug or systemic disease inducedimmunosuppression. We do include patientsreceiving oral corticosteroid therapy as this is anot uncommon situation for patients admittedon medical “take”.(2) Patients who have been in hospital withinthe previous 10 days and may have hospitalacquired pneumonia. Patients admitted fromhealthcare facilities such as nursing homes andresidential homes will be commented on sepa-rately.(3) Children with CAP. Such guidelines arebeing published by the BTS guidelines com-mittee on pneumonia in children.

Summary+ The guidelines are aimed to aid the

management of unselected adults with CAPseen by a general practitioner or admitted tohospital as an emergency, on general medi-cal “take”, in the UK or in other countrieswith similar care systems.

+ They are not aimed at patients with knownpredisposing conditions admitted withpneumonia to specialist units such as oncol-ogy, haematology, palliative care, infectiousdiseases or AIDS units.

1.7 DefinitionsDEFINITION OF CAP

The diagnosis in hospital will be made with thebenefit of a chest radiograph. In the commu-nity the recognition and definition of CAP bygeneral practitioners in the UK, without thebenefit of investigations or radiology, posesgreater problems and the diagnosis will beinvariably based only on clinical features.

DEFINITION OF CAP IN A COMMUNITY SETTING

The clinical definition of CAP that has beenused in community studies has varied widelybut has generally included a complex of symp-toms and signs both from the respiratory tractand regarding the general health of thepatients. Features such as fever (>38°C), pleu-ral pain, dyspnoea, and tachypnoea and signson physical examination of the chest (particu-larly when new and localising) seem most use-ful when compared with the gold standard ofradiological diagnosis of CAP38 [II].

Woodhead et al 40 found that 39% of adultstreated with antibiotics for an acute lower res-piratory tract infection associated with newfocal signs on chest examination had evidenceof CAP on chest radiograph compared with 2%of patients who did not have new focal chestsigns [II]. By contrast, Melbye et al 41 foundthat respiratory symptoms and signs were ofonly minor value in diVerentiating patientswith radiographic pneumonia in a study of 71

patients suspected by their general practition-ers of having CAP [II]. The clinical findingsreported by the general practitioners to bemost suggestive to them of CAP (typicalhistory of cough, fever, dyspnoea, chest pains,and lung crackles on examination) had lowpredictive values; only a short duration ofsymptoms (less than 24 hours) was of signifi-cant predictive value.

Various prediction rules have been publishedfor the diagnosis of CAP [II] but have generallyshown the need for confirmatory radiographicevidence. Statistical modelling was used byDiehr et al 42 to predict the presence of CAP in1819 adults presenting as hospital outpatientswith acute cough, 2.6% of whom had CAP onchest radiographic examination [II]. The pres-ence of fever (>37.8°C), raised respiratory rate(>25 breaths/min), sputum productionthroughout the day, myalgia and night sweats,and absence of sore throat and rhinorrhoeawere the only clinical features that predictedCAP when included in a diagnostic rule whichhad a sensitivity of 91% and a specificity of40%.

For the purposes of these guidelines CAP in thecommunity has been defined as:+ symptoms of an acute lower respiratory tract

illness (cough and at least one other lowerrespiratory tract symptom);

+ new focal chest signs on examination;+ at least one systemic feature (either a symp-

tom complex of sweating, fevers, shivers,aches and pains and/or temperature of 38°Cor more);

+ no other explanation for the illness, which istreated as CAP with antibiotics.

DEFINITION OF CAP IN PATIENTS ADMITTED TO

HOSPITAL (WHEN CHEST RADIOGRAPHY IS

AVAILABLE)Studies of CAP from diVerent countries haveused very diVerent definitions and inclusioncriteria10 38 43; most have required a combina-tion of symptoms, signs, and radiologicalfeatures. The BTS study of CAP used a defini-tion which included an acute illness withradiographic shadowing which was at least seg-mental or present in more than one lobe andwas not known to be previously present or dueto other causes.14 Like most studies, cases wereexcluded if pneumonia occurred distal to aknown carcinoma or foreign body.

For the purposes of these guidelines CAP in hos-pital has been defined as:+ symptoms and signs consistent with an

acute lower respiratory tract infection asso-ciated with new radiographic shadowing forwhich there is no other explanation (e.g. notpulmonary oedema or infarction);

+ the illness is the primary reason for hospitaladmission and is managed as pneumonia.

DEFINITION OF THE TERMS “ATYPICAL

PNEUMONIA” AND “ATYPICAL PATHOGENS”Another issue relevant to the aetiology, diagno-sis, management, and prognosis of CAP is theuse of the descriptive term “atypical” whendescribing pneumonia or groups of pathogens.The term “atypical pneumonia” has outgrown


www.thoraxjnl.com



its historical usefulness and we do not recom-mend its continued use as it implies, incor-rectly, a distinctive clinical pattern (see section4.2).

We do, however, use the term “atypicalpathogens” which, for the purposes of theseguidelines, are defined as infections caused byMycoplasma pneumoniae, C pneumoniae, Cpsittaci, and Coxiella burnetii. These pathogensare characterised by being diYcult to diagnoseearly in the illness and are sensitive to antibiot-ics other than beta-lactams such as macrolides,tetracyclines, or fluoroquinolones. They areconcentrated intracellularly which is the usualsite of replication of these pathogens. We con-clude that the term “atypical pathogens” is stilluseful to clinicians in guiding discussion aboutaetiology and management of CAP.

Legionella species, although sharing some ofthese characteristics, are not considered to bean “atypical pathogen” for the purpose of thisdocument as there are diVerent species andthese can be acquired both in the communityand hospital environment.

DEFINITION OF THE TERM “ELDERLY”There is no agreed age cut oV to define theterm “elderly” and published guidelines haveused very diVerent definitions.2 3 5–7 Whenreferring to published research, whereverpossible we define the age limits that are usedin studies of CAP in older patients. Whenmaking recommendations we arbitrarily usethe term “elderly” to include those adults aged75 years and over.

1.8 Guidelines committee membershipAs recommended by guideline developers,44 45 arepresentative group of clinical specialists, amethodologist, and an information specialistwere selected to join the committee. Theseincluded:+ four practising general physicians with a

special interest in respiratory medicine andactive research interest in respiratory infec-tion (DH, GD, JTM, MAW)

+ two general practitioners (PS, WFH)+ one clinical microbiologist (TB)+ two physicians with an interest in infectious

diseases (DN, RGF)+ a specialist registrar trainee in general and

respiratory medicine (WSL)+ a clinical epidemiologist/health services re-

searcher (JW)+ a medical librarian/information scientist

(RM)

1.9 Involvement with other groupsPS and WFH also acted as mandated repre-sentatives for the Royal College of GeneralPractitioners, RGF for the British InfectionSociety, DN for the British Society forAntimicrobial Chemotherapy, and GD andJTM for the Standards of Care Committee ofthe British Thoracic Society.

Designated representatives of the followingprofessional groups later provided formalreview, comments on, and endorsement of thedraft guidelines synopsis: Royal College ofPhysicians of London (including the Clinical

EVectiveness and Evaluation Unit and theAdvisory Committees on General InternalMedicine, Respiratory Medicine and Geriat-rics), the Public Health Laboratory Service(including the Committee on Respiratory andSystemic Infections and the Primary CareCoordinating Committee), the British Geriat-rics Society (including the BGS ExecutiveCommittee, the Policy Committee and theSpecial Interest Group in Respiratory Medi-cine), and the British Lung Foundation. TheLower Respiratory Tract Infection GuidelinesCommittee of the Scottish IntercollegiateGuidelines Network (SIGN) provided formalreview and comments on the draft as part ofthe peer review process.

As CAP is usually an acute self-limiting dis-ease we did not feel that patient involvement onthe committee was appropriate. However, wedid involve a group of 200 patients who hadrecently recovered from CAP in validating andrevising a British Lung Foundation patientinformation leaflet on pneumonia46 [III]. Theleaflet, revised during 2001, is available onrequest from the British Lung Foundationheadquarters (78 Hatton Gardens, LondonEC1N 8LD, UK) and regional oYces.

1.10 Scope of the task and questionsettingThe broad remit of the group was determinedby the BTS Standards of Care Committee andincluded producing updated and evidencebased guidelines for the management of CAPin adults over 16 years for the UK. The grouprefined this remit by considering documentedproblems in the current management of CAPboth in primary care and in hospital practiceand issues arising from previously publishedguidelines for CAP, such as the 1993 BTSguidelines.1

A postal questionnaire was sent to one con-sultant respiratory physician at each of the 263UK hospitals listed in the BTS Directory ofRespiratory Services enquiring about localwritten guidelines for the empirical manage-ment of CAP and their views about the 1993BTS CAP guidelines. There were 215 re-sponses (82%) which were reviewed anddiscussed by the committee when assessingquestion setting and guideline development.47

From these sources the nine broad clinicalareas listed below were identified. Each issuewas expanded into specific clinical questionswhich were structured to facilitate easy litera-ture searches.48

1.11 Literature search, assessmentstrategy, and critical appraisalLITERATURE SEARCH

Search strings developed by North ThamesRegional Library service and published on thewebsite (www.nthames-health.tpmde.ac.uk/evidence_strategies.htm) were adapted to ourpurpose by combining them with free text andkey word terms for CAP to produce 16 searchstrategies. These were applied to locate allEnglish language studies relevant to theaetiology, diagnosis, severity staging, investiga-tion, prognosis, complications, or treatment of


www.thoraxjnl.com



CAP in adults over 16 years. Initial searcheswere conducted on Medline (1966 onwards),Embase (1980 onwards), and the CochraneLibrary in February 1998. These searches wererepeated in May 1999 and again in January andSeptember 2000 (on the latter occasionssupplemented by a search of the NationalLibrary of Medicine PreMedline database fea-turing articles not yet fully indexed). A lowyield of relevant references in sections on anti-biotic management, non-antibiotic manage-ment, and complications led to a series of addi-tional searches being conducted for thesesections.

ASSESSMENT OF RELEVANCE

One individual (WSL) read the title andabstract of each article retrieved by theliterature searches and decided whether thepaper was definitely relevant, possibly relevant,or not relevant to the project. For each paper inthe first and second categories the full paperwas ordered and allocated to the relevantsection(s). During the 2 year period of theproject a total of 578 papers were judged to berelevant and circulated for critical appraisal.

CRITICAL APPRAISAL

At least two clinical experts were identified foreach of the main clinical topic areas:+ Incidence and mortality (JTM, WSL)+ Aetiology and epidemiology (MAW, GD)+ Clinical and radiological features (DH, TB)+ General and microbiological investigations

(DH, TB)+ Severity assessment (WSL, JTM)+ General management in hospital (GD,

WSL, MAW)+ Antibiotic therapy (RGF, DH, DN)+ Complications and failure to improve (GD,

WSL)+ Prevention including vaccination (GD,

WFH)Each expert independently judged the clini-

cal relevance and scientific rigour of each paperassigned to them using generic study appraisalchecklists (see appendices 1 and 2) adaptedfrom published checklists.45 49–51 With the widerange of study designs relevant to the ques-tions,52 no single set of pass/fail criteria couldbe applied across all topic areas. Subject areaexperts determined the relevance of studies toCAP in the UK by examining the patientgroups studied, the healthcare setting (primarycare, etc), and the healthcare system where thismight influence patient case mix at presenta-tion.

Experts sorted studies into those relevant totheir own topic area, those relevant to otherareas (which were copied on to experts in otherrelevant sections), or those not relevant. Theapproximate agreement rates between asses-sors for this process was between 70% and 90%for diVerent sections. They also added relevantstudies from their own knowledge and personalresearch reference lists.

Experts individually assessed the literatureselected and wrote a short document describ-ing study findings and answering specificpre-identified questions. These documents

were discussed by the whole committee toidentify areas of overlap and gaps. Given theheterogeneous nature of the patient groups,outcomes, interventions, and tests studied, noformal meta-analysis was carried out.

The reliability of the evidence in each studywas graded from Ia to IVb using a new genericlist of evidence levels (see appendix 3)developed from existing insights and check-lists.44 53 Disagreements were resolved by dis-cussion. Where relevant, individual referencesused in this document are followed by an indi-cation of the evidence level in square brackets.

1.12 Grading of recommendations anddrafting of guidelinesGRADING OF RECOMMENDATIONS

Recommendations were graded from A+ to Das indicated by the strength of the evidence aslisted in the table in appendix 4.

DRAFTING OF GUIDELINES

Once the outline guideline sections were com-plete and summarised, they were circulated torepresentatives of the professional bodies givenabove for comments.

A detailed guideline synopsis was circulatedto the membership of the BTS prior to its win-ter 1999 conference. The guidelines werediscussed in a plenary session at the conferenceand, in addition, 213 structured feedbackforms from BTS members were analysed.

Revisions were made, which also included anupdated literature search performed in Febru-ary and September 2000, and the documentcontent was finalised by the end of 2000. Thus,antibiotics which were not licensed for use inthe UK by the end of 2000 could not beincluded in our recommendations. Similarlyrelevant products that are not licensed in theUK but are available in other countries are notreviewed. Updates to include any new relevantinformation or products are planned for thefuture.

1.13 Plans for updating these guidelinesFollowing the BTS protocol for guidelinesrevisions, the committee will meet on anannual basis and review new published evi-dence obtained from a structured literaturesearch, comment on any newly licensed andrelevant antibiotics, and issue guideline up-dates or revisions as necessary. Importantchanges will be posted on the BTS website(www.brit-thoracic.org.uk). The membershipof the guideline committee will change overtime on a rolling programme, dictated by theBTS Standards of Care Committee policy forguideline committee membership.

1.14 Auditing management of CAPThe management of CAP is a suYciently com-mon and important issue to warrant the devel-opment of audit measures of the process of careand outcome to evaluate the quality of care forCAP using guidelines as a standard of manage-ment. There is evidence from Hiriani andMacfarlane22 [III], Gleason et al 54 [II], and


www.thoraxjnl.com



Gilbert et al 55 [II] that guidelines do guide andstandardise management, but with less meas-urable eVect on outcome.

The issue of choosing quality indicators andaudit tools for CAP has recently been exten-sively reviewed 56[III]. With guidance from thisreview, an audit tool has been developed by thecommittee, refined by the BTS Audit Com-mittee, and tested in pilot hospitals. This willbe made available through the BTS website(www.brit-thoracic.org.uk).

1.15 Implementation of the guidelinesWe expect that these guidelines will act as aframework for local development or modifica-tion of protocols after discussion with local cli-nicians and management. The subsequent dis-semination, implementation, and evaluation of

these guidelines should be undertaken by thehospital Quality and Clinical EVectivenessgroup in conjunction with relevant committeessuch as those responsible for therapeutics,antibiotic prescribing, or protocol develop-ment. Countries with similar health servicesystems will also find the framework of value,adapting the guidelines to take into accountany relevant national diVerences in diseasepresentation and the availability of investiga-tions and antimicrobial agents.

For maximal long term impact, the guide-lines should be the subject of continuingeducation and quality improvement activities.Production of summary statements, algo-rithms, pocket sized reminders, wall charts,and material specific to primary care will behelpful for both individual doctors and depart-ments working with patients with CAP.


www.thoraxjnl.com



2 Incidence, mortality and economicconsequences

2.1 How common is adult CAP in thecommunity and in hospital?Prospective population studies from the UK40

[II], Finland 57 [Ib], and North America58 [Ib]have reported an annual incidence of CAPdiagnosed in the community of between 5 and11 per 1000 adult population. Pneumonia,diagnosed clinically by general practitioners,accounts for only 5%40 [Ib] to 12%59 [Ib] of allcases of adult lower respiratory tract infectiontreated with antibiotics by general practitionersin the community in the UK.

The incidence varies markedly with age,being much higher in the very young and theelderly. In the Finnish study the annualincidence in the 16–59 age group was 6 per1000 population, 20 for those aged 60 yearsand over, and 34 per 1000 population for thoseaged 75 and over57 [Ib]. A similar pattern wasreported from Seattle, USA58 [Ib].

Population based studies of the incidence ofCAP requiring hospitalisation have reportedoverall incidences of 1.1 per 1000 adult popu-lation per annum in Canada60 [Ib], 2.6 per1000 in Spain25 [II], 2.7 per 1000 populationin Ohio, USA61 [Ib], and 4 per 1000 popula-tion in Pennsylvanian hospitals, USA62 [III].

The proportion of adults with CAP whorequire hospital admission in the UK has beenreported as being between 22%40 [Ib] and42%63 [III]. This figure varies in other coun-tries, probably depending on the structure ofthe primary and secondary healthcare systems.In a Finnish prospective longitudinal popula-tion study 42% were admitted to hospital57 [Ib].A 50% admission rate was reported in onestudy from Spain but this only includedpatients referred by their general practitioner tothe hospital emergency service for confirma-tion of the diagnosis of CAP25 [II].

In Seattle, USA 15% were hospitalised64 [Ib].In the Pneumonia Patient Outcomes researchmulticentre, prospective cohort study of CAPin America, 41% of adults studied weremanaged initially as outpatients and theremainder were admitted to hospital. Of thoseinitially treated as outpatients, only 7.5% weresubsequently admitted, 56% because of theCAP and the rest because of worsening of acomorbid illness65 [Ib].

The proportion of adults admitted to hospi-tal with CAP who require management on anICU varies from 5% in the BTS multicentrestudy66 [II] to 10% in a Spanish study [II].67

Between 8%25 [II] and 10%21 [III] of medicaladmissions to an ICU are for severe CAP.

Summary+ The annual incidence in the community is

5–11 per 1000 adult population [Ib].+ CAP accounts for 5–12% of all cases of

adult lower respiratory tract infection man-aged by general practitioners in the commu-nity [Ib].

+ The incidence varies markedly with age,being much higher in the very young and theelderly [Ib].

+ The incidence of CAP requiring admissionto hospital varies between 1.1 and 4 per1000 population [Ib].

+ Between 22% and 42% of adults with CAPare admitted to hospital [Ib].

+ Between 5% and 10% of adults admitted tohospital with CAP are managed on an ICU[II].

2.2 What is the mortality of CAP?The reported mortality of adults with CAPmanaged in the community is low at less than1%25 [II], 40 [Ib], 65 [Ib]. Deaths in the commu-nity due to CAP are rare in the UK. In onestudy only seven cases were identified by coro-ners’ post mortem examinations over 1 year inNottingham, a large urban city of threequarters of a million, giving an incidence of 1per 100 000 68 [III].

The reported mortality of adults admitted tohospital with CAP has varied widely. The BTSmulticentre study reported a mortality of5.7%14 [II] but did not study patients over theage of 74 years. Other UK studies havereported mortalities of 8%12 [II], 12%11 [Ib],and 14%69 [Ib]. Countries with similar health-care systems have reported hospital mortalityrates of 4%57 [Ib], 7%70 [II], 8%71 [Ib], and10%72 [Ib]. Mortality figures from NorthAmerican hospital studies have tended to behigher, probably because more patients withCAP are provided with ambulatory care asoutpatients and only those with more severepneumonia or co-morbid disease are admittedto hospital.

The mortality of patients with severe CAPrequiring admission to an intensive care unit(ICU) is high. This is likely to be particularlyevident in health services such as the NationalHealth Service where ICU beds are at apremium such that only critically ill patients inneed of assisted ventilation can be admitted.ICU based studies in the UK report mortalityrates of over 50%21 [III], 22 [III], 66 [III],73 [III]. Nearly all of the patients requiredassisted ventilation. By contrast, the mortalityrate in a large multicentre study of severe CAPin four French ICUs reported a mortality rateof 35% with a ventilation rate of only 52%74

[Ib]. Similar figures were reported fromanother ICU based study in France75 [II]. In aspecialist ICU in Spain a mortality rate of 22%was reported, rising to 36% in the 61% ofpatients who required assisted ventilation67

[II].

Summary+ The reported mortality rate of adults with

CAP managed in the community in the UKis very low at less than 1% [Ib].


www.thoraxjnl.com



+ The reported mortality rate of adults admit-ted to hospital with CAP in the UK has var-ied between 5.7% and 12% [Ib].

+ The mortality rate of patients with severeCAP requiring admission to an ICU in theUK is high at over 50% [III].

2.3 What are the economic consequencesof CAP?A prevalence based burden of illness studyestimated that CAP in the UK incurred a directhealthcare cost of £441 million annually at1992–3 prices. The average cost for managingpneumonia in the community was estimated at£100 per episode compared with £1700–5100for hospitalised patients. Hospitalisationaccounted for 87% of the total annual cost63

[III].A similar exercise for the USA calculated

that annual CAP costs amounted to $8.4billion, 52% of the costs being for the inpatientcare of 1.1 million patients and the remainingcosts for the 4.4 million outpatient consulta-tions. The average length of stay in hospitalvaried between 5.8 days for those under 65years of age and 7.8 days for older patients76

[III]. A prospective study of costs and outcomeof CAP from five hospitals in North Americaconcluded that the costs of antibiotic treatmentvaried widely but had no eVect on outcome ormortality. Patients treated in hospitals with thelowest costs did not demonstrate worse medi-cal outcomes55 [Ib].

Summary+ The direct costs associated with CAP are

high and mostly associated with inpatientcare costs [III].

+ Substantial cost savings could probably bemade by strategies to prevent CAP, reducethe requirement for hospital admission, andshorten the length of hospital stay forpatients with non-severe CAP [III].

2.4 What comments can be made aboutcost eVectiveness of diVerent treatments?We are not able to provide any structuredguidance on this subject. Modern guidelinesshould attempt to provide information not onlyon clinical management, but also on theassessment of robust published data on the costeVectiveness of therapies. However, it wasnoted that there is a clear deficiency of goodquality comparative clinical data which wouldallow meaningful comparisons of managementand antibiotic strategies for CAP, whetherassessing for clinical or cost eVectivenessoutcomes.

Summary+ We have not attempted a systematic ap-

praisal of current pharmacoeconomic evi-dence for CAP and do not give a structuredview on cost eVectiveness.

+ Cost eVectiveness data pertinent to UKpractice do not exist at the time of writingand are an area for further research.


www.thoraxjnl.com



3 Aetiology and epidemiology

3.1 IntroductionNo two studies of the aetiology of CAP are thesame. Apparent diVerences in the observedfrequency of pathogens, while possibly real,may also be due to a number of other factorsincluding health care delivery (distribution ofmanagement between primary and secondarycare; hospital and ICU admission practices);population factors such as age mix, thefrequency of alcoholism, comorbid diseases,immunosuppression, and malignancy; andstudy factors such as the type and number ofsamples collected, investigations performed,and interpretation of results. Frequently suchdetails are not explicitly stated in the studymethodology and, although we have notincluded studies which do not comply withcertain standards, apparently similar studiesmay hide very diVerent methodologies. Withthe exception of the elderly, few adequatelypowered studies using the same methodologyhave been used to compare diVerent popula-tion groups. Conclusions about observeddiVerences in the following data must thereforebe treated with caution.

Many of the statements in the following textarise from a comparison of studies rather thandata from individual studies. For this reasonevidence grades follow statements to justify theconclusions, as well as individual references.

3.2 What are the causes of adult CAP inthe UK?These are set out in table 2, together withdetails of the relevant references (and gradingof evidence from those individual references),grouped together according to whether pa-tients have been managed in the community, inhospital, or on an ICU. For all these groups acommon range of pathogens is regularlyidentified as causing CAP [Ib]. Although a sin-gle pathogen is identified in 85% of patientswhere an aetiology is found, the true frequencyof polymicrobial CAP is not known and

observed figures are dependent on the intensityof the investigation. S pneumoniae is the mostfrequently identified pathogen [Ib]. Therelative frequency of pathogens in patientsmanaged in the community and in hospital isprobably similar, but the absence of more thanone study in the community makes furtherconclusions uncertain. Legionella species and Saureus are identified more frequently in pa-tients managed on the ICU [Ib]. The apparentdiVerence in the frequency of M pneumoniaemay depend on whether or not a study isperformed in an epidemic year [II]. Gramnegative enteric bacilli, C psittaci, and C burnetiiare uncommon causes of CAP [Ib].

3.3 What are the causes of adult CAP insimilar populations elswhere in theworld?The results and references of relevant studiesfrom the remainder of Europe, Australia andNew Zealand, and North America are com-pared in tables 3, 4, and 5. For patientsmanaged in the community and in hospital, thefrequency of pathogens is broadly similar tothat in the UK [II]. This suggests that aspectsof these guidelines will be applicable to othercountries as well as the UK. The absence ofstudies using sensitive methods for pneumo-coccal polysaccharide capsular antigen detec-tion for the identification of S pneumoniae maybe the explanation for the lower frequency out-side the UK (fig 3). The apparent diVerences inM pneumoniae may relate to the presence orabsence of epidemics at the time of the study(fig 4). C pneumoniae is identified regularly inEurope and North America as well as in arecent UK study [II].

There are no North American studies ofpatients managed in an ICU. Antibiotic resist-ant S pneumoniae appears to be no morefrequent in severely ill patients admitted to theICU than in those managed on an ordinaryhospital ward in a country where suchresistance is common77 [Ib]. Studies of patientswith severe CAP from Europe suggest a lowerfrequency of legionella and a higher frequencyof Gram negative enteric bacilli infections thanin the UK. These diVerences may be real ormethodological [IVa].

3.4 How does the aetiology diVer incertain geographical areas?Specific studies suggest a higher frequency ofcertain pathogens in the geographical areasshown in table 6 [II].

3.5 Is the aetiology diVerent in specificpopulation groups?THE ELDERLY

Three UK studies (two using a definition ofelderly of over 65 years of age but excludingthose aged over 7913 [Ib] and one using a defi-nition of over 75 years69 [Ib]) have reported

Table 2 CAP studies conducted in the UK in diVerent settings

Community(1 study*, n=236)

In hospital(5 studies†, n=1137)

Intensive care unit(4 studies‡ n=185)

Mean(%) 95% CI

Mean(%) 95% CI

Mean(%) 95% CI

S pneumoniae 36.0 29.9 to 42.1 39 36.1 to 41.8 21.6 15.9 to 28.3H influenzae 10.2 6.3 to 14.0 5.2 4.0 to 6.6 3.8 1.5 to 7.6Legionella spp 0.4 0.01 to 2.3 3.6 2.6 to 4.9 17.8 12.6 to 24.1S aureus 0.8 0.1 to 3.0 1.9 1.2 to 2.9 8.7 5.0 to 13.7M catarrhalis ? 1.9 0.6 to 4.3 ?Gram negative

enteric bacilli 1.3 0.3 to 3.7 1.0 0.5 to 1.7 1.6 0.3 to 4.7M pneumoniae 1.3 0.3 to 3.7 10.8 9.0 to 12.6 2.7 0.9 to 6.2C pneumoniae ? ? 13.1 9.1 to 17.2 ? ?C psittaci 1.3 0.3 to 3.7 2.6 1.7 to 3.6 2.2 0.6 to 5.4C burnetii 0 0 to 1.6 1.2 0.7 to 2.1 0 0 to 2.0All viruses 13.1 8.8 to 17.4 12.8 10.8 to 14.7 9.7 5.9 to 14.9Influenza A & B 8.1 4.9 to 12.3 10.7 8.9 to 12.5 5.4 2.6 to 9.7Mixed 11.0 7.0 to 15.0 14.2 12.2 to 16.3 6.0 3.0 to 10.4Other 1.7 0.5 to 4.3 2 1.3 to 3 4.9 2.3 to 9.0None 45.3 39.0 to 51.7 30.8 28.1 to 33.5 32.4 25.7 to 39.7

*Reference 40 [Ib].†References 11 [Ib], 12 [Ib], 14 [Ib], 15 [Ib], 69 [Ib].‡References 21 [Ib], 22 [Ib], 66 [Ib], 73 [II].


www.thoraxjnl.com



data on the comparative frequency of patho-gens in the elderly compared with that in ayounger population. The results are combinedin fig 5. For most pathogens their frequency isthe same in the young and in the elderly, but Mpneumoniae and legionella infection are lessfrequent in the elderly [Ib]. M pneumoniae andother atypical pathogens were found to occurmore frequently in patients aged <60 years inone other study78 [Ib]. H influenzae may also bemore commonly identified in the elderly [II].Gram negative enteric bacilli were no morecommon in elderly patients [III], although thishas been reported in at least one other study79

[II]. No diVerence in the frequency ofpathogens according to age was found in onestudy of patients with severe CAP80 [III].

PATIENTS WITH CHRONIC OBSTRUCTIVE

PULMONARY DISEASE (COPD)There are no relevant UK studies. H influenzaeand M catarrhalis may be more frequent. OneDanish study directly compared those with andwithout COPD and found no diVerence inpathogen frequency, but numbers were smallso real diVerences may have been missed81 [II].A Spanish study of patients with COPD but

with no control group found a pathogen distri-bution similar to that described in studies ofCAP in the general population82 [II]. A furtherSpanish study found S pneumoniae, Enterobacte-riaceae, P aeruginosa, and mixed infections tooccur more frequently in those with chroniclung disease77 [Ib]. COPD was found morefrequently in patients with bacteraemic pneu-mococcal pneumonia than in those with othercommunity acquired pneumonias in onestudy83 [Ib].

PATIENTS WITH DIABETES

Diabetes was found to be more frequent inpatients with bacteraemic pneumococcal pneu-monia than in those with either non-bacteraemic pneumococcal pneumonia or allcommunity acquired pneumonias in onestudy83 [Ib].

NURSING HOME RESIDENTS

There are no UK studies. Aspiration43 [II],84 [II], Gram negative enteric bacilli85 [III], andanaerobes [IVb] may occur more frequentlythan in matched elderly patients. NorthAmerican studies, which suggest these diVer-ences, may not be relevant to the UKpopulation and healthcare system. Legionellainfections and atypical pathogens are uncom-mon43 [II], 86 [III].

ALCOHOLIC PATIENTS

There are no UK studies. Aspiration87 [II],pneumococcal infection overall77 [Ib], bacter-aemic pneumococcal infection77 [Ib], 83 [II],Gram negative enteric bacilli79 [II], Legion-ella88 [III], atypical pathogens77 [Ib], Cpneumoniae77 [Ib], anaerobes [IVb], and mixedinfections77 [Ib] may be more frequent.

PATIENTS ON ORAL STEROIDS

There are no UK studies. Infection withLegionella species may be more frequent89 [III].

ASPIRATION PNEUMONIA

There are no UK studies. Most studies of CAPexclude such patients. Anaerobic bacteria andGram negative enteric bacilli may be morecommon87 [III], 90 [III].

Table 3 Studies of CAP conducted in the community

UK(1 study*, n=236)

Rest of Europe(6 studies†, n=654)

North America(1 study‡, n=149)

Mean(%) 95% CI

Mean(%) 95% CI

Mean(%) 95% CI

S pneumoniae 36.0 29.9 to 42.1 8.4 6.4 to 10.8 ? ?H influenzae 10.2 6.3 to 14.0 1.1 0.4 to 2.2 ? ?Legionella spp 0.4 0.01 to 2.3 2.8 1.6 to 4.3 0.7 0.01 to 3.7S aureus 0.8 0.1 to 3.0 0 0.0 to 0.7 ? ?M catarrhalis ? 0 0.0 to 0.6 ?Gram negative

enteric bacilli1.3 0.3 to 3.7 0.2 0.0 to 1.0 ? ?

M pneumoniae 1.3 0.3 to 3.7 13.3 10.7 to 15.9 26.2 19.3 to 34.0C pneumoniae ? ? 8.7 6.5 to 11.3 14.8 9.5 to 21.5C psittaci ? ? 2.0 1.1 to 3.4 14.8 9.5 to 21.5C burnetii 0 0 to 1.6 0.8 0.3 to 1.9 2.7 0.7 to 6.7All viruses 13.1 8.8 to 17.4 12.4 9.9 to 14.9 8.1 4.2 to 13.6Influenza A & B 8.1 4.9 to 12.3 6.3 4.5 to 8.4 6.0 2.8 to 11.2Mixed 11.0 7.0 to 15.0 4.7 2.8 to 7.3 4.7 1.9 to 9.4Other 1.7 0.5 to 4.3 2.0 1.1 to 3.4 0 0 to 2.5None 45.3 39.0 to 51.7 53.7 49.8 to 57.5 50.3 42.0 to 58.6

*Reference 40 [Ib].†References 25 [Ib], 370 [II], 371 [Ib], 372 [Ib], 373 [Ib], 374 [Ib].‡Reference 29 [II].

Table 4 Studies of CAP conducted in hospital

UK(5 studies*, n=1137)


Australia & New Zealand(3 studies‡, n=453)

North America(4 studies§, n=1306)

Mean (%) 95% CI Mean (%) 95% CI Mean (%) 95% CI Mean (%) 95% CI

S pneumoniae 39 36.1 to 41.8 19.4 18.4 to 20.4 38.4 33.9 to 42.9 11.3 9.5 to 13.0H influenzae 5.2 4.0 to 6.6 3.9 3.4 to 4.4 9.5 7 to 12.6 6.3 5.0 to 7.7Legionella spp 3.6 2.6 to 4.9 5.1 4.6 to 5.7 7.5 5.3 to 10.3 4.8 3.7 to 6.0M catarrhalis 1.9 0.6 to 4.3 1.2 1.0 to 1.5 3.1 1.4 to 6.1 1.2 0.5 to 2.5S aureus 1.9 1.2 to 2.9 0.8 0.5 to 1.1 2.9 1.5 to 4.9 3.8 2.9 to 5.0Gram negative enteric bacilli 1 0.5 to 1.7 3.3 2.8 to 3.7 4.6 2.9 to 7 5.3 4.1 to 6.6M pneumoniae 10.8 9.0 to 12.6 6 5.4 to 6.6 14.6 11.3 to 17.8 4.1 3.1 to 5.3C pneumoniae 13.1 9.1 to 17.2 6.3 5.5 to 7.3 3.1 1.4 to 6.1 5.9 4.3 to 7.8C psittaci 2.6 1.7 to 3.6 1.4 1.1 to 1.8 1.4 0.5 to 3.2 0.1 0 to 0.7C burnetii 1.2 0.7 to 2.1 0.9 0.6 to 1.1 0 0 to 3.4 2.3 1.5 to 3.7All viruses 12.8 10.8 to 14.7 9.5 8.6 to 10.3 10.6 7.8 to 13.4 8.9 7.4 to 10.6Influenza A & B 10.7 8.9 to 12.5 5.3 4.6 to 6.1 6.4 4.3 to 9.1 5.9 4.5 to 7.6Mixed 14.2 12.2 to 16.3 6.3 5.5 to 7.1 19.6 16 to 23.3 8.5 7.0 to 10.3Other 2 1.3 to 3 2 1.7 to 2.4 4 2.4 to 6.2 8.0 6.6 to 9.7None 30.8 28.1 to 33.5 50.7 49.5 to 52.0 31.6 27.3 to 35.8 40.7 38.1 to 43.4

*References 11 [Ib], 12 [Ib], 14 [Ib], 15 [Ib], 69 [Ib].†References 24 [Ib], 77 [Ib], 78 [Ib], 79 [Ib], 81 [Ib], 122 [Ib], 156 [Ib], 190 [Ib], 193 [Ib], 372 [Ib], 373 [Ib], 375 [Ib], 376 [Ib], 377 [Ib], 378 [II], 379 [Ib], 380[Ib], 381 [Ib], 382 [Ib], 383 [Ib], 384 [Ib], 385 [Ib]‡References 70 [Ib], 71 [Ib], 72 [Ib].§References 43 [Ib], 113 [Ib], 386 [Ib], 387 [Ib].


www.thoraxjnl.com



3.6 What are the epidemiological patternsof pathogens causing CAP and is thisinformation useful to the clinician?Streptococcus pneumoniaeS pneumoniae occurs most commonly in thewinter91 [II]. Outside the UK epidemics haveoccurred in overcrowded settings such asmens’ shelters and prisons92 [II], 93 [II].

Legionella speciesLegionella infection is most common inSeptember and October in the UK(www.phls.co.uk/facts) [II]; 52% (95% CI 49to 54) of UK cases are related to travel, 91%(95% CI 87 to 94)94 95 of these relating to travelabroad96 [II], 97 [II], 98 [II], 99 [II], 100 [II]. Clus-ters of cases are linked to Mediterraneanresorts, especially Turkey and Spain, but only23% (95% CI 19 to 26)97 99 of cases occur inclusters96 [II], 97 [II], 98 [II], 99 [II], 100 [II]. Epi-demics occur related to water containingsystems in buildings96 [II].

Mycoplasma pneumoniaeEpidemics spanning three winters occur every4 years in the UK101 [II] as shown in fig 4.

Chlamydia pneumoniaeEpidemics occur in the community and inclosed communities102 [II], 103 [II], 104 [II]. Itsdirect pathogenic role as a cause of—asopposed to being associated with—CAP is notclear. Evidence that antibiotic treatment di-rected against this organism alters the course ofthe illness is lacking. When identified, otherbacterial pathogens such as S pneumoniae areoften identified in the same host26 [II], 28 [II],105 [II]. Patients may recover when antibioticsto which C pneumoniae is not sensitive aregiven105 [II].

Chlamydia psittaciInfection is acquired from birds and animalsbut human to human spread may occur [II].Epidemics are reported in relation to infectedsources at work—for example, poultry or duckworkers [II]. Only 20% of UK cases have a his-tory of bird contact106 [II].

Coxiella burnetiiCases are most common in April to June, pos-sibly related to the lambing and calving season[II]. Epidemics occur in relation to animalsources (usually sheep), but a history of occu-pational exposure is only present in 7.7% (95%CI 6.2 to 9.4) of cases107 [II].

Staphylococcus aureusIt is more common in the winter months.Coincident influenza type symptoms are re-ported in 39% (95% CI 27 to 53)11 12 14 15 [II].Evidence of coincident influenza virus infec-tion is found in 39% (95% CI 17 to 64) ofthose admitted to hospital11 12 14 15 [II], and50% (95% CI 25 to 75) of those admitted to anICU21 22 66 73 [II].

Influenza virusAnnual epidemics of varying size are seen dur-ing the winter months108 [II]. Pneumonia com-plicates 2.9% (95% CI 1.4 to 5.4) of cases in

Table 5 Studies of CAP conducted in the intensive care unit

UK(4 studies*, n=185)


Mean (%) 95% CI Mean (%) 95% CI

S pneumoniae 21.6 15.9 to 28.3 21.8 19.4 to 24.2H influenzae 3.8 1.5 to 7.6 5.3 4.1 to 6.8Legionella spp 17.8 12.6 to 24.1 5.5 4.2 to 7.2S aureus 8.7 5.0 to 13.7 7.0 5.6 to 8.6M catarrhalis ? ? 3.8 2.4 to 5.9Gram negative enteric bacilli 1.6 0.3 to 4.7 8.6 7.1 to 10.4M pneumoniae 2.7 0.9 to 6.2 2.0 1.3 to 3.0C pneumoniae ? ? 6.6 2.5 to 13.8C psittaci 2.2 0.6 to 5.4 0.9 0.4 to 1.9C burnetii 0 0 to 2.0 0.7 0.3 to 1.4Viruses 9.7 5.9 to 14.9 4.0 2.7 to 5.6Influenza A & B 5.4 2.6 to 9.7 2.3 1.1 to 4.2Mixed 6.0 3.0 to 10.4 5.0 2.4 to 9.1Other 4.9 2.3 to 9.0 8.4 6.8 to 10.1None 32.4 25.7 to 39.7 43.3 40.4 to 46.2

*References 21 [Ib], 22 [Ib], 66 [Ib], 73 [II].†References 67 [Ib], 75 [Ib], 77 [Ib], 187 [Ib], 188 [Ib], 388 [Ib], 389 [Ib], 390 [Ib], 391 [Ib],392 [II].

Sensitivemethods used

Sensitivemethods not used

Combinedmean

90

80

70

60

50

40

30

20

10

0

%

Figure 3 Frequency of identification of pneumococcal infection in studies of adultsadmitted to hospital in Europe in relation to the use of sensitive detection methods forS pneumoniae (with 95% CI).

350

300

250

200

150

100

50

0

Nu

mb

er o

f re

po

rts

200019991998199719961995

Year19941993199219911990

Figure 4 Laboratory reports of mycoplasma infections to the Communicable DiseaseSurveillance Centre, England and Wales, 1990–2000 (4 weekly)101 [II].

Table 6 Pathogens which are more common as a cause of CAP in certain geographicalregions

Pathogen Geographical area References

Legionella spp Countries bordering theMediterranean

28 [II], 382 [II]

C burnetii North West Spain 393 [II]C burnetii Canada 394 [II]Klebsiella pneumoniae South Africa 395 [II], 396 [II]Burkholderia pseudomallei South East Asia and Northern

Australia94 [II], 397 [II], 398 [II],399 [II]

Gram negative enteric bacilli Italy 383 [II]Mycobacterium tuberculosis Non-industrialised countries 94 [II], 95 [II]


www.thoraxjnl.com



the community109 [Ib]. The frequency ofstaphylococcal pneumonia in patients withinfluenza symptoms is not known. Of adults

with CAP admitted to UK hospitals in whominfluenza infection is confirmed, 10% (95% CI4.1 to 19.5) have coincident S aureus infection[II]. Of those admitted to an ICU, thecorresponding figure is 67% (95% CI 35 to90)21 22 66 73 [II].

Summary+ The low frequency of legionella, staphylo-

coccal, C psittaci, and C burnetii infection inpatients with CAP in both the communityand in hospital, together with the likely highfrequency of the relevant risk factors (out-lined above) in the general populationsuggests that routine enquiry about suchfactors is likely to be misleading [IV].

+ Only in those with severe illness where thefrequency of legionella and staphylococcalinfection is higher, may enquiry aboutforeign travel and influenza symptoms be ofpredictive value [IV].

+ Knowledge of increased mycoplasma activityin the community during an epidemic periodmay help guide the clinician to the increasedlikelihood of mycoplasma infection [IV].

Figure 5 DiVerence in causative pathogens between young and elderly patients. Verticalaxis shows the percentage diVerence in frequency between young and the elderly groups forpooled data from three UK studies (with 95% confidence intervals) [Ib].13 69 Sp =Streptococcus pneumoniae, Hi = Haemophilus influenzae, Lp = Legionella spp, Sa =Staphylococcus aureus, Mcat = Moraxella catarrhalis, GNEB = Gram negative entericbacilli, Mp = Mycoplasma pneumonia, Cp = Chlamydia pneumoniae, Cpsi = Chlamydiapsittaci, Cb = Coxiella burnetti, AllV = viruses, Flu = influenza viruses, Oth = otherorganisms, None = no pathogen identified.

20

15

10

5

0

_5

_10

Sp Hi Lp Sa Mcat GNEB

Less common in the elderly

More common in the elderly

Mp Cp Cpsi Cb AllV Flu Oth None


www.thoraxjnl.com



4 Clinical features

4.1 Can CAP be reliably diVerentiatedfrom other respiratory conditions byclinical features alone?Diagnosing pneumonia clinically without achest radiograph is inaccurate110 [Ia], but thepresence of normal vital signs on chestexamination makes an underlying diagnosis ofpneumonia unlikely110 [Ia]. A review of pub-lished studies reported that there were no indi-vidual clinical findings that reliably diagnosedCAP110 [Ia]. The problem is compounded bypoor interobserver reliability in eliciting respi-ratory signs111 [II]. Although most patients withCAP can be managed successfully in the com-munity by their general practitioner withoutinvestigations, distinguishing CAP from othercauses of respiratory symptoms and signs canbe diYcult, particularly where the presence ofcomorbidity such as left ventricular failure,chronic lung disease, or COPD complicate theclinical picture. The elderly can present a par-ticularly diYcult diagnostic challenge becausethey more frequently present with non-specificor absent symptoms and signs112 [II].

Summary+ The diagnosis of CAP on the basis of history

and physical findings is inaccurate without achest radiograph [Ia].

4.2 Can the aetiology of CAP be predictedfrom clinical features?There have been a large number of publica-tions looking at the possibility of predicting theaetiological agent from clinical features.

Fang et al113 [II] found no overall distinctiveclinical features at presentation that enabledprediction of the aetiological agent and sug-gested that the term “atypical” pneumoniashould be abandoned (see sections 1 and 8).Similarly Farr et al114 [II] reported thataetiology could not be predicted reliably usingfive clinical variables. For patients with severeCAP admitted to the ICU, clinical features hadlittle value in predicting the aetiological agent75

[II] with the exception of those patients withfever (>39°C) or chest pain who were statisti-cally more likely to have pneumococcal pneu-monia. Similarly, pleuritic chest pain was foundto be less likely in those patients with “atypical”pathogens115 [II]. The term “atypical patho-gen” is defined in section 1.

Summary+ The likely aetiological agent causing CAP

cannot be accurately predicted from clinicalfeatures [II].

+ The term “atypical” pneumonia should beabandoned as it incorrectly implies thatthere is a characteristic clinical presentationfor patients with infection caused by “atypi-cal” pathogens [II].

4.3 Specific clinical features of particularrespiratory pathogensCLINICAL FEATURES

The clinical features associated with specificpathogens are described below and summa-rised in box 1.

Streptococcus pneumoniae+ Increasing age, comorbidity, acute onset,

high fever and pleuritic chest pain

Bacteraemic Streptococcus pneumoniae+ Female sex, excess alcohol, diabetes mel-

litus, chronic obstructive pulmonary dis-ease, dry cough

Legionella pneumophila+ Younger patients, smokers, absence of

comorbidity, diarrhoea, neurologicalsymptoms, more severe infection, evi-dence of multisystem involvement (e.g.abnormal liver function tests, elevatedserum creatine kinase)

Mycoplasma pneumoniae+ Younger patients, prior antibiotics, less

multisystem involvement

Chlamydia pneumoniae+ Longer duration of symptoms before

hospital admission, headache

Coxiella burnetii+ Male sex, dry cough, high fever

Box 1 Some clinical features reported to be morecommon with specific pathogens (references aregiven in the text).

Streptococcus pneumoniaeOne study using discriminant function analysisfound pneumococcal aetiology to be morelikely in the presence of cardiovascular comor-bidity, an acute onset, and pleuritic chest pain,and less likely if patients had a cough or flu-likesymptoms or had received an antibiotic beforeadmission116 [III].

Bacteraemic pneumococcal pneumonia wasfound to be more likely in those patients whohad at least one of the following features:female sex, history of no cough or a non-productive cough, history of excess alcohol,diabetes mellitus, or COPD83 [II].

Mycoplasma pneumoniaeOne study compared CAP due to M pneumo-niae to patients with pneumococcal or le-gionella pneumonia40 [III] and reported thatpatients with mycoplasma pneumonia wereyounger and less likely to have multisysteminvolvement and were more likely to havereceived an antibiotic before admission. Bycontrast, another report117 [II] found nodistinctive clinical features in patients with Mpneumoniae.

Legionella pneumophilaNumerous studies have reported a variety ofclinical features to be more common withlegionella pneumonia. In a study comparing


www.thoraxjnl.com



clinical features of diVerent CAP pathogens,patients with L pneumophila were more likely tohave encephalopathy, elevated levels of liverenzymes, haematuria and, less commonly, tohave upper respiratory tract symptoms118 [III].A subsequent study by the same group,however, found no significant clinical diVer-ences119 [III]. A study in Israel reportedpatients with L pneumonia to have no diVeren-tiating clinical features apart from beingyounger with a low incidence of comorbidity120

[II].Two studies have compared cases of le-

gionella and pneumococcal pneumonia. Onereported that patients with legionella infectionwere less likely to be smokers, have pleuralsymptoms and arthromyalgia and were morelikely to be older or to be dyspnoeic than thosewith pneumococcal pneumonia121 [II], whilethe other found that patients with legionellawere more likely to be alcoholics, smokers, tohave already received an antibiotic, and to havegastrointestinal or neurological symptoms butwere less likely to have purulent sputum orpleuritic chest pain122 [II]. A recent prospectivestudy also showed that patients with legionellapneumonia have a low incidence of comorbid-ity and an increased frequency of diarrhoea andraised serum creatinine kinase88 [II].

Chlamydia pneumoniaeA comparative study of patients with Cpneumoniae and S pneumoniae pneumoniafound the former more likely to present withheadaches and a longer duration of symptomsbefore hospital admission105 [II]. A study fromIsrael reported no distinguishing clinical fea-tures for chlamydial pneumonia, except that itaVected older patients when compared withpneumococcal and mycoplasma infections123

[II]. A recent study reported that, in caseswhere C pneumoniae was the only pathogenidentified, the illness was generally mild withnon-specific symptoms124 [II].

Coxiella burnetiiCAP due to C burnetii (Q fever) causesnon-specific clinical features125 [II], 126 [II].Two reviews of Q fever have reported thatinfection was more common in younger menand that dry cough and high fever werecommon127 [III], 128 [III]. Epidemiological fea-tures are discussed in section 3.

Klebsiella pneumoniaeThe clinical features of bacteraemic pneumo-nia due to S pneumoniae and K pneumoniae havebeen compared129 [III]. In the latter group menwere more commonly aVected and presentedwith a lower platelet count and leucopenia.Alcoholics were at particular risk of bacterae-mic and fatal Klebsiella pneumonia.

Some rarer community respiratory pathogensCommunity acquired acinetobacter pneumo-nia is seen more often in older patients with ahistory of alcoholism and has a high rate ofmortality130 [III]. CAP due to Streptococcusmilleri may indicate a dental or abdominalsource of infection131 [III], while CAP causedby viridans streptococci is associated with aspi-ration132 [III].

4.4 CAP in the elderly: are risk factorsand clinical features diVerent?The classic symptoms and signs of pneumoniaare less likely in the elderly while non-specificfeatures, especially confusion, are more likely112

[II], 115 [II], 133 [II]. Comorbid illness is morecommon in older patients with CAP and twostudies have found absence of fever is morelikely than in younger patients113 [II], 134 [II].

Case controlled studies of pneumonia ac-quired in nursing homes have shown that bothaspiration and comorbidity were more com-mon in nursing home acquired pneumoniathan in others with CAP84 [II]. The inpatientmortality rate for nursing home acquiredpneumonia was higher than for age matchednon-nursing home acquired pneumonia pa-tients135 [II].

The relationship between the aetiology ofCAP and the age of the patient is discussed insection 3.

Summary+ Elderly patients with CAP more frequently

present with non-specific symptoms andhave comorbid disease and a higher mor-tality rate, and are less likely to have a feverthan younger patients [II].

+ Aspiration is a risk factor for CAP in elderlypatients, particularly nursing home resi-dents [II].


www.thoraxjnl.com



5 Radiological, general and microbiologicalinvestigations

5.1 Are there characteristic features thatenable the clinician to predict the likelypathogen from the chest radiograph?In a comparative study of cases of legionellapneumonia, pneumococcal pneumonia, myco-plasma pneumonia and psittacosis no uniqueradiological pattern was found although somediVerences were reported136 [III]. The lowerlobes were aVected most commonly, regardlessof aetiology.

Homogenous shadowing was less commonin mycoplasma pneumonia than in the othertypes. Multilobe involvement at presentationwas more likely in bacteraemic pneumococcalpneumonia than in non-bacteraemic pneumo-coccal pneumonia or legionella pneumonia.Pleural eVusions were more common inbacteraemic pneumococcal disease. Lymph-adenopathy was noted in some cases of myco-plasma infections but not in the other types ofinfection.

The frequency of radiographic deteriorationafter hospital admission and the rates ofresolution of radiographic shadowing diVeredaccording to the underlying aetiology (seebelow). In one study of 149 patients, pneumo-coccal infections were found to be associatedwith multilobe involvement more often thanpneumonia caused by atypical pathogens29 [II],a finding not reported in another comparativestudy137 [II]. Other studies have found nodiVerence between C pneumoniae and S pneu-moniae pneumonia138 [II], between C pneumo-niae and a variety of other causes of CAP123 [II],or between cases of Q fever pneumonia andthose due to other pathogens139 [II].

CAP caused by S aureus appears to be morelikely to present with multilobar shadowing,cavitation, pneumatoceles, or spontaneouspneumothorax140 [III].

K pneumoniae has been reported to producechest radiograph changes with a predilectionfor upper lobes (especially the right), a bulginginterlobar fissure, and abscess formation withcavitation. However, a prospective study of 15proven cases of klebsiella pneumonia, mostlyhospital acquired, found that although the rightupper lobe was most likely to be involved, nocase had a bulging interlobar fissure or cavita-tion141 [II]. A bulging interlobar fissure is prob-ably just a reflection of an intense inflammatoryreaction that can occur in any severe infectionsuch as pneumonia due to S aureus140 [III].

There are few data on the role of high reso-lution CT lung scans in CAP. One study hasreported a diVerence in CT appearances in 18patients with CAP due to bacterial infectionscompared with 14 patients with atypical patho-gens142 [III]. A smaller study has reported thathigh resolution CT scans may improve theaccuracy of diagnosing CAP compared withchest radiography alone143 [II]. Similarly, CTlung scans have better sensitivity than standard

chest radiographs in patients with mycoplasmapneumonia144 [II]. CT lung scans may be use-ful in subjects where the diagnosis is in doubt145

[III], but in general there seems little role forCT scanning in the usual investigation of CAP.

Summary+ There are no characteristic features of the

chest radiograph in CAP that allow a confi-dent prediction of the likely pathogen [II].

5.2 How quickly do chest radiographsimprove after CAP?Radiographic changes resolve relatively slowlyafter CAP, and lag behind clinical recovery.Complete resolution of chest radiographicchanges occurred at 2 weeks after initialpresentation in 51% of cases, by 4 weeks in64%, and at 6 weeks in 73% in one study ofCAP146 [II]. Clearance rates were slower in theelderly, those with more than one lobe involvedat presentation, in smokers, and inpatientsrather than outpatients. Multivariate analysisshowed that only age and multilobe involve-ment were independently related to the rate ofclearance. Age was also a major factorinfluencing the rate of radiographic recovery inthe BTS multicentre CAP study14 [Ib]. Whenchest radiographs of patients with bacteraemicpneumococcal pneumonia were followed, only13% had cleared at 2 weeks and 41% at 4weeks147 [III]. Pneumonias caused by atypicalpathogens clear more quickly. The clearancerate has been reported to be faster formycoplasma pneumonia than for legionella orpneumococcal pneumonia which may take 12weeks or more136 [III]. In a series of patientswith C burnetii pneumonia, 81% of the chestradiographs had returned to normal within 4weeks127 [III].

Legionella pneumonia seemed to be particu-larly slow to resolve136 [III]. In this study radio-graphic deterioration after admission to hospi-tal was more common with legionella (65% ofcases) and bacteraemic pneumococcal pneu-monia (52%) than with non-bacteraemicpneumococcal (26%) or mycoplasma pneumo-nia (25%). Residual pulmonary shadowing wasfound in over 25% of cases of legionella andbacteraemic pneumococcal pneumonia. Dete-rioration after admission has also been re-ported in more than 50% of cases of S aureuspneumonia140 [III]. Radiographic deteriorationafter hospital admission appears to be morecommon in older patients (aged 65 years orover)134 [II].

Summary+ Radiological resolution often lags behind

clinical improvement from CAP, particularlyfollowing legionella and bacteraemic pneu-mococcal infection [III].


www.thoraxjnl.com



+ Pneumonia caused by atypical pathogensclears more quickly than pneumonia causedby bacterial infection [III].

+ Radiological resolution is slower in theelderly and where there is multilobe involve-ment [Ib].

Recommendation+ In a patient who is improving clinically

and for whom there are no concerningclinical features, it will usually not benecessary to perform further investiga-tions just because radiological improve-ment lags behind clinical recovery[B+].

5.3 When should the chest radiograph berepeated during recovery and what actionshould be taken if the radiograph has notreturned to normal?Repeat chest radiographs are probably oftenordered unnecessarily following CAP148 [IVa].Although it is usual practice to repeat the chestradiograph on discharge from hospital andagain at “routine” hospital clinic follow upabout 6 weeks later, there is no evidence onwhich to base a recommendation regarding thevalue of this practice in patients who haveotherwise recovered satisfactorily. It is also notknown whether there is any value in arrangingclinical follow up in a hospital clinic rather thanwith the patient’s general practitioner.

The main concern is whether the CAP was acomplication of an underlying condition suchas lung cancer. This concern will depend on avariety of factors such as age, smoking status,pre-existing conditions such as COPD, and theclinical condition of the patient. In a study of236 adults presenting to their general prac-titioner with a clinical diagnosis of CAP, 10were found to have underlying lung cancer oninvestigation. There was a high frequency oflung cancer in older smokers (six of 36 (17%)smokers aged over 60 years), suggesting that achest radiograph is particularly indicated inthis group of patients with CAP in the commu-nity40 [II].

Studies of CAP in hospital often excludepatients found to have lung cancer, making itdiYcult to assess how frequently lung cancerpresents acutely with CAP. In one study of 162adults hospitalised with suspected CAP thediagnosis was accepted in only 127, 10 (6%) ofwhom were found to have cancer11 [II].Another study found only 13 (1.3%) of 1011patients hospitalised with CAP to have anunderlying lung cancer on investigation149 [III].Eight of these were detected on the admissionchest radiograph and the others were detectedbecause of unsatisfactory clinical recovery.They concluded that a convalescent radio-graph was useful in detecting occult lung can-cer only if signs or symptoms persisted after amonth or so. The practice of performing abronchoscopy before discharge from hospitalon patients admitted with CAP has been inves-tigated150 [III]. In patients aged over 50 orthose who were current or ex-smokers, 14%

were found to have an abnormality at bron-choscopy (bronchial carcinoma was diagnosedin 11%).

Recommendations+ The chest radiograph need not be

repeated before hospital discharge inthose who have made a satisfactoryclinical recovery from CAP [D].

+ A chest radiograph should be arrangedafter about 6 weeks for all those patientswho have persistent symptoms or physi-cal signs or who are at higher risk ofunderlying malignancy (especiallysmokers and those over 50 years),whether or not they have been admittedto hospital [C].

+ Further investigations which may in-clude bronchoscopy should be consid-ered in patients with persisting signs,symptoms, and radiological abnormali-ties at around 6 weeks after completingtreatment [C].

+ It is the responsibility of the hospitalteam to arrange the follow up plan withthe patient and the general practitionerfor those patients admitted to hospital[D] (see section 7.5).

5.4 Why are non-microbiologicalinvestigations performed in CAP?General investigations are performed to assessseverity (see section 6), to assess the impact onor detect the presence of any co-morbiddisease, to provide some pointer to the particu-lar aetiological agent or group of pathogens, toidentify complications, and to monitorprogress (see section 9).

5.5 What general investigations should bedone in a patient with suspected CAP inthe community?

Recommendations+ General investigations, including a

chest radiograph, are not necessary forthe majority of patients with CAP whoare managed in the community [C].

+ Out of hours and emergency generalpractitioner assessment centres shouldconsider obtaining pulse oximeters toallow for simple assessment of oxygena-tion [D] (see section 7).

5.6 What general tests should be done onall patients admitted to hospital?It is normal practice to perform a routine bio-chemical and haematological profile on admis-sion. A white cell count of >15 × 109/l stronglysuggests a bacterial (particularly pneumococ-cal) aetiology151 [III], although lower counts donot exclude a bacterial cause. A white cellcount of >20 × 109/l or <4 × 109/l is an indica-tor of severity (see section 6). Urea, electro-lytes, and liver function tests are performed toassess severity (see section 6) and for the iden-tification of underlying or associated renal orhepatic disease.


www.thoraxjnl.com



There is some recent evidence that measure-ment of C reactive protein (CRP) may have auseful role in the management of hospitalisedadult patients with CAP. Raised levels of CRPon admission are a relatively more sensitivemarker of pneumonia than an increasedtemperature or raised white cell count152 [II].One study found that only 5% of patientsadmitted with CAP had CRP levels of<50 mg/l153 [III], while another reported thatall patients with CAP had levels above 50 mg/land 75% of patients had levels above 100 mg/l152

[II]. CRP levels are generally higher in patientswho have not received antibiotics beforeadmission152 [II], 154 [III]. Higher CRP levelshave been reported in patients with pneumo-coccal pneumonia (especially if complicated bybacteraemia) than in those with mycoplasma orviral pneumonias154 [III]. One study has alsoreported that measurement of CRP can help todistinguish pneumonia from exacerbations ofCOPD using an arbitrary cut oV of 100 mg/l152

[III].Serial measurements of CRP may be espe-

cially useful for monitoring the response ofpatients to treatment. One study found that themedian time for a 50% reduction in the CRPlevel was 3.3 days153 [III]. A CRP level thatdoes not fall by 50% within 4 days suggestsfailure of treatment or the development ofcomplications such as empyema or antibioticassociated diarrhoea.

Further prospective studies are required todefine the role of CRP measurements in themanagement of patients with CAP. Thespecificity of CRP measurement in this patientpopulation requires further evaluation as anumber of other respiratory conditions—for example, neoplasia and pulmonaryinfarction—and other non-respiratory bacterialinfections can cause an increase in levels ofCRP.

Summary+ The published evidence to date suggests that

measurement of CRP on admission may behelpful in distinguishing pneumonia fromother acute respiratory illnesses and mayalso allow useful comparison with a repeatmeasurement in patients who subsequentlyfail to improve [III].

Recommendations+ All patients should have the following

tests performed on admission:+ Chest radiograph [C].+ Full blood count [B–].+ Urea, electrolytes and liver function

tests [C].+ CRP when locally available [B–].+ Oxygenation assessment. Oxygen satu-

ration should be measured on admis-sion. Those with SaO2 <92% or withfeatures of severe pneumonia shouldhave an arterial blood gas measure-ment. It is essential to record theinspired oxygen concentration whenmeasuring oxygen saturation andblood gases to allow correct interpret-ation of the results (see section 7.3) [C].

5.7 Why are microbiologicalinvestigations performed in patients withCAP?Establishing the microbial cause of CAP isuseful for several reasons:(1) Identification of pathogens and antibioticsensitivity patterns permits selection of optimalantibiotic regimens.(2) Targeted and narrow spectrum antibiotictherapy limits drug costs, the threat ofantibiotic resistance, and adverse drug reac-tions such as C diYcile associated diarrhoea.(3) Specific pathogens have public health orinfection control significance, including le-gionella, psittacosis, Q fever, influenza A, andpenicillin resistant pneumococci. Patients withthese infections should be identified quickly sothat appropriate treatment and control meas-ures can be implemented.(4) Microbiological investigations allow moni-toring of the spectrum of pathogens causingCAP over time. This allows trends regardingaetiology and antibiotic sensitivity to betracked for public health needs.

Unfortunately, microbiological investiga-tions are insensitive and often do not contrib-ute to initial patient management155 [III]. Indetailed prospective aetiology studies themicrobial cause is not found in 25–60% ofpatients40 [II], 156 [II], and the yield is evenlower in routine hospital practice157 [III],158 [III].

Recommendations+ It is not necessary or appropriate to

perform a full range of microbiologicalinvestigations on every patient withCAP. The investigations performedshould be guided by the severity ofpneumonia, epidemiological risk fac-tors, and the response to treatment [D].

5.8 Which microbiological investigationsshould be performed in patients withsuspected CAP in the community?Comments about the pros and cons of diVerentmicrobiological investigations are given belowin section 5.9. Many of these investigations willnot be appropriate for patients with CAP man-aged in the community. Such patients are notusually severely ill, are at low risk of death, anddelays in transport of specimens to the labora-tory reduce the yield of bacterial pathogensespecially S pneumoniae from sputum cultures.The results are often received too late by thegeneral practitioner to be of much practicalvalue in initial management.

Recommendations+ For patients managed in the commu-

nity, microbiological investigations arenot recommended routinely [D].

+ Examination of sputum should be con-sidered for patients who do not respondto empirical antibiotic treatment [D].

+ Examination of sputum for Mycobacte-rium tuberculosis should be consideredfor patients with a persistent productivecough, especially if malaise, weight loss,


www.thoraxjnl.com



night sweats or risk factors for tubercu-losis (e.g. ethnic origin, social depriva-tion, the elderly) are present [D].

+ Serological investigations may be con-sidered during outbreaks (e.g. Legion-naires’ disease) or epidemic myco-plasma years, or when there is aparticular clinical or epidemiologicalreason [D].

5.9 What microbiological investigationsshould be performed in patients admittedto hospital with CAP?The investigations that are recommended forpatients admitted to hospital are summarisedin table 7. More extensive microbiologicalinvestigations are recommended only for pa-tients with severe CAP unless there areparticular clinical or epidemiological featuresthat warrant further microbiological studies.Comments and recommendations regardingspecific investigations are given below.

BLOOD CULTURES

Microbial causes of CAP that can be associatedwith bacteraemia include S pneumoniae, Hinfluenzae, S aureus, and K pneumoniae. Isola-tion of these bacteria from blood cultures inpatients with CAP is highly specific in deter-mining the microbial aetiology. Bacteraemia isalso a marker of illness severity. However, manypatients with CAP do not have an associatedbacteraemia. Even in pneumococcal pneumo-nia the sensitivity of blood cultures is at mostonly 25%47 [II], 103 [II], and is even lower forpatients given antibiotic treatment beforeadmission14 [II].

Recommendation+ Blood culture is recommended for all

patients admitted with CAP, preferablybefore antibiotic treatment is com-menced [D].

SPUTUM CULTURES

Sputum cultures may identify the causativeagent in CAP including unexpected or anti-biotic resistant pathogens such as S aureus or

penicillin resistant pneumococci. Routine spu-tum cultures are, however, neither very sensi-tive nor specific2 [Ia], and often do notcontribute to initial patient management159

[II].

Problems include:(1) The inability of patients to produce goodspecimens.(2) Prior exposure to antibiotics.(3) Delays in transport and processing.(4) DiYculty in interpretation due to contami-nation of the sample by upper respiratory tractflora which may include potential pathogenssuch as S pneumoniae and coliforms (especiallyin patients already given antibiotics).

Recommendations+ Sputum samples should be sent for cul-

ture and sensitivity tests from patientsadmitted to hospital with non-severeCAP who are able to expectorate puru-lent samples and have not received priorantibiotic treatment. Specimens shouldbe transported rapidly to the laboratory[D].

+ Sputum cultures should also be per-formed for patients with severe CAP orthose who fail to improve [D].

SPUTUM GRAM STAIN

The value of performing a Gram stain onexpectorated sputum has been widely debated.A meta-analytical review concluded that thesensitivity and specificity of Gram stains ofsputum in patients with CAP varied substan-tially in diVerent settings160 [Ia]. The presenceof large numbers of Gram positive diplococciin purulent samples from patients with CAPcan indicate pneumococcal pneumonia161 [II].There are many factors that determine the reli-ability and usefulness of Gram stain results,which are summarised below.

Advantages+ Quick and inexpensive.+ Can assess quality of samples (cytological

content) with rejection of poor quality sam-ples.

Table 7 Recommendations for the microbiological investigation of patients admitted to hospital with community acquired pneumonia (CAP)

Routine investigations in hospital for all patients with non-severe CAP Routine investigations in hospital for all patients with severe CAP

+ Blood cultures (minimum 20 ml) + Blood cultures (minimum 20 ml)+ Sputum for routine culture and sensitivity tests for those who have not

received prior antibiotics (± Gram stain*)+ Sputum or other respiratory sample† for Gram stain, routine culture and

sensitivity tests+ Clotted (acute) serum sample for store‡ + Pleural fluid, if present, for microscopy, culture and sensitivity+ Pleural fluid, if present, for microscopy, culture and sensitivity + Pneumococcal antigen tests (if available): urine, sputum or blood

For selected patients only+ Investigations for legionella pneumonia:§ (a) urine for legionella antigen;

(b) sputum or other respiratory sample for legionella culture and directimmunofluorescence (if available); (c) initial and follow up legionella serology

+ Investigations for legionella pneumonia: (a) urine for legionella antigen;(b) sputum or other respiratory sample for legionella culture and directimmunofluorescence (if available); (c) initial and follow up legionella serology

+ Follow up viral and atypical pathogen serology (to be run in parallel withinitial serum sample)¶

+ Investigations for atypical and viral pathogens: (a) if available, sputum orother respiratory sample for direct immunofluorescence (or other antigendetection test) to Chlamydia spp, influenza A & B, parainfluenza 1–3,adenovirus, respiratory syncytial virus, Pneumocystis carinii (if at risk); (b)initial and follow up viral and “atypical pathogen” serology

*The routine use of sputum Gram stain is discussed in the text.†Consider obtaining lower respiratory tract samples by more invasive techniques such as bronchoscopy (usually after intubation) or percutanous fine needle aspira-tion for those who are skilled in this technique.

‡The date of onset should be clearly indicated on the laboratory request form.§Patients with clinical or epidemiological risk factors (travel, occupation, co-morbid disease). Investigations should be considered for all patients with CAP duringoutbreaks.

¶For patients unresponsive to â-lactam antibiotics or those with a strong suspicion of an “atypical” pathogen on clinical, radiographic, or epidemiological grounds.


www.thoraxjnl.com



+ Can aid the interpretation of culture resultsand occasionally give an early indication ofpossible aetiology.

Disadvantages+ Strict criteria for interpretation require

appropriate operator training.+ Validity of results is directly related to the

experience of the interpreter162 [II].+ Sputum Gram stain correlates poorly with

culture results in conditions other thanCAP163 [II]. This poses practical diYcultiesfor laboratories that frequently have tointerpret results with little or no clinicalinformation.

Recommendations+ Laboratories should be able to oVer a

reliable Gram stain for patients withsevere CAP or complications as, onoccasions, this can give an immediateindicator of the likely pathogen. Routineperformance or reporting of sputum

Gram stain on all patients is unneces-sary but can aid the laboratory inter-pretations of culture results [D].

+ Laboratories performing sputum Gramstains should adhere to strict and locallyagreed criteria for interpretation andreporting of results [B+].

SEROLOGY FOR RESPIRATORY PATHOGENS

Respiratory serological tests usually compriseantibody tests for the atypical pathogens (Mpneumoniae, Chlamydia spp, C burnetii), influ-enza A virus, influenza B virus, adenovirus,respiratory syncytial virus, and L pneumophila.Many laboratories still rely on complementfixation tests (CFTs) which are time consum-ing and inconvenient to perform and have poorsensitivity and specificity. Other tests arebecoming increasingly available, but for someof the “atypical” pathogens there are noalternative assays.

There is little value in testing single serumsamples taken within 7 days of the onset of

Figure 6 Suggested algorithm for serological testing in patients with CAP.

Patient admitted to

hospital with CAP

Take 5–10 ml clottedblood sample (serum)

Initial serum

for store

Onset of symptoms<7 days before admission?

Unresponsive to beta-lactamantibiotics before admission?

Features of severe CAP?

No

No

No

No follow up serum sample.

Initial stored sample not tested

No

Unresponsive to beta-lactamantibiotics?

Specific epidemiological orpublic health indication for

performing serology?

Features of severe CAP?

No

No

Following hospital

admission with CAP

Yes

Yes Initial serum for

“atypical pathogens”

and legionella serology

Date of onset must

be clearly indicated

on request form

Yes

YesFollow up serum for

“atypical pathogen” and

legionella serology 7–10

days after admission

Initial and

follow up samples tested

in parallel

Yes

Yes


www.thoraxjnl.com



CAP. Such samples can be stored until the fol-low up (convalescent) sample is taken 7–10days later and the paired samples can be testedin parallel.

However, raised antibody titres, particularlyto L pneumophila or M pneumoniae, may befound in some patients on or soon after admis-sion to hospital, particularly if the onset ofsymptoms is more than 7 days before admis-sion. It is thus important that the date of onsetof symptoms is clearly indicated on serologicalrequest forms so that serum samples takenmore than 1 week into the illness can be testedimmediately.

A suggested algorithm for performing sero-logical investigations is shown in fig 6.

Recommendations+ Paired serological tests should be per-

formed for all patients with severe CAP,those who are unresponsive to â-lactamantibiotics, and for selected patientswith particular epidemiological riskfactors or in whom a specific microbio-logical diagnosis is important for publichealth measures [D].

+ Serological tests should be extended toall patients admitted to hospital withCAP during outbreaks and whenneeded for the purposes of surveillance.The criteria for performing serologicaltests in these circumstances should beagreed locally between clinicians, labo-ratories, and public health oYcers [D].

NON-CULTURAL TESTS FOR S PNEUMONIAE

Pneumococcal antigen detectionPneumococcal antigens can be detected invarious body fluids during active pneumococ-cal infection including sputum, pleural fluid,serum, and urine. Antigen detection is lessaVected by prior antibiotic treatment and thedetection of antigenaemia is correlated withclinical severity164 [IVb].

Various techniques have been used to detectpneumococcal polysaccharide antigens orC-polysaccharide165 [IVb]. Counterimmuno-electrophoresis is the least sensitive techniquebut has been studied the most. Latex agglutina-tion has improved sensitivity but produces poorresults with urine samples. Enzyme immu-noassays (EIAs) are promising in terms ofimproved sensitivity and specificity but havenot been rigorously evaluated, and a commer-cial immunochromatographic test for detectionof antigen in urine has recently been intro-duced.

The detection of pneumococcal antigen inserum or urine is reasonably specific but lesssensitive. Higher sensitivity is found withsputum, but specificity is compromised bycross reactions with “viridans” streptococciand false positive results due to oropharyngealcarriage of S pneumoniae.

Pneumococcal antigen detection has notbeen widely adopted in the UK due to cost,lack of sensitivity, and lack of “robustness” in aroutine diagnostic setting165 [IVb]. However,the development of a well validated technique

with good sensitivity and specificity for detec-tion of pneumococcal antigen in urine wouldcomplement the now established legionellaurine antigen test (see below) for patients withsevere CAP, and further work in this area isawaited.

Pneumococcal serologyThe detection of antibodies to the pneumococ-cal toxin pneumolysin has been reported to beboth sensitive (80–90%) and specific in thediagnosis of pneumococcal infection166 [II].The antibody response is usually delayed and,to date, this technique has been reserved forepidemiological studies.

Pneumococcal polymerase chain reactionPolymerase chain reaction (PCR) based meth-ods for detection of pneumococcal DNA inclinical samples are still under development forroutine diagnosis.

Recommendations+ There is currently insuYcient evidence

to recommend widespread use of pneu-mococcal antigen tests or serologicaltests in CAP [D].

+ Antigen tests should be used for pa-tients with severe CAP, if availablelocally [D].

TESTS FOR LEGIONNAIRES’ DISEASE

Legionella pneumonia can be severe andcarries a significant mortality. Prompt diagno-sis is important both for patient managementand for public health investigations. Riskfactors for legionella infection include recenttravel (within 10 days of onset), certainoccupations, recent repair to domestic plumb-ing systems, and immunosuppression.

Urine antigen detectionDetection of L pneumophila urinary antigen byEIA is now established as a highly specific(>95%) and sensitive (∼80%) test167 [III].Rapid results can be obtained at an early stageof the illness, and this is a valuable method forthe early diagnosis of legionella infection99

[III].Several commercial assays are available

including a rapid immunochromatographictest. These assays principally detect infectionwith L pneumophila serogroup 1 and do notdetect antigen from other Legionella species.

Legionella direct immunofluorescence (DIF) testsL pneumophila can be detected by DIF on inva-sive respiratory samples such as bronchial aspi-rates. L pneumophila specific reagents should beused and not hyperimmune rabbit antiserawhich are poorly specific. The value ofperforming DIF on expectorated sputum sam-ples is less well established.

CultureEvery eVort should be made to diagnose byculture of Legionella species from clinical sam-ples (principally respiratory samples). Cultureis 100% specific and is the only method ofdetecting infection with Legionella species other


www.thoraxjnl.com



than L pneumophila. Culture is also valuable forepidemiological investigations, allowing pheno-typic and genotypic comparison of clinical andenvironmental legionella strains.

Problems with culture include the inabilityof many patients with legionella pneumonia toproduce sputum samples; prior antibiotictreatment; laboratory time and cost in process-ing samples; and lack of rapid results (culturesneed to be incubated for up to 10 days). Manylaboratories do not set up legionella cultures onrespiratory samples unless specifically re-quested to do so.

SerologyThe diagnosis by determination of antibodylevels is well established and has been themainstay of diagnosis in the past. Serologicalassays previously employed in the UK werehighly specific, although false positive resultsdue to a serological cross reaction may occur inpatients with recent campylobacter infection168

[II]. Serological reagents for diagnosis oflegionella infection are no longer available fromthe Public Health Laboratory Service, compro-mising the use of this diagnostic method in theUK, although commercial assays are currentlyunder evaluation.

PCRDetection of legionella DNA by PCR fromrespiratory samples, blood, and urine is stillonly available as a research tool.

Recommendations+ Investigations for legionella pneumonia

are recommended for all patients withsevere CAP, for other patients with spe-cific risk factors, and for all patientswith CAP during outbreaks [D].

+ Rapid testing and reporting for le-gionella urine antigen should be avail-able in at least one laboratory perregion [D].

+ Legionella culture should be specificallyrequested by clinicians on laboratoryrequest forms from patients with severeCAP, or where Legionnaires’ disease issuspected on epidemiological grounds.

+ Legionella cultures should be routinelyperformed on invasive respiratory sam-ples (e.g. obtained by bronchoscopy)from patients with CAP [D].

TESTS FOR M PNEUMONIAE

The mainstay of diagnosis at the present time isby serological testing. Culture of M pneumoniaeis generally not available in diagnostic laborato-ries.

The most common serological assay used isthe CFT but various alternative assays such asmicroparticle agglutination and EIAs are alsoavailable. The CFT is still regarded as the“gold standard” to which other assays havebeen compared, although it does lack somesensitivity and specificity. A comparison ofvarious mycoplasma antibody assays (includ-ing IgM and CFTs) concluded that no singleassay has significantly better sensitivity andspecificity than the others169 [III].

Raised CFT titres are usually detected noearlier than 10–14 days after the onset ofmycoplasma infection, but the insidious onsetand slow progression of symptoms means thatmany patients admitted to hospital with myco-plasma CAP have raised titres on or shortlyafter admission.

Genomic detection of M pneumoniae inrespiratory specimens by amplification tech-niques such as PCR is currently underdevelopment.

Recommendation+ Serological assay with CFTs is widely

available and should remain the main-stay of diagnosis [C].

TESTS FOR CHLAMYDIA SPECIES

CultureIt is not appropriate for routine diagnosticlaboratories to attempt culture of Chlamydiaspecies from respiratory samples from patientswith CAP as special laboratory precautions arerequired. C psittaci is a “category 3 pathogen”indicating a high risk pathogen that may putlaboratory staV at risk of serious illness ifinfected occupationally. C pneumoniae is verydiYcult to grow in the laboratory; culture isslow, time consuming, expensive, and insensi-tive.

Antigen detectionChlamydial antigen can be detected in respira-tory samples using direct immunofluorescence(DIF) with species and genus specific mono-clonal antibodies170 [II]. Genus specific re-agents are not available for C psittaci which isantigenically highly diverse. DIF requiresexpertise in slide preparation and reading andis not available in all diagnostic laboratories. Cpneumoniae can also be detected by DIF onthroat swabs with a comparable sensitivity tosputum.171 However, antigen may be detectedfor several months after “acute” infection,making interpretation diYcult.

Chlamydial antigen can also be detected inrespiratory samples by EIA with a comparablesensitivity to PCR,172 but this approach re-quires further study.

SerologyVarious serological assays are used in the diag-nosis of respiratory chlamydial infections. TheCFT is available in most diagnostic serologylaboratories. Microimmunofluorescence(MIF) and whole cell immunofluorescence(WHIF) are specialised reference tests. SeveralEIAs have been described and at least one iscommercially available in the UK. Each ofthese assays has advantages and disadvantages,and there are particular problems in theserological diagnosis of C pneumoniae infec-tions.

The CFT uses a genus specific antigen and isrelatively sensitive and specific for diagnosingpsittacosis. However, in adults, most infectionswith C pneumoniae are re-infections and thesegenerate only a weak or absent CFT response.The MIF and WHIF tests require considerableexperience to read and interpret. They can


www.thoraxjnl.com



detect a species specific response, although thismay be delayed for 4–6 weeks, especially with Cpneumoniae re-infections. They may also miss Cpsittaci infections, depending on the particularserovars included in the test, and there areconflicting reports regarding the accuracy ofthese tests in reliably distinguishing chlamydialspecies173 [IVb].

A commercial EIA has been used withsuccess174 [III] but has not been shown to besignificantly superior to the CFT.

Molecular techniquesAmplification of chlamydial DNA by PCRusing genus or species specific primers hasbeen reported from a variety of respiratorysamples, but these molecular techniques are

still confined to research/reference laboratoriesat the present time.

Recommendations+ Chlamydial antigen detection tests

should be available for invasive respira-tory samples from patients with severeCAP or where there is a strong suspi-cion of psittacosis [D].

+ The CFT remains the most suitable andpractical serological assay for routinediagnosis of respiratory chlamydial in-fections [B–]. There is no currentlyavailable serological test that canreliably detect infections due to Cpneumoniae.


www.thoraxjnl.com



6 Severity assessment

6.1 Why is severity assessmentimportant?CAP presents to physicians both in primaryand secondary care as a wide spectrum ofillness from mild and self-limiting to a lifethreatening and occasionally fatal disease. Thisbreadth of illness severity is reflected in thevariable mortality rates reported by studies ofCAP in diVerent clinical settings.

The decision regarding the most appropriatesite of care, including whether admission tohospital is warranted, is the first and singlemost important decision in the overall manage-ment of CAP. It has consequences both for thelevel of treatment received by the patient aswell as the overall costs of treatment63 [III].This decision is best informed by an accurateassessment of the severity of illness at presenta-tion and the likely prognosis. Recognition ofpatients at low risk of complications and there-fore suitable for outpatient treatment has thepotential for reducing inappropriate hospitali-sation and consequent inherent morbidity andcosts.

When hospital admission is required, furthermanagement is also influenced by illness sever-ity. This includes the extent of microbiologicalinvestigation, the choice of initial empiricalantimicrobial agents, route of administration,duration of treatment, and level of nursing andmedical care. Early identification of patients athigh risk of death allows initiation of appropri-ate antibiotic treatment and admission to anICU where assisted ventilation can be readilyinitiated if necessary.

6.2 What clinical factors andinvestigations are associated with a poorprognosis on univariate and multivariateanalysis?A large number of studies conducted in hospi-tal and ICUs have employed univariate analysisto identify risk factors associated with a poorprognosis. In hospital mortality has been themost common outcome measure. Some studieshave used admission to the ICU as the mainoutcome measure175 [Ib]. However, diVerencesin ICU admission criteria make it diYcult tocompare results from these studies. This isreflected in the widely varying rates ofadmission of patients with CAP to ICU, rang-ing from 1–3% in New Zealand, 5% in the UK,12–18% in the USA, to 35% in Germany14

[Ib], 43 [Ib], 71 [Ib], 176 [II], 177 [III], 178 [III]. Inthese guidelines we have concentrated only onstudies that have used mortality as the mainoutcome measure.

Univariate studies have suggested that over40 diVerent parameters are associated withmortality. However, an independent associ-ation of only a few of these risk factors withmortality have been demonstrated by studiesemploying multivariate analysis.

AGE

Age has consistently been shown to be signifi-cantly associated with mortality in studies froma variety of countries14 [Ib], 43 [Ib], 71 [Ib],179 [Ia], 180 [Ib], 181 [Ib], 182 [III]. However,recent studies of CAP in the elderly (>65 years)have suggested that, within this population, ageby itself is not of prognostic importance183 [Ib],184 [Ib], 185 [II]. Importantly, the lack of pyrexia,which is easily overlooked in clinical practice,has been identified by some as an importantprognostic factor13 [II], 184 [Ib]. Being previ-ously bedridden and having been admittedfrom a nursing home are other poor prognosticfactors that may be interrelated and warrantattention43 [Ib], 135 [II], 184 [Ib]. Admission cri-teria to nursing homes may diVer in the UKfrom the countries where these studies wereconducted, so validation of these findings isdesirable.

Based on current evidence, there is no firmreason to support the use of a diVerent set ofclinical features in assessing disease severity inthis group of patients.

PRESENCE OF COEXISTING ILLNESSES

Concomitant congestive cardiac failure181 [Ib],186 [III], coronary artery disease186[III], stroke43

[Ib], 179 [III], 186 [III], diabetes mellitus179 [Ia],187 [II], chronic lung disease11 [II], 187 [II],cancer43 [Ib] and other coexisting illnesses havebeen shown to predict death in patients withCAP180 [Ib], 188 [III], 189 [III]. However, themagnitude of the contribution of these coexist-ing illnesses to disease severity is diYcult toascertain due to variations in disease definitionand problems in determining the severity ofthese conditions themselves. This may partlyexplain the low predictive power of thepresence of coexisting illnesses as a risk factorfor death on multivariate analysis, despite thelarge number of studies that have demon-strated significance on univariate analysis.

RESPIRATORY RATE

A raised respiratory rate has emerged as one ofthe most reliable indicators of disease severityin both univariate and multivariate analysisacross all age groups14 [Ib], 43 [Ib], 71 [Ib],177 [III], 179 [Ia], 184 [Ib], 189 [Ib], 190 [Ib]. Al-though a linear correlation of respiratory ratewith mortality has been reported191 [III], inclinical practice a respiratory rate of 30breaths/min or more is accepted as indicatingsevere disease14 [Ib], 71 [Ib], 189 [Ib]. Increasedawareness and better documentation of thisvaluable clinical sign is strongly recommended.

MENTAL STATUS

An altered mental state has been identified asan independent risk factor for mortality in atleast two large studies14 [Ib], 180 [III]. It alsoappears to be useful in the elderly, althoughdiVerences in definition make integration ofresults diYcult182 [III], 184 [Ib]. The more


www.thoraxjnl.com



widespread use of the Abbreviated Mental Testscore192 [II] (summarised in box 2) as a meas-ure of acute confusion will help futurevalidation of the importance of this clinical signas part of a severity prediction rule71 [Ib].

The Abbreviated Mental Test (eachquestion scores 1 mark, total 10 marks)+ Age+ Date of birth+ Time (to nearest hour)+ Year+ Hospital name+ Recognition of two persons (e.g. doctor,

nurse)+ Recall address (e.g. 42 West Street)+ Date of First World War+ Name of monarch+ Count backwards 20 → 1

Box 2 Summary of questions asked of a patientto assess the degree of mental confusion as part ofthe severity assessment. A score of 8 or less has beenused to define mental confusion in the modifiedBTS severity rule.

BLOOD PRESSURE

A low systolic (<90 mm Hg) or diastolic(<60 mm Hg) blood pressure or the presenceof septic shock on admission to hospital arerecognised poor prognostic factors13 [II],14 [Ib], 71 [Ib], 114 [III], 179 [Ia], 182 [III], 184 [Ib],187 [II], 189 [III], 193 [II]. In the intensive caresetting septic shock, defined as (a) a sustaineddecrease (for at least 1 hour) in systolic bloodpressure of at least 40 mm Hg from baseline or(b) a resultant systolic blood pressure of<90 mm Hg after adequate volume replace-ment in the absence of antihypertensive drugsoccurring more than 12 hours after admissionto the ICU is an additional adverse sign74 [II],185 [II].

OXYGENATION

Hypoxaemia is an important adverse findingacross all ages and the need to apply positiveend expiratory pressure (PEEP) or an inspiredoxygen concentration of over 60% to maintainadequate oxygenation are both poor prognosticindicators67 [II], 71 [Ib]. Respiratory failure andthe need for mechanical ventilation, whetherupon admission to the ICU or subsequently, isa predictor of mortality in itself 43 [Ib], 67 [II],74 [Ib], 75 [II].

WHITE CELL COUNT

Both leucopenia (white cell count <4 × 109)and leucocytosis (>20 × 109) on admissionhave been associated with mortality on univari-ate analysis177 [III], 182 [III], 188 [III], 194 [III].Results from multivariate analyses have beenvariable and suggest that leucopenia may be themore important variable14 [Ib], 179 [Ia].

RADIOGRAPHIC CHANGES

Bilateral involvement or the involvement ofmore than two lobes on the chest radiographare poor prognostic factors43 [Ib], 74 [III, Ib],179 [Ia], 184 [Ib], 188 [III], 193 [II]. The presence

of bilateral parapneumonic eVusions has alsobeen associated with an increased risk of deathin one large study195 [Ib].

In patients admitted to an ICU, progressionof chest radiographic changes is associated withmortality67 [II]. However, outside an ICUrepeated radiological investigations are notroutinely recommended as a means of severityassessment unless otherwise indicated.

MICROBIOLOGY

A positive blood culture, regardless of thepathogen isolated, is associated with a pooroutcome43 [Ib], 67 [II], 74 [Ib], 75 [II], 179 [Ia],188 [III], 194 [III]. In bacteraemic patients,pathogens particularly associated with severedisease include S pneumoniae, Gram negativeenteric bacilli, S aureus, and P aeruginosa11 [II],43 [Ib], 67 [II], 75 [II]. Although Legionella spe-cies have not been shown to be specificallyassociated with increased mortality in studiesemploying multivariate analysis of risk factors,it is the second most common pathogenisolated in patients with CAP admitted toICUs in the UK (see section 6). It shouldtherefore always be considered in patients withsevere pneumonia. However, as clinical fea-tures at presentation are generally unhelpful inaccurately diVerentiating between causativepathogens in CAP, patients should be managedaccording to other more readily availablepredictors of severity in the first instance.

6.3 What predictive models for assessingseverity on or shortly after hospitaladmission have been tested?Clinical assessment of disease severity isdependent on the experience of the attendingclinician, but such clinical judgement has beenshown to result in apparent underestimation ofseverity71 [Ib]. No single prognostic factor ofmortality is adequately specific and sensitive.Various severity scoring systems and predictivemodels have therefore been developed in anattempt to help the clinician identify patientswith pneumonia and a poor prognosis at anearly stage.

None of the predictive models developedthus far allow the unequivocal categorisation ofpatients into definite risk groups and it is unre-alistic to expect this196 [IVb]. Until recently, thefew studies that have examined the impact ofseverity based practice guidelines on clinicaloutcomes have not demonstrated significantbeneficial eVects22 [III], 54 [II], 197 [II]. How-ever, new data are emerging to suggest thatsuch practice guidelines may be useful in iden-tifying patients with CAP who are suitable forambulatory outpatient care and who thereforedo not require hospital admission198 [II]. Astudy of a clinical practice guideline embracinga strict antibiotic policy together with criteriafor hospital admission and for regular reviewand the switch from intravenous to oral antibi-otics resulted in reductions in hospital admis-sion, use of intravenous antibiotics, and lengthof hospital stay199 [Ib]. These encouragingresults are preliminary and will need confirma-tion in a wider setting such as the UK.


www.thoraxjnl.com



CONCLUSIONS REGARDING PREDICTIVE MODELS

+ None of the available predictive modelsallows the unequivocal categorisation ofpatients into definite risk groups.

+ Predictive models based on severity are bestviewed as useful adjuncts to clinical assess-ment.

+ Regular reassessment of severity during thecourse of hospital stay is mandatory if treat-ment is to be adjusted appropriately, avoid-ing the morbidity of overtreatment as well asthe complications of undertreatment.

FURTHER DETAILS OF TWO PREDICTIVE MODELS

Predictive models have been developed toaddress two main areas in the management ofCAP.

Approach A: Predicting patients at low risk ofdeathBased on work carried out in the USA, a Pneu-monia Severity Index (PSI) for stratifyingpatients into five risk classes according to therisk of mortality has been published189 [Ib].Patients identified as being in risk class I are atlow risk of death and adverse outcomes, with amortality risk of 0.1–0.4%. In an observationalstudy of those patients in risk class I judged tobe suitable for outpatient treatment, the rate ofsubsequent hospitalisation within 30 days was5.5%65 [II].

This index relies on the identification of two“pre-existing” patient features (age over 50years and the presence of certain coexistingchronic illnesses) and five adverse clinicalfeatures (mental state, respiratory rate, systolicblood pressure, pulse rate, and temperature—with twice as much risk weighting being placedon the first three features) for the categorisa-tion of patients into risk class I. Its use in theprimary care setting to identify patients at lowrisk of death is therefore feasible although atpresent there are no UK validation studies. Theomission of chronic lung disease (CLD) as anadverse chronic illness in the PSI is notable andis based on the absence of an independentassociation of chronic lung disease withmortality in the studies from which this rulewas derived. This may be because the contribu-tion of underlying chronic lung disease to clini-cal outcome is partly accounted for bymeasures of respiratory rate and oxygenationstatus. In a clinical intervention trial conductedin Canada in which about 30% of patients hadchronic lung disease, no harm was demon-strated from using the PSI to stratify patientsinto groups suitable for home or hospital man-agement199 [Ib].

Approach B: Predicting patients at high risk ofdeathIn 1987 the BTS derived a rule for predictingmortality based on three easily measured clinicalparameters14 [Ib]. Patients were found to have a21-fold increased risk of death if they had two ormore of the following “core” adverse prognosticfeatures: (a) admission respiratory rate >30/min, (b) admission diastolic blood pressure<60 mm Hg, and (c) blood urea >7 mmol/lduring admission. The sensitivity of the rule is

88%, specificity 79%, false negative rate 12%,false positive rate 21%, positive predictive value(PPV) 19%, and negative predictive value(NPV) 99%.

This rule has been validated by groups in theUSA and New Zealand and appears to remainuseful when all three parameters are measuredat the time of admission. From these studiesthe BTS rule has an overall sensitivity and spe-cificity of approximately 80%71 [Ib], 177 [III].However, there are some recent data to suggestthat the BTS rule is less sensitive at predictingmortality in older patients (65 years andabove)182 [III].

A modification of this rule to add mentalconfusion on hospital admission as a fourth“core” adverse prognostic feature was firstevaluated in 199170 [II]. Since then, incorpora-tion of confusion measured as a score of 8 orless on a 10 point Abbreviated Mental TestScore (box 2)192 [II] has been shown to increasethe sensitivity of the rule at the expense ofspecificity69 [II], 71 [Ib]. This modified BTSrule has recently been prospectively validatedin the UK in adults of all ages69 [Ib]. Combin-ing results from the derivation and validationstudies yields a sensitivity of 83%, specificity70%, negative predictive value 97%, and posi-tive predictive value 26%. The number of“core” adverse prognostic features present cor-related well with mortality, with a mortality of2.4% in the presence of no “core” features, 8%with one “core” feature, 23% with two, 33%with three, and 83% with all four “core”features present69 [Ib]. It has been proposedthat these four “core” adverse prognosticfeatures could be remembered as the CURBseverity score (Confusion, Urea, Respiratoryrate, Blood pressure)69 [Ib].

The use of the APACHE II scoring system aswell as other specialised ICU scoring systemsfor organ system failure in CAP has beendescribed to predict outcome74 [IIb], 154 [Ib],185 [IIb]. However, general use of these scoringsystems outside the intensive care setting is dif-ficult, time consuming, and likely to beimpractical.

6.4 What severity assessment strategy dowe recommend for CAP?We have been keen to recommend one severityassessment strategy that is simple to rememberand practical to implement both in thecommunity and in hospital and that is adaptedfrom the results of studies previously discussed.

This strategy places patients into threegroups including those with:(a) low risk of death (0.1–0.4%): they may besuitable for either treatment at home, as anoutpatient or for early discharge from themedical assessment unit or medical ward;(b) high risk of death (22–30%): they should bemanaged using the recommendations forsevere pneumonia;(c) some increased risk compared with the lowrisk group because they demonstrate some indi-vidual adverse prognostic features independ-ently associated with death, but for which a riskratio cannot be calculated from a compilation ofthe data in the available literature. Clinical


www.thoraxjnl.com



judgement is essential when deciding on themanagement of all patients with CAP, and willbe particularly important for this group.

We emphasise the importance of assessingand recording the presence of “core” adverseprognostic features for all patients, whether inthe community or in hospital, the presence ofany of which has been associated with anincreased risk of death in studies. Patients whohave two or more “core” adverse prognosticfeatures and who therefore fulfil the modifiedBTS severity prediction rule are at high risk ofdeath. Where the risk assessment using themodified BTS severity prediction rule isunclear, clinical judgement can also be helpedby assessing “additional” adverse prognosticfeatures that have been independently associ-ated with outcome. Three of the “core” adverseprognostic features are derived from simpleclinical observations and therefore will bereadily available to doctors when assessingpatients in the community. Age less than 50years and the absence of coexisting disease arealso useful “pre-existing” features for identify-ing those with a very good prognosis who maybe suitable for home treatment.

We also recommend how and when to reviewseverity status after initial assessment.

SUMMARY OF THE FOUR “CORE” ADVERSE

PROGNOSTIC FEATURES (CURB SCORE)69

These should be assessed for all patients:+ Confusion: new mental confusion defined as

an Abbreviated Mental Test score of 8 orless [Ib]

+ Urea: raised >7 mmol/l (for patients beingseen in hospital) [Ib]

+ Respiratory rate: raised >30/min [Ib]+ Blood pressure: low blood pressure (systolic

<90 mm Hg and/or diastolic <60 mm Hg)[Ib]

SUMMARY OF “ADDITIONAL” ADVERSE

PROGNOSTIC FEATURES

These can aid clinical judgement as they arereadily available for patients in hospital.Oxygen saturation measurements may beavailable to some general practitioners in thecommunity who have oximeters.+ Hypoxaemia (SaO2 <92% or PaO2 <8 kPa),

regardless of inspired oxygen concentration[Ib]

+ Bilateral or multilobe involvement on thechest radiograph [Ib]

SUMMARY OF “PRE-EXISTING” ADVERSE

PROGNOSTIC FEATURES

Absence of these features, which are easilyidentified in all patients, can be useful in iden-tifying those at low risk of death:+ Age 50 years and over [Ib]+ Presence of coexisting disease [Ib]

6.5 Identifying those patients who canusually be safely treated at home (alsosummarised in fig 7)

SummaryPatients are at low risk of death if they fulfil thefollowing three criteria and hence do notdisplay any adverse prognostic features:

Figure 7 Severity assessment used to determine the management of CAP in patients in the community.*The social circumstances and wishes of the patient should also be considered.

Home

management*

Refer to

hospital*

Consider “core” adverse prognostic features:

• new mental confusion• respiratory rate 30/min or more• BP <90 mmHg systolic or ≤60 mmHg diastolic

1 featurepresent

Nonepresent

Clinical judgement

Consider “pre-existing”

adverse prognostic features:

• Age 50 years or over or• Any coexisting chronic illness?

2 or morefeatures present

NO

YES

Consider “additional”

adverse prognostic feature

if available:

• SaO2 <92%


www.thoraxjnl.com



+ Age less than 50 years [Ib]+ No coexisting disease—that is, they do not

have any one of congestive cardiac failure,cerebrovascular disease, chronic renal dis-ease, chronic liver disease, or neoplastic dis-ease [Ib]

+ No “core” adverse prognostic features [Ib].In practice, the general practitioner will notusually have the result of blood ureameasurements and will rely on assessingthe three simple “core” clinical features ofconfusion, respiratory rate, and bloodpressure.

Recommendations+ Patients who display no adverse prog-

nostic features are at low risk of deathand do not normally require hospitali-sation for clinical reasons [D].

+ Patients who display two or more“core” adverse prognostic features areat high risk of death and should bereferred urgently to hospital [D].

+ For all other patients the decision totreat at home or refer to hospital is amatter of clinical judgement [D].

+ When deciding on home treatment, thepatient’s social circumstances andwishes must be taken into account in allinstances [D].

6.6 Identifying those with severe CAPfrom those with non-severe CAP inhospital (also summarised in fig 8)

Summary+ The key to identifying those with severe

pneumonia remains assessing the four“core” adverse prognostic features [Ib].

+ “Pre-existing” and “additional” adverseprognostic features also relate to outcome[Ib] and their assessment can be helpful toclinical judgement.

Recommendations+ Patients who have two or more “core”

adverse prognostic features (fig 7) are athigh risk of death and should bemanaged as having severe pneumoniaaccording to the recommendations out-lined in sections 7 and 8 [A–].

+ Patients who display one “core” adverseprognostic feature are at increased riskof death. The decision to treat suchpatients as having severe or non-severepneumonia is a matter of clinical judge-ment, preferably from an experiencedclinician. This decision can be assistedby considering “pre-existing” prognos-tic features (including age and the pres-ence of pre-existing chronic illness) and“additional” adverse prognostic features(including hypoxaemia and the extent ofradiographic shadowing) [D].

Figure 8 Severity assessment used to determine the management of CAP in patients admitted to hospital.*The social circumstances and the wishes of the patient and general practitioner should also be considered.

Consider managingas outpatient*

Manage in hospitalas severe CAP

Manage in hospitalas non-severe CAP*

Consider “core” adverse prognostic features:

• new mental confusion• urea >7 mmol/l• respiratory rate 30/min or more• systolic BP <90 mmHg or diastolic BP ≤60 mmHg

1 featurepresent

Nonepresent

Clinical judgement

Consider “pre-existing”


• Age 50 years or over or• Any coexisting chronic illnesses?

Consider “additional”


• PaO2 <8 kPa/SaO2 <92% (any FiO2)• CXR: bilateral/multilobar shadowing

2 or morefeatures present

NO

YES


www.thoraxjnl.com



+ Patients who display no adverse prog-nostic features can be managed as hav-ing non-severe pneumonia and may besuitable for outpatient treatment orearly hospital discharge [B+].

6.7 Reviewing severity status after initialassessment

Summary+ Regular and structured clinical review and

reassessment of disease severity facilitatesthe stepping down and stepping up ofantibiotic management [Ib].

Recommendations+ Regular assessment of disease severity

is recommended for ALL patientsfollowing hospital admission. The“post take” round by a senior doctorand the medical team provides oneearly opportunity for this review[D].

+ All patients who display one or more“core” adverse prognostic features onadmission should be reviewed medicallyat least 12 hourly until shown to beimproving [D].


www.thoraxjnl.com



7 General management

7.1 What general management strategyshould be oVered to patients treated inthe community?Patients with CAP may present with fever,cough, sputum production or pleuritic painand usually have localised signs on chest exam-ination. They should be advised to rest andavoid smoking200 [IIb] and, especially whenfebrile, be encouraged to drink plenty of fluids.It is important to relieve pleuritic pain usingsimple analgesia such as paracetamol ornon-steroidal anti-inflammatory drugs. Physio-therapy is of no proven benefit in acutepneumonia201 [III]. Nutritional status appearsimportant both to the outcome and the risk ofacquiring pneumonia and nutritional supple-ments may be helpful in prolonged illness.Patients with pneumonia are often catabolicand those over 55 years who are malnourishedappear to be at greater risk of developing pneu-monia115 [III], 202 [III].

Patients with pneumonia often becomehypoxic because pulmonary blood flow takesplace through unventilated lung tissue. Theclinical signs of hypoxia are non-specific andoften diYcult to recognise in the early stages.They include altered mental state, dyspnoea,and tachypnoea. Respiratory rate should there-fore always be assessed. Central cyanosis isunreliable both as a clinical sign and also as anindicator of tissue hypoxia. In contrast, pulseoximetry which measures arterial oxygen satu-ration (SaO2) is in most situations a simple andreliable method of assessing oxygenation.However, poor peripheral perfusion, jaundice,and pigmented skin can produce a falsely lowsaturation, and carboxyhaemoglobin can resultin a falsely high saturation. It is recommendedthat pulse oximetry, with appropriate training,should become more widely available ingeneral practice for use in the assessment ofpatients who may have pneumonia and otheracute respiratory illnesses. SaO2 below 92% in apatient with CAP is an “additional” adverseprognostic feature (see section 6.4) and also anindication for oxygen therapy203 [IVb] whichwill usually require urgent referral to hospital.

Patients who fall outside the low risk severitycriteria for CAP should be assessed for theneed for hospital referral (section 6). Socialfactors will also play an important part in thedecision to refer a patient to hospital. Patientswith severe pneumonia should be admitted tohospital and managed, where possible, withinput from a physician with an interest inrespiratory medicine.

Recommendations+ Patients with suspected CAP should be

advised not to smoke, to rest, and todrink plenty of fluids [D].

+ Pleuritic pain should be relieved usingsimple analgesia such as paracetamol[D].

+ Nutritional supplements should be con-sidered in prolonged illness [C].

+ The need for hospital referral should beassessed using the criteria recom-mended in section 6 [C].

+ Pulse oximetry, with appropriate train-ing, should become increasingly avail-able to general practitioners for assess-ment of severity and oxygenrequirement for patients with CAP andother acute respiratory illnesses [D].

7.2 What review policy should be adoptedin patients managed in the community?When to review a patient with CAP in thecommunity will be determined by the initialseverity assessment and other factors such asreliable help in the home. Patients assessed asbeing at low risk should improve on appropri-ate treatment within 48 hours, at which timeseverity reassessment is recommended. Thosewho fail to improve within 48 hours should beconsidered for hospital admission. Patientswho do not fulfil the criteria for low risk sever-ity and are being managed at home will requiremore frequent review.

Recommendations+ Review of patients in the community

with CAP is recommended after 48hours or earlier if clinically indicated.“Core” and “additional” adverse prog-nostic features should be assessed aspart of the clinical review [D].

+ Those who fail to improve after 48 hourstreatment should be considered forhospital admission or chest radio-graphy [D].

7.3 What general management strategyshould be oVered to patients in hospital?INITIAL MANAGEMENT

There is some evidence that use of a criticalcare pathway for patients referred to hospitalcan reduce the hospital admission of “low risk”patients and can also rationalise inpatientmanagement199 [Ib].

All patients referred to hospital with CAPshould have a chest radiograph (if not alreadyperformed in the community) and should haveoxygenation assessed by pulse oximetry, prefer-ably while breathing air. Those with SaO2

<92% should have arterial blood gas measure-ments, as should all patients with features ofsevere pneumonia. Knowledge of the inspiredoxygen concentration is essential to theinterpretation of blood gas measurements andshould be clearly recorded with the blood gasresult.

Continuous oxygen therapy is indicated forthose patients with PaO2 <8 kPa, hypotensionwith systolic blood pressure <100 mm Hg,metabolic acidosis with bicarbonate<18 mmol/l, or respiratory distress with arespiratory rate of >24 breaths/min203 [IVb].The aim of oxygen therapy should be to main-tain PaO2 at >8 kPa or SaO2 >92%. In nearly all


www.thoraxjnl.com



cases of CAP, unless complicated by severechronic obstructive pulmonary disease withventilatory failure, high concentrations of oxy-gen of 35% or more are indicated and can besafely used.

High concentration oxygen therapy given topatients with pre-existing chronic obstructivepulmonary disease who may have carbon diox-ide retention can reduce hypoxic drive andincrease ventilation-perfusion mismatching. Insuch patients initial treatment with low oxygenconcentrations (24–28%) should be progres-sively increased on the basis of repeated arterialblood gas measurements, the aim being to keepPaO2 >6.65 kPa without causing a fall inarterial pH below 7.26204 [IVb], in line with themanagement strategy recommended in theBTS guidelines on the management ofCOPD.39 In severe cases non-invasive ventila-tion or respiratory stimulants may be of valueand transfer to a high dependency unit or ICUshould be considered.

Patients admitted with pneumonia should beassessed for volume depletion and may requireintravenous fluids. While physiotherapy is of noproven value in acute pneumonia205 [III], a sin-gle randomised trial has suggested that “bottle-blowing” into an underwater seal bottle on 10occasions daily can shorten the time inhospital206 [II]. There is also evidence that, inselected patients, particularly those with con-current COPD, continuous positive airwayspressure (CPAP) treatment can occasionally beof additional value207 [II]. In cases of severepneumonia requiring prolonged hospital ad-mission, increased nutritional support—whether enteral, parenteral, or via a nasogastrictube—should be arranged.

Recommendations+ All patients should receive appropriate

oxygen therapy with monitoring of oxy-gen saturations and inspired oxygenconcentration (see section 5.3) with theaim of maintaining PaO2 at >8 kPa andSaO2 at >92%. High concentrations ofoxygen can safely be given in uncompli-cated pneumonia [D].

+ Oxygen therapy in patients withpre-existing COPD complicated byventilatory failure should be guided byrepeated arterial blood gas measure-ments [C].

+ Patients should be assessed for volumedepletion and may require intravenousfluids [C].

+ Nutritional support should be given inprolonged illness [C].

MONITORING IN HOSPITAL

Pulse, blood pressure, respiratory rate, tem-perature, oxygen saturation (with a recordingof the inspired oxygen concentration at thesame time), and mental status should bemeasured initially at least twice daily. Thosewith severe pneumonia who require continu-ous oxygen or cardiovascular support shouldbe monitored more frequently.

The acute phase reactant CRP is a sensitivemarker of progress in pneumonia153 [III],

154 [III]. A fall in CRP of 50% over 4 days isconsistent with a good clinical response. A fallin the level of CRP of less than 50% or apersistently high or rising white cell countsuggests failure of antibiotic treatment ordevelopment of an infective complication suchas an empyema (see sections 5.6 and 9).

Failure to improve over 4 days is anindication to repeat the chest radiograph.

Recommendations+ Temperature, respiratory rate, pulse,

blood pressure, mental status, oxygensaturation, and inspired oxygen con-centration should be monitored andrecorded initially at least twice dailyand more frequently in those withsevere pneumonia or requiring regularoxygen therapy [C].

+ The CRP level should be remeasured[B–] and the chest radiograph repeated[C] in patients who are not progressingsatisfactorily.

7.4 What advice should be givenregarding ICU management of CAP?Severity assessment is an important part ofhospital management as it can identify thosepatients at increased risk of death (section 6).Patients who fulfil the severity criteria forsevere CAP on admission and who do notrespond rapidly should be considered fortransfer to a high dependency unit or an ICU(see section 6). Persisting hypoxia with PaO2

<8 kPa despite maximal oxygen administra-tion, progressive hypercapnia, severe acidosis(pH <7.26), shock, or depressed consciousnessare also indications for transfer to the ICU forassisted ventilation and cardiovascular sup-port208 [IVb].

Bronchoscopy after intubation may be valu-able to remove retained secretions, to obtainfurther samples for culture, and to excludeendobronchial abnormalities such as carci-noma. Hydrocortisone given to patients in theICU with severe pneumonia does not alter lev-els of circulating inflammatory cytokines suchas tumour necrosis factor (TNF)á209 [III], andthere is no evidence that systemic cortico-steroids are of benefit. Hospital acquired venti-lator associated pneumonia can occur inapproximately 14% of patients mechanicallyventilated for severe CAP and causes increasedmortality80 [III]. Other aspects of ICU man-agement are outside the scope of theseguidelines.

Recommendations+ Patients with CAP admitted to ICU

should be managed by specialists withappropriate training in intensive careand respiratory medicine [D].

+ Bronchoscopy can be valuable to re-move retained secretions, obtain sam-ples for culture for other microbiologi-cal investigations, and to excludeendobronchial abnormality [C].


www.thoraxjnl.com



7.5 What arrangements should be madefor follow up after hospital discharge andby whom?It is usual practice to arrange “routine” followup at a hospital clinic and to repeat the chestradiograph about 6 weeks after discharge.However, there is no evidence on which to basea recommendation regarding the value of thispractice in patients who have otherwise recov-ered satisfactorily. It is also not known whetherthere is any value in arranging clinical follow upin a hospital clinic rather than with the patient’sgeneral practitioner. The main concern iswhether the CAP was a complication of anunderlying condition such as lung cancer (seesection 5.3).

At discharge or at follow up patientsshould be oVered access to information aboutCAP (see section 1.9). In one study of200 patients who had recently recoveredfrom CAP a patient information leaflet wasjudged to be very helpful by the majority ofpatients46 [III]. An updated leaflet on CAP isavailable on request from the British LungFoundation headquarters (78 Hatton Gardens,London EC1N 8LD, UK) and UK regionaloYces.

Recommendations+ Clinical review should be arranged for

all patients at around 6 weeks, eitherwith their general practitioner or in ahospital clinic [D].

+ At discharge or at follow up patientsshould be oVered access to informationabout CAP such as a patient infor-mation leaflet [D].

+ It is the responsibility of the hospitalteam to arrange the follow up plan withthe patient and the general practitioner[D].

+ Guidelines on whether or not to repeatthe chest radiograph or perform fur-ther investigations at that time aregiven in section 5.3:+ A chest radiograph should be ar-

ranged at that time for those patientswho have persistence of symptoms orphysical signs or who are at higherrisk of underlying malignancy (espe-cially smokers and those over 50years) [C].

+ Further investigations which mayinclude bronchoscopy should be con-sidered in patients with persistingsigns, symptoms, and radiologicalabnormalities about 6 weeks aftercompleting treatment [C].


www.thoraxjnl.com



8 Antibiotic management

8.1 IntroductionAntimicrobial chemotherapy is essential to themanagement of CAP. While mild pneumoniamay be self-limiting, the timely use of appropri-ate antibiotics abbreviates illness, reduces therisk of complications, and lowers mortality.

Few pneumonias are defined microbiologi-cally at initial assessment and hence most pre-scribing is empirical, especially when managedin the community. In hospitalised patients theaetiology may be determined, thereby permit-ting modification of the initial empiricalregimen. However, in practice this applies tothe minority of infections158 [II]. Clinical,epidemiological, and radiographic informationis rarely predictive of the microbial aetiology.

The pathogens responsible for CAP arediverse and vary in their ability to cause severedisease179 [Ib]. For example, bacteraemicpneumococcal pneumonia caused by serotype3 is more likely to cause fatal disease than CAPcaused by M pneumoniae.

Severity assessment and the association ofpre-existing co-morbid disease is essential inpredicting prognosis and, in turn, determinesmanagement, choice of antibiotic therapy, andits method of administration (see section 6).

Age will also have some influence on thelikely causative pathogens and consequentlythe choice of empirical therapy in somecircumstances. For the purpose of this discus-sion on antibiotic management we havearbitrarily chosen the term “elderly” to includethose aged 75 years and over.

8.2 The term “atypical” pathogenThe definitions used are described in sections1.7 and 4.4. In summary, for the purposes ofthese guidelines infections caused by Mpneumoniae, C pneumoniae, C psittaci, and Cburnetii are captured in the phrase “atypical”pathogens. Legionella species, although sharingsome similar characteristics, are not consideredto be an “atypical” pathogen for the purpose ofthis document. We do not recommend the useof the term “atypical” pneumonia.

8.3 Local issues aVecting the choice ofantibiotic regimenThe choice of antibiotic regimen may have con-sequences beyond the management of the indi-vidual patient. Overprescribing of macrolidesand â-lactams, especially when administeredparenterally in the management of hospitalisedpatients with CAP, possibly resulted from tooloose an interpretation of “severe pneumonia”.Furthermore, the application of the earlierpneumonia guidelines to community acquiredlower respiratory tract infections other thanpneumonia and hospital acquired pneumoniaadded to this overuse. This has increased thecost of management and, in the case ofcefotaxime and other injectable cephalosporins,has been linked to an increase in complicating CdiYcile associated diarrhoea and enteropathy16

[II], 19 [Ib], 210 [IVa], 211 [II], 212 [II], 213 [Ib],214 [Ia], 215 [Ib]. Alternative agents such as theaminopenicillins, with or without a â-lactamaseinhibitor, can also induce C diYcile associateddiarrhoea216 [Ia]. Although the excess and inap-propriate use of such antibiotics for treatinglower respiratory tract infections has been veryimportant, C diYcile associated diarrhoea is alsolinked to hospital hygiene and cross infectionpractices217 218 [IVa] and is particularly prevalentin elderly patients on medical wards. An impor-tant aspect of implementing these guidelineslocally and auditing their use will be theemphasis with which they are directed at themanagement of pneumonia and not at non-pneumonic lower respiratory tract infectionssuch as exacerbations of COPD. Hospitals thatcontinue to have a problem with C diYcileenteropathy in spite of attention to the appro-priate use of these guidelines for pneumoniashould consider choosing the alternative regi-mens over the preferred regimen for a period, astrategy that anecdotally has been associatedwith a reduction in cases of antibiotic associateddiarrhoea. With regard to the latter, a fluoroqui-nolone is suggested although the evidence for alower risk of C diYcile enteropathy remainslimited214 [Ia], 219 [IVb], 220 [Ib], 221[II].

8.4 Antibiotic resistance of respiratorypathogensResistance among respiratory pathogens isincreasing and is of concern. Beta-lactamaseproduction among H influenzae varies geo-graphically but ranges from 2% to 17% in vari-ous parts of the UK 222 [II], 223 [II]. However,this is an uncommon cause of pneumonia and,unless local data suggest otherwise, there isinsuYcient justification to include aâ-lactamase resistant antibiotic regimen ininitial empirical treatment of non-severe CAP.M catarrhalis is an even rarer cause of CAP forwhich the same argument applies.

S aureus is widely resistant to penicillin224 [II]and an increasing number are now methicillinresistant (MRSA). When this occurs in thecommunity it generally reflects hospitalisationwithin the recent past or residence in a nursinghome. Hence, â-lactamase unstable penicillins(penicillin G, aminopenicillins) and, in the caseof MRSA, isoxazolyl penicillins (flucloxacillin,cloxacillin) and cephalosporins are inappropri-ate for such infections.

Antibiotic resistance of S pneumoniae is ofgreater concern because of the dominance ofthis organism as a cause of CAP and becausepenicillin and macrolide resistance are fre-quently linked223 [II], 225 [II]. Resistanceamong pneumococci is the result of alterationsin one or more of the penicillin bindingproteins which reduces their aYnity for penicil-lin. This, in turn, leads to a requirement forhigher drug concentrations to bring aboutdeath of the organism. Of the >90 knownpneumococcal serotypes, a small number have


www.thoraxjnl.com



been responsible for penicillin resistanceworldwide, among which selected clones (suchas 23F, 9V, and 6B) have become widelydisseminated. However, despite these con-cerns, the clinical importance of in vitropenicillin resistance among S pneumoniaeremains uncertain when treating pneumo-coccal pneumonia226 [II], 227 [II], 228 [IVb], 229

[IVa], 230 [II], 231 [II]. This is reflected in thecontinued ability of current doses of penicillinsto inhibit strains of intermediate susceptibility(minimum inhibitory concentration (MIC) ofpenicillin 0.1–1.0 mg/l), as well as many strainsexhibiting higher level resistance (defined by anMIC of >1 mg/l). Resistance to erythromycinis the result of genetic mutations that eitheraVect the target site (erm gene mutations) orresult in elimination of the drug by an eZuxpump (mef gene mutation). The distribution ofsuch strains diVers internationally and prob-ably explains the variation in the clinical impactof such resistance since erm gene mutations arelinked to high level resistance. Tetracyclines arenot widely used in the treatment of CAP andresistance among S pneumoniae is relatively low.Likewise, reduced susceptibility of S pneumo-niae to fluoroquinolones is beginning to bereported 232 [II]. The trends in penicillin, eryth-romycin, and tetracycline resistance in Englandand Wales are shown in fig 9.

Legionella pneumophila and Legionella speciesin general remain susceptible to rifampicin,macrolides, and the fluoroquinolones, al-though low level resistance has been found invitro in some isolates. However, the clinicalsignificance of these observations remainsunclear.233

Linked to the issue of antibiotic resistance isthe problem of use and abuse of antibiotic pre-scribing. Excessive or inappropriate use ofantibiotics has been highlighted in the Houseof Lords Select Committee Report234 and bythe Department of Health.34 Proper patientselection for treatment and the correct use ofagents are emphasised in this document.

8.5 Newer antibioticsSince the first BTS guidelines were publishedthere have been a number of new antibioticslicensed for the treatment of CAP. Theseinclude new macrolides (clarithromycin andazithromycin) and several fluoroquinoloneswith greater in vitro activity against respiratorypathogens and, in particular, S pneumoniae,regardless of their susceptibility to penicillin.Their role in the initial empirical and specificantibiotic management of CAP is discussedbelow. Much has also been learned about anti-biotic distribution within the lung, based onstudies that detect penetration within bronchialtissue, alveolar macrophages, and epitheliallining fluid obtained at bronchoscopy 235 [Ia],236 [Ia]. Furthermore, in the development ofthese new agents there is substantial evidencethat links antimicrobial performance in vivo tothe pharmacokinetic profile of an agent and thein vitro susceptibility of target pathogens. Suchpharmacodynamic approaches are likely toinfluence dosage regimens increasingly in thefuture237 [Ia], 238 [II], 239 [II], 240 [Ia].

8.6 International diVerences inrecommendations and clinical studies ofmanagementIn defining the UK choice of empirical andspecific treatment for CAP, it is apparent thatthe international diVerences in publishedrecommendations cannot be entirely based ongeographical variation in the distribution andantibiotic susceptibility of pathogens responsi-ble. There is clearly variation in medical prac-tice with regard to licensing, availability,choice, dose, route of administration, andduration of treatment which is more a reflec-tion of local custom and practice than robustscientific evidence. The literature review for theperiod 1981–99 provided only 16 acceptablearticles relevant to the antibiotic managementof CAP54 [Ib], 221 [II], 241 [Ib], 242 [Ib], 243 [Ib],244 [Ib], 245 [Ib], 246 [Ib], 247 [Ib], 248 [Ib], 249 [Ib],250[Ib], 251 [Ib], 252 [Ib], 253 [Ib], 254 [Ib]. Theremainder were rejected for the followingreasons: inadequately powered studies or aretrospective design,15 255–272 antibiotic notavailable in the UK or withdrawn,27 273–306 studypopulation or management unrepresentative ofnormal clinical practice in the UK,253 307–315 orthey included mixed lower respiratory tractinfections including CAP.253 286 316 317

Few of the studies reviewed were conductedwithin a healthcare system comparable to thatof the UK. Others were designed to support thelicensing of new therapies. For this reason theyare primarily designed to demonstrate equiva-lence between the new agent and comparatortherapy, which may or may not have beenselected in accordance with current standardmanagement. This invariably makes it diYcultto oVer evidence based recommendations sincesuperiority of a particular regimen is rarelyidentified. Likewise, matters of diVerentialsafety for the various regimens is diYcult toassess since this information is essentially abyproduct of these licensing studies, is rarelystandardised, and has often not been comparedwith current standard treatment.

Figure 9 Resistance (%) to penicillin (high and intermediate), erythromycin andtetracycline among selected S pneumoniae isolates (blood and cerebrospinal fluid) fromlaboratories reporting to the Public Health Laboratory Service.

1989 1990 1991 1992 1993Year

1994 1995 1996 1997 1998

16

14

12

10

8

6

4

2

0

Res

ista

nce

(%

)

PenicillinErythromycinTetracycline


www.thoraxjnl.com



8.7 Formulation of theserecommendationsFor these reasons the recommendations fortreatment have been made on the basis ofassessing a matrix of laboratory, clinical, phar-macokinetic and safety data, interpreted in aninformed manner. While this remains anunsatisfactory basis for making robust evidencebased recommendations, it highlights the needfor appropriate, prospective, randomised con-trolled studies designed to address the manykey questions that will enable the managementof CAP to be placed on a sounder basis. Theresponsibility for this presents a challenge tomedical practitioners, healthcare systems,grant giving bodies, and industry. We have alsoonly considered antibiotics licensed and avail-able in the UK at the time these guidelineswere prepared.

As stated by Finch and Woodhead318 [IVa],“it is important to recognise that these are sim-ply guidelines and reflect our interpretation ofgood practice within an evolving area. Guide-lines cannot capture every clinical situation andit therefore remains the responsibility of thephysician to balance the history and clinicalfeatures, assess the importance of risk factorsand interpret local epidemiology and labora-tory data in order to make the best judgementfor an individual patient”.

(A) EMPIRICAL TREATMENT

8.8 What are the principles and practiceof empirical antibiotic choice for CAP inthe community?Most patients with pneumonia are treated suc-cessfully in the community in the absence ofany microbial definition of an infecting micro-organism(s). The decision to manage a patientin the community is based on a range of factorswhich include an assessment that the pneumo-nia is non-severe, that oral treatment is appro-priate and will be complied with, and that thesocial circumstances and available care for anindividual are satisfactory.

Empirical therapy is primarily directed at Spneumoniae which remains the leading cause ofCAP (section 3)179 [Ia]. Apart from M pneumo-niae, atypical pathogens, Legionella species andâ-lactamase producing bacteria are uncommonin the community setting. M pneumoniaeexhibits epidemic periodicity every 4–5 yearsand largely aVects younger persons. A policyfor initial empirical therapy that aimed alwaysto cover this pathogen was considered inappro-priate. The possibility of recommending a tet-racycline was considered on the basis of lowerresistance rates among pneumococci (fig 9)and activity against atypical pathogens. How-ever, this would require a major change in cur-rent prescribing practice by general practition-ers in the community where the majority ofpneumonias are treated, and hence would limitcompliance with the recommendations. Inaddition, there are safety concerns in women ofchildbearing potential.

For these reasons, as well as the issues ofcurrent practice, cost, wide experience and

drug tolerance, amoxicillin remains the pre-ferred agent. The alternative agent for thoseintolerant of amoxicillin is erythromycin54 [Ib],primarily for its activity against S pneumoniaealthough recognising that it is also activeagainst M pneumoniae, other atypical patho-gens, and Legionella species. Gastrointestinalintolerance to erythromycin is the majorindication for substituting an alternative mac-rolide such as clarithromycin or modifiedformulations of erythromycin. Concern overthe rising frequency of in vitro resistance of Spneumoniae to macrolides (which is oftenlinked to penicillin resistance) is recognised (fig9), yet published clinical evidence for clinicalfailure of macrolides in the treatment of pneu-mococcal pneumonia remains limited 319 [III],320 and controversial321 [IVa], 322 [IVa].

The newer macrolides323 [IVa] and fluoro-quinolones324 [IVa] have microbiologicalstrengths in vitro, yet in published studies todate have not been shown to be moreeYcacious than standard therapy in treatingpatients with CAP. They are clearly more costlyand, in the case of most fluoroquinolones,experience currently remains limited. Con-cerns also remain with regard to the safety ofthe new fluoroquinolones, with several com-pounds having been withdrawn because seri-ous adverse events have become apparent aftermarketing.324

The association of H influenzae and, to amuch lesser extent M catarrhalis, with acuteexacerbations of COPD is recognised325 [II].However, both remain uncommon causes ofCAP. When CAP does arise with thesepathogens, an even smaller percentage of suchpatients will be infected with â-lactamase pro-ducing strains. To illustrate the clinical signifi-cance of such resistance for managing CAP, itis estimated that 5% of cases of CAP may becaused by H influenzae, of which 15% may beâ-lactamase producing strains in the UK.Thus, of 500 patients with CAP, only four maybe infected with such antibiotic resistantstrains.

A view that specific pathogens are associatedwith other comorbid diseases (for example, Hinfluenzae and COPD) to increase the risk ofCAP is not supported by the literature. Forthese reasons these guidelines do not oVeralternative regimens for patients with orwithout comorbid illness, while recognisingthat such diseases can aVect the severity ofCAP in an individual.

The current concerns over the increasingprevalence of pneumococci with reducedsusceptibility to penicillin is recognised. How-ever, the incidence of highly resistant strains(MIC >4 mg/l) remains uncommon in theUK. Furthermore, the rarity of documentedclinical failures among penicillin resistantpneumococcal pneumonia, if treated withadequate doses of penicillin, is the basis forendorsing oral amoxicillin as first line therapy,but at an increased dosage of 500–1000 mgthree times daily.


www.thoraxjnl.com



Recommendations (table 8)+ Amoxicillin remains the preferred

agent but at a higher dose than previ-ously recommended [D].

+ A macrolide (erythromycin or clari-thromycin) is oVered as an alternativechoice and for those patients who arehypersensitive to penicillins [D].

+ Those with features of severe infectionshould be admitted urgently to hospital[C].

8.9 Should general practitionersadminister antibiotics before transfer tohospital in patients who need admission?There is no direct evidence upon which to pro-vide clear guidance on this question. There is,however, some circumstantial evidence to sug-gest that early antibiotics are of benefit insevere pneumonia.

Delay in prescribing antibiotics for patientsin hospital with diagnosed pneumonia is asso-ciated with a worse outcome326 [III] and, inpatients dying from CAP, the majority have notreceived prior antibiotics even though most

had visited a general practitioner in theprevious few days. In a national confidentialenquiry into CAP deaths in young adults inEngland and Wales, 20 of the 27 fatal casesinvestigated had seen their general practitionerfor the illness and only nine had received anti-biotics327 [II]. In the multicentre BTS study ofCAP in 1982, none of the patients who diedfrom pneumococcal pneumonia had receivedan antibiotic before admission. The authorsconcluded that some deaths may have beenpreventable and recommended that an anti-biotic active against S pneumoniae should bestarted as soon as pneumonia is recognised14

[Ib]. In a study of CAP from New Zealand sig-nificantly fewer (p=0.05) of those who diedhad received antibiotics before admission(20%) than those who survived (42%)60 [Ib].Currently, less than half of adults admitted tohospital in the UK with severe CAP havealready received antibiotics from their generalpractitioner22 [III], 66 [III]. Many deaths andrequirements for assisted ventilation occur inthe first few days of admission for severe CAP14

[Ib], 69 [Ib], 71 [Ib]. All of these studies providefurther support to the suggestion that, in casesof diagnosed pneumonia, antibiotics should begiven as early as possible, if necessary beforehospital admission.

Delays do occur between general prac-titioner assessment in the community, arrang-ing admission, confirmation of the diagnosis inhospital, and the start of treatment. Thesedelays are probably inevitable and will be exac-erbated by transport distances and ambulanceavailability and prioritisation, bed availability,and triage in the medical assessment unit oraccident and emergency department. Delaysbetween admission and receiving antibiotics ofover 6 hours have been reported for youngeradults dying in hospital of CAP (mean delay260 minutes20 [III]). This study was conductedbefore acute assessment units were introducedinto most UK hospitals, a change which mayhave speeded up the management of medicaladmissions. Although it is likely that mostpatients with low risk CAP which proves to benon-fatal come to no harm in the time that thiswhole process takes, it appears far from ideal todelay giving antibiotics when the general prac-titioner is reasonably sure of the diagnosis.

From time to time general practitioners dosee patients who are severely ill with whatappears to be pneumonia. In such circum-stances treatment should commence as soon aspossible, providing it does not delay transfer tohospital. When general practitioners feel treat-ment in such circumstances is needed, itshould aim to cover pneumococcal pneumo-nia, the commonest cause of severe CAP, withparenteral penicillin G or oral amoxycillin 1 gorally (or erythromycin in cases of penicillinsensitivity). General practitioners are likely tocarry such antibiotics with them as parenteralpenicillin is recommended as the immediatetreatment for suspected meningococcal infec-tion. Ambulance services should allocate a highpriority for transfer to hospital of patients withpneumonia.

Table 8 Preferred and alternative initial empirical treatment regimens for adults withCAP managed in the community

Preferred Alternative*

Home treated, not severe+ Amoxicillin 500 mg–1.0 g tds po + Erythromycin 500 mg qds po or

clarithromycin 500 mg bd† po

bd = twice daily; tds = 3 times/day; qds = 4 times/day; po = oral.*An alternative regimen is provided for those intolerant of or hypersensitive to the preferred regi-men.†Clarithromycin may be substituted for those with gastrointestinal intolerance to oral erythromy-cin and also has the benefit of twice daily dosage.

Table 9 Preferred and alternative initial empirical treatment regimens for adults withCAP seen and managed in hospital

Preferred Alternative*

Hospital treated, not severe (admitted fornon-clinical reasons or previously untreatedin the community)

+ Amoxicillin 500 mg–1.0 g tds po + Erythromycin 500 mg qds po orclarithromycin 500 mg bd† po

Hospital treated, not severeOral:+ Amoxicillin 500 mg–1.0 g tds po plus

erythromycin 500 mg qds po orclarithromycin 500 mg bd po

+ Fluoroquinolone with some enhancedpneumococcal activity, e.g. levofloxacin 500mg od po‡ (the only such licensed agent inthe UK at the time of writing)

If intravenous treatment needed:+ Ampicillin 500 mg qds iv or benzylpenicillin

1.2 g qds iv plus erythromycin 500 mg qdsiv or clarithromycin 500 mg bd iv

+ Levofloxacin 500 mg od iv‡

Hospital treated, severe+ Co-amoxiclav 1.2 g tds or cefuroxime 1.5 g

tds or cefotaxime 1 g tds or ceftriaxone 2 god (all iv) plus erythromycin 500 mg qds ivor clarithromycin 500 mg bd iv (with orwithout rifampicin 600 mg od or bd iv§)

+ Fluoroquinolone with some enhancedpneumococcal activity, e.g. levofloxacin 500mg bd iv, po‡ plus benzylpenicillin 1.2 g qdsiv

od = once daily; bd = twice daily; tds = 3 times/day; qds = 4 times/day; iv = intravenous; po = oral.*An alternative regimen is provided for those intolerant of or hypersensitive to the preferred regi-men, or where there are local concerns over C diYcile associated diarrhoea related to â-lactam use.†Clarithromycin may be substituted for those with gastrointestinal intolerance to oral erythromy-cin and also has the benefit of twice daily dosage.‡Levofloxacin is the only currently UK licensed fluoroquinolone with some enhanced activityagainst S pneumoniae and comes in an oral and parenteral formulation. In the future other fluoro-quinolones such as moxifloxacin, gemafloxacin, and gatifloxacin are likely to extend this choice,when licensed in the UK. Oral moxifloxacin is already licensed and is available in some othercountries.§Concurrent administration of rifampicin reduces the serum level of macrolides; the clinical sig-nificance of this is not known.Recommendations for switching from a parenteral drug to the equivalent oral preparation areshown in box 3.


www.thoraxjnl.com



Prescribing antibiotics does have an influ-ence on some microbiological investigations14

[Ib]. However, when general practitioners feela patient is severely ill or circumstances suggestthat delays in transfer will slow assessment andtreatment in hospital, concern over the poten-tial eVect on subsequent investigations is not areason to withhold treatment.

It is important to state that, as with the wholeof this document, these comments refer topatients with CAP and not to the much largergroup of patients with non-pneumonic lowerrespiratory tract infection or exacerbations ofCOPD.

Summary+ Delays occur in the process of admitting

patients to hospital with CAP and in admin-istration of antibiotics [III].

+ There is direct and indirect evidence thatadministering antibiotics early is importantin the outcome of CAP, particularly when itis assessed as severe [Ib].

+ Less than half of patients admitted withsevere CAP have received antibiotics beforeadmission, even though they may have seentheir general practitioner [III].

+ Most deaths from CAP occur shortly afteradmission [Ib].

+ Antibiotics given before admission cannegatively influence the results of subse-quent microbiological investigations [Ib],but this is not seen as a reason for withhold-ing antibiotics if the general practitionerfeels they are indicated.

Recommendations+ For those patients referred to hospital

with suspected CAP, general practition-ers may consider administering antibi-otics immediately where the illness isconsidered to be life threatening orwhere there are likely to be delays (>2hours) in admission [D].

8.10 What are the principles and practiceof empirical antibiotic choice for adultshospitalised with non-severe CAP?Approximately 20% of patients with CAP areadmitted to hospital in the UK (section 2). Thereasons for hospitalisation vary and includeseverity of the infection, an unsatisfactoryresponse to treatment initiated by the generalpractitioner, significant co-morbid illness, andnon-clinical reasons such as inappropriatehome circumstances suitable for communitymanagement.

The principles of antibiotic selection fornon-severe CAP managed in hospital are simi-lar to those for its management in the commu-nity. The predominant pathogen will be Spneumoniae. However, overall, atypical patho-gens and Legionella species account for ap-proximately 20% of defined infections. Forthese reasons a combined â-lactam/macrolideregimen is recommended (table 9). Oralamoxicillin and erythromycin or clarithromy-cin is the preferred regimen. When oral therapyis inappropriate, parenteral ampicillin or peni-cillin G are oVered as alternatives to oral

amoxicillin. Erythromycin by the intravenousroute is given four times daily; thus clarithro-mycin, given twice a day, is oVered as the pre-ferred macrolide for parenteral therapy and analternative for oral administration.

The new fluoroquinolones may be given asan alternative in hospitalised patients inspecific circumstances. While these fluoroqui-nolones possess in vitro activity against S pneu-moniae, including strains with reduced suscep-tibility to penicillin, experience to date intreating such infections remains limited. Therole of the new fluoroquinolones is addressedlater in section 8.18. At the time of completingthese guidelines only levofloxacin is licensedand available in the UK.

Regardless of the regimen selected, it is criti-cal that the antibiotics are administeredpromptly (within 2 hours of admission) and, inthe case of the patient with severe pneumonia,without delay by the admitting doctor in theadmissions ward or by the general practitionerif delays are expected in the hospital admissionprocess (see section 8.9). Delays in administra-tion of antibiotics are related adversely to mor-tality326 [Ib].

In practice, it is recognised that a significantnumber of patients with non-severe pneumoniaare admitted to hospital for non-clinicalreasons—for example, advanced age, personalor family preference, inadequate home care, oradverse social circumstances—who mightotherwise be adequately managed in thecommunity. Others will be admitted who havenot received antibiotic treatment. They cannotbe considered to have failed community treat-ment and initial treatment with a single agentas for the “home treated, not severe” group(table 8) is considered appropriate. Further-more, M pneumoniae is an important contribu-tor to the overall incidence of atypical patho-gens but is an infrequent cause of CAP inelderly patients. This provides further justifica-tion for monotherapy in the hospitalisednon-severe elderly patient. In all such circum-stances patient management requires carefulclinical judgement and regular reviews.

Switch from parenteral drug to theequivalent oral preparation+ This should be made as soon as clinically

appropriate, in the absence of microbio-logically confirmed infection.

+ Change is straightforward when there isan eVective and equivalent oral andparenteral formulation.

+ In the case of the parenteral cepha-losporins, the oral switch to co-amoxiclav625 mg tds is recommended rather thanto oral cephalosporins.

+ For those treated with benzylpenicillin +levofloxacin, oral levofloxacin with orwithout oral amoxicillin 500 mg–1.0 g tdsis recommended.

Box 3 Recommendations for switching from aparenteral drug to the equivalent oral preparation.


www.thoraxjnl.com



Following initial assessment and empiricaltherapy, progress should be monitored care-fully. The route and choice of antibiotictreatment will require adjustment, either bystepping up and broadening the spectrum ofmicrobiological activity in the light of clinicaldeterioration or as a result of positive microbio-logical information, or stepping down withimprovement as discussed below. The review ofantibiotic treatment forms an obvious andessential part of the regular clinical review ofpatients with CAP.

Recommendations (table 9)+ Most patients can be adequately treated

with oral antibiotics [C].+ Combined oral treatment with amoxi-

cillin and a macrolide (erythromycin orclarithromycin) is preferred for pa-tients who require hospital admissionfor clinical reasons [D].

+ Oral monotherapy should be consid-ered in the following circumstances:+ Amoxicillin monotherapy may be

considered for (i) those previouslyuntreated in the community or (ii)those admitted to hospital for non-clinical reasons who would otherwisebe treated in the community [D]. Thelatter factor will often apply to theelderly in whom it is also recognisedthat infection with atypical pathogensrequiring macrolide therapy is un-common (see section 3).

+ Monotherapy with a macrolide maybe suitable for patients who havefailed to respond to an adequate courseof amoxicillin prior to admission.Deciding on the adequacy of priortherapy is diYcult and is a matter ofindividual clinical judgement. It istherefore recommended that combi-nation antibiotic therapy is the pre-ferred choice in this situation andthat the decision to adopt mono-therapy is reviewed on the “posttake” round within the first 24 hoursof admission (see section 9.1) [D].

+ When oral treatment is contra-indicated, recommended parenteralchoices include intravenous ampicillinor benzylpenicillin together with eryth-romycin or clarithromycin [D].

+ A fluoroquinolone active against Spneumoniae is an alternative regimenfor those intolerant of penicillins ormacrolides or where there are localconcerns over C diYcile associateddiarrhoea. However, experience withsuch newer fluoroquinolones in thetreatment of CAP and their interactionand side eVect profile is at present lim-ited and further reported experience isrequired [B–]. Levofloxacin is currentlythe only recommended agent licensedin the UK.

8.11 What are the principles and practiceof empirical antibiotic choice for adultshospitalised with severe CAP?Mortality is greatly increased in those withsevere pneumonia (see section 6). The illnessmay progress before microbiological infor-mation is available.

Preferred and alternative initial treatmentregimens are summarised in table 9 and mostlyinclude combination therapy with broad spec-trum â-lactams and a macrolide. While S pneu-moniae remains the predominant pathogen, Saureus and Gram negative enteric bacilli,although uncommon, carry a high mortality179

[Ia], hence the recommendation for broadspectrum â-lactam regimens in those withsevere CAP [C]. Parenteral antibiotics are rec-ommended in those with severe CAP, regard-less of the patient’s ability or otherwise to takeoral medication [C]. This is to ensure prompthigh blood and lung concentrations of anti-biotic.

Patients admitted to hospital with CAPcaused by Legionella species are more likely tohave severe pneumonia179 [Ia], so the initialempirical antibiotic regimen should also cap-ture this pathogen within its spectrum of activ-ity.328 EYcacy data from prospective controlledclinical trials are not available. However, a ret-rospective study suggests a reduction inmortality for those treated with a third genera-tion cephalosporin plus a macrolide54 [III],although no additional benefit has been notedin another study329 [II]. Currently, a macrolideor a fluoroquinolone is preferred for treatinglegionella infections. The choice of a macrolideis based on a combination of in vitro pharma-cokinetic and animal model data supplementedwith limited clinical information330 [IVb] [C].Azithromycin is currently only available fororal administration and is therefore inappropri-ate for treating severe infection. For life threat-ening infection where Legionella species couldbe present, the addition of rifampicin is recom-mended despite the absence of clinical datasupporting its benefit [C]. The risk of loweringmacrolide plasma and tissue concentrations asa result of induction of the P 450 enzyme sys-tem is acknowledged,331 332 but there are no datato indicate that this is of clinical significancewhen treating CAP.

Fluoroquinolones are oVered as an alterna-tive, despite limited data on their use in severeCAP324 [IVa]. Levofloxacin is the only licensedagent available in the UK at the time of writing,and is marketed in parenteral and oral formu-lations. Since the latter is 98% bioavailable, thisindicates that it can be used in severe pneumo-nia provided there are no contraindicationsto oral administration. While it has modestactivity against pneumococci in vitro, thepublished evidence for eYcacy in severe CAPis reassuring199 [Ib], 247 [Ib], 270 [Ib], 333 [Ib],334 [Ib]. However, until more clinical experi-ence is available, we recommend combining itwith another agent active against S pneumoniaesuch as parenteral benzylpenicillin in cases ofsevere CAP. Future Gram positive fluoroqui-nolones such as gatifloxacin, gemafloxacin andmoxifloxacin, which are currently unlicensed in


www.thoraxjnl.com



the UK, may be found through clinical experi-ence to have greater activity against S pneumo-niae than levofloxacin and hence may be eVec-tive as monotherapy.

Recommendations+ Patients with severe pneumonia should

be treated immediately after diagnosiswith parenteral antibiotics [B–].

+ An intravenous combination of a broadspectrum â-lactamase stable antibioticsuch co-amoxiclav or a second genera-tion (e.g. cefuroxime) or third genera-tion (e.g. cefotaxime or ceftriaxone)cephalosporin together with a mac-rolide (e.g. clarithromycin or erythro-mycin) is preferred [C].

+ For those who are intolerant ofâ-lactam or macrolide therapy or wherethere are local concerns over C diYcileassociated diarrhoea, a fluoroquinolonewith enhanced activity against pneumo-cocci together with benzylpenicillin isoVered as an alternative [D]. Levo-floxacin is currently the only such fluo-roquinolone licensed in the UK.

8.12 When should the intravenous or theoral route be chosen?Parenteral administration of antibiotics iswidely and often unnecessarily used in manag-ing hospitalised patients including those withCAP245 [Ib], 335 [IVa]. Approximately 30–50%of patients admitted to hospital will initiallyrequire treatment with parenteral antibiotics245

[Ib]. Apart from the discomfort to the patientof inserting intravenous devices, there aresignificant complications, notably infection. Inaddition, the total cost of parenteral regimensgreatly exceeds orally administered treatment.

Factors determining the route of administra-tion are summarised in box 4. Parenteral anti-biotics are clearly indicated for patients unableto swallow where there is concern aboutadequate absorption of drug from the gut, andin the presence of severe pneumonia. However,many antibiotics are well absorbed followingoral administration and achieve their maxi-mum plasma concentration within 1–2 hours.

Parenteral treatment+ severe pneumonia+ impaired consciousness+ loss of swallowing reflex+ functional or anatomical reasons for mal-

absorption

Oral treatment+ community managed+ hospital managed, non-severe with no

other contraindications

Box 4 Indications for parenteral and oralantibiotic treatment of adult CAP.

Recommendation+ The oral route is recommended in those

with non-severe pneumonia admittedto hospital provided there are no con-traindications to oral treatment [B+].

8.13 When should the intravenous routebe changed to oral?As stated above, parenteral antibiotic treatmentis widely and often unnecessarily used inhospitalised patients with non-severe pneumo-nia. This in part reflects custom and practicebut, in addition, may be driven by too liberal aninterpretation of the “criteria” for identifyingsevere CAP for which parenteral agents arerecommended. The current practice of medi-cine surrounding emergency medical admis-sions may also be a factor in the choice ofparenteral administration where it providesgreater confidence to admitting junior medicalstaV that the patient is receiving the “best”management. Oral treatment was clearly morewidely adopted in the past.335 Publishedevidence indicating comparable eYcacy ofparenteral and oral regimens is limited but hasbeen shown for intravenous cefuroxime andoral levofloxacin247 [Ib].

The choice and timing of any change tooral treatment will be aVected by severalfactors. These include the absence of anycontraindications to oral administration, theavailability of any microbiological infor-mation regarding aetiology of the infection,and clear evidence that the patient is respond-ing to initial treatment. Some of the criteriaindicating improvement are summarised inbox 5.

+ Resolution of fever for >24 hours+ Pulse rate <100 beats/min+ Resolution of tachypnoea+ Clinically hydrated and taking oral fluids+ Resolution of hypotension+ Absence of hypoxia+ Improving white cell count+ Non-bacteraemic infection+ No microbiological evidence of legionella,

staphylococcal or Gram negative entericbacilli infection

+ No concerns over gastrointestinal ab-sorption

Box 5 Features indicating response to initialempirical parenteral treatment permitting consid-eration of substitution with oral antibiotics.

There can be no rigid recommendationconcerning the timing of transfer to oral treat-ment and further studies are needed305 [II],336 [Ia], 337 [II]. Any decision must be individu-alised on the basis of assessing all factors.Nonetheless, the recommended guideline isthat oral treatment be considered in a patientwho has shown clear evidence of improvementand whose temperature has resolved for aperiod of 24 hours. The features indicatingresponse to parenteral treatment are summa-rised in box 5. This policy will allow a signifi-cant proportion of patients with non-severepneumonia to be safely transferred to anoral regimen after a period of initial paren-teral treatment198 [II], 248 [Ib], 266 [Ia], 305 [II],336 [Ia], 338 [IVa].


www.thoraxjnl.com



Recommendations+ Patients treated initially with

parenteral antibiotics should be trans-ferred to an oral regimen as soon asclinical improvement occurs and thetemperature has been normal for 24hours, providing there is no contraindi-cation to the oral route. Pointers toclinical improvement are given in table9 [B+].

+ The choice of route of administrationshould be reviewed initially on the “posttake” round and then daily [D].

+ Ward pharmacists could play an impor-tant role in facilitating this review byhighlighting prescription charts whereparenteral antibiotic treatment contin-ues [D].

8.14 Which oral antibiotics arerecommended on completion ofintravenous treatment?The selection of agents for oral administrationfollowing initial intravenous treatment is basedon antimicrobial spectrum, eYcacy, safety, andcost considerations (summarised in box 3).Although it may appear logical to select the oralformulation of a parenteral agent, this is notessential and such oral agents may not meet thecriteria for selection. For macrolides, oral clari-thromycin is better tolerated than oral erythro-mycin241 [Ib], 242 [Ib] but is more expensive. Aclinical judgement can be made whether tochange to oral monotherapy in those who haveresponded favourably to parenteral combina-tion therapy or where there is microbiologicaldocumentation of the nature of the infection, inwhich case the recommendations in table 10should be adopted.

Recommendation+ The choice of antibiotics when switch-

ing from the intravenous to the oralroute is summarised in box 3 [C].

8.15 For how long should antibiotics begiven?The precise duration of antibiotic treatment forthe management of microbiologically docu-mented and non-documented CAP is not sup-ported by robust evidence. The Summary ofProduct Characteristics (formerly the DrugData Sheets) for many agents used in the treat-ment of CAP mention a range of treatmentdurations which sometimes diVer inter-nationally.

In the case of documented infections there isevidence that pneumonia caused by Legionellaspecies may require 14 days of treatment or upto 21 days in severe infections330 [IVa].Likewise, intracellular pathogens responsiblefor pneumonia sometimes respond slowly andhence a 2 week treatment regimen has beenproposed for atypical pathogens.

The aim of antibiotic treatment is to ensureelimination of the target pathogen in the short-est time. In uncomplicated infections this islikely to occur rapidly (within 3 days) withmany common respiratory pathogens such as Spneumoniae. The resolution of pneumoniainvolves not only the elimination of theinvading pathogen and its products, but alsothe subsidence of the host inflammatoryresponse which together are responsible for themany clinical and radiographic features ofpneumonia.

Until we have more precise methods to iden-tify microbiological and clinical end pointsreliably, the duration of treatment will remainsubject to clinical judgement and will vary withthe individual patient, disease severity, andspeed of resolution. Table 10 oVers guidancebased largely on custom and practice butmodified where evidence exists.

Recommendations (table 10)+ For patients managed in the commu-

nity and most of those admitted to hos-pital with non-severe and uncompli-cated pneumonia, treatment withappropriate antibiotics for 7 days isrecommended [C].

+ For patients with severe microbiologi-cally undefined pneumonia, 10 days oftreatment is proposed. This should beextended to 14–21 days where legionella,staphylococcal, or Gram negative en-teric bacilli pneumonia are suspectedor confirmed [C].

8.16 Failure of initial empirical treatmentIn those patients who fail to respond to initialempirical treatment, several possibilities needto be considered. The first is whether the cor-rect diagnosis been made. Radiographic reviewis recommended for both community and hos-pital managed patients. This may also indicatecomplications of CAP such as pleural eVusion/empyema, lung abscess, or worsening pneu-monic shadowing. This aspect is considered indetail in section 9.

The initial empirical antibiotic regimen mayneed to be reassessed. However, compliancewith and adequate absorption of an oralregimen should first be considered.

Microbiological data should be reviewed andfurther specimens examined with a view toexcluding less common pathogens such as Saureus, atypical pathogens, Legionella species,viruses, and Mycobacteria species. It should alsobe noted that mixed infections can arise inapproximately 10% of patients admitted tohospital with CAP. In the absence of anymicrobiological indicators of infection, themanagement of those failing initial empiricaltreatment will vary according to the severity of

Table 10 Duration of antibiotic management by place of care, severity, andmicrobiological data

Place/severity or pathogen Duration of treatment (days)

Home treated, not severe (microbiologically undefined) 7Hospital treated, not severe (microbiologically undefined) 7Hospital treated, severe (microbiologically undefined) 10Legionella infection 14–21“Atypical” pathogen 14Pneumococcal infection (uncomplicated) 7Staphylococcal infection 14–21Gram negative enteric bacilli 14–21


www.thoraxjnl.com



the illness at reassessment. In patients man-aged in the community with non-severe pneu-monia a macrolide could be substituted foramoxicillin. However, when the patient’s con-dition has deteriorated, admission to hospitalshould be considered.

In the hospital managed, non-severely illpatient the addition of a macrolide is recom-mended in those patients initially managedwith amoxicillin alone. Changing to a newfluoroquinolone such as levofloxacin provides asecond alternative.

In the severely ill patient already receiving aâ-lactam/clarithromycin regimen it is recom-mended that rifampicin should be added. Inaddition, urgent referral to a respiratory physi-cian should be made for clinical assessmentincluding the possible need for bronchoscopicsampling.

Recommendations+ When a change in empirical antibiotic

treatment is considered necessary, amacrolide could be substituted for oradded to the treatment for those withnon-severe pneumonia treated withamoxicillin monotherapy in the com-munity or in hospital [C].

+ For those with non-severe pneumoniain hospital on combination therapy,changing to a fluoroquinolone witheVective pneumococcal cover is anoption [C].

+ The addition of rifampicin may be con-sidered for those with severe pneumo-nia not responding to combinationantibiotic treatment [C].

(B) SPECIFIC PATHOGEN DIRECTEDANTIBIOTIC TREATMENT

8.17 What are the optimum antibioticchoices when specific pathogens havebeen identified?In routine clinical practice only about one thirdto one quarter of patients with CAP admitted

to hospital will be defined microbiologically. Ofthese some, such as mycoplasma, chlamydial,and Q fever infection will be diagnosed late inthe illness on the basis of seroconversion,reducing the opportunity for early targetedtreatment. In patients managed in the commu-nity, very few will be microbiologically defined.

When a pathogen has been identified,specific treatment as summarised in table 11 isproposed. In transferring patients from empiri-cal to pathogen targeted treatment, the regi-men and route of administration will be deter-mined by the continued need for parenteraltreatment and known drug intolerance. Hence,table 11 provides preferred and alternativeregimens for intravenous or oral administra-tion. However, it should be remembered thatapproximately 10% (see section 3) of infectionswill be of mixed aetiology, although many ofsuch co-pathogens will be viral and hence notinfluenced by antibiotic choice. These recom-mendations are again based on a synthesis ofinformation which includes the in vitro activityof the drugs, appropriate pharmacokinetics,and clinical evidence of eYcacy gleaned from avariety of studies. The choice of agent may bemodified following the availability of sensitivitytesting or following consultation with a special-ist in microbiology, infectious disease, or respi-ratory medicine.

S pneumoniae highly resistant to penicillin(MIC >4 mg/l) is currently uncommon in theUK. However, it is important that the situationis monitored and in future higher doses ofpenicillins or alternative regimens may need tobe considered.

S aureus is an uncommon cause of CAP inthe UK. Most community isolates are sensitiveto methicillin, although the recent increase inMRSA in hospitalised patients may result insubsequent readmission with an MRSA infec-tion which may include CAP. Options formethicillin sensitive and resistant infections are

Table 11 Recommended treatment of microbiologically documented pneumonia* (local specialist advice should also be sought)

Pathogen Preferred Alternative

S pneumoniae Amoxicillin 500 mg–1.0 g† tds po orbenzylpenicillin 1.2 g qds iv

Erythromycin 500 mg qds po orclarithromycin 500 mg bd po orcefuroxine 0.75–1.5 g tds iv orcefotaxime 1–2 g tds iv or ceftriaxone 2 g od iv

M pneumoniae andC pneumoniae

Erythromycin 500 mg qds po or iv orclarithromycin 500 mg bd po or iv

Tetracycline 250–500 mg qds po orfluoroquinolone‡ po or iv

C psittaci andC burnetii

Tetracycline 250–500 mg qds po or 500 mg bd iv Erythromycin 500 mg qds orclarithromycin 500 mg bd, both po or iv

Legionella spp Clarithromycin 500 mg bd po or iv ± rifampicin§ 600 mg od or bd po/iv Fluoroquinolone po or iv‡

H influenzae Non-â-lactamase-producing: amoxicillin 500 mg tds po or ampicillin 500 mg qds ivâ-lactamase-producing: co-amoxiclav 625 mg tds po or 1.2 g tds iv

Cefuroxime 750 mg–1.5 g tds iv orcefotaxime 1–2 g tds iv or ceftriaxone 2 g od iv orfluoroquinolone‡ po or iv

Gram negative entericbacilli

Cefuroxime 1.5 g tds orcefotaxime 1–2 g tds iv orceftriaxone 1–2 g bd iv

Fluoroquinolone‡ iv orimipenem 500 mg qds iv ormeropenem 0.5–1.0 g tds iv

P aeruginosa Ceftazidime 2 g tds iv plus gentamicin or tobramycin (dose monitoring) Ciprofloxacin 400 mg bd iv orpiperacillin 4 g tds iv plus gentamicin or tobramycin(dose monitoring)

S aureus Non-MRSA: flucloxacillin 1–2 g qds iv ± rifampicin 600 mg od or bd po or ivMRSA: vancomycin 1 g bd iv (dose monitoring)

Teicoplanin 400 mg bd iv ± rifampicin 600 mg od orbd po/iv

*Can be modified once the results of sensitivity testing are available.†A higher dose of 1.0 g tds is recommended for infections documented to be caused by less susceptible strains (MIC >1.0 mg/l).‡Currently UK licensed and available suitable fluoroquinolones include ciprofloxacin, ofloxacin and levofloxacin.§Concurrent administration of rifampicin reduces the serum level of macrolides; the clinical relevance of this is not known.


www.thoraxjnl.com



based on parenteral administration in view ofthe serious nature of staphylococcal pneumo-nia.

Recommendation+ If a specific pathogen has been identi-

fied, the antibiotic recommendationsare summarised in table 11. Localmicrobiological advice is recom-mended [C].

8.18 What is the role of the newerantibiotics such as fluoroquinolones, newmacrolides, and oxazolidinonederivatives?Several new antibiotics have become availablesince the 1993 BTS guidelines were published.Among these are the fluoroquinolones (withimproved activity against S pneumoniae com-pared with ciprofloxacin and ofloxacin), themacrolide clarithromycin, and the azalideazithromycin.

The largest class of agents are the fluoroqui-nolones. These include levofloxacin which iscurrently licensed in the UK and gatifloxacin,gemifloxacin, and moxifloxacin which are notlicensed in the UK at the time of writing306 [Ib].CAP is likely to be included among theirlicensed indications. Many of these new agentsare active against respiratory “atypical” patho-gens and Legionella species.328 Their activityagainst S pneumoniae may also give them amore important role in the future should peni-cillin and erythromycin resistance continue toincrease and be associated with demonstrableclinical failure. However, caution should re-main since pneumococcal resistance to thefluoroquinolones, including levofloxacin, isalready recognised and has been associatedwith clinical failures, although varying qualita-tively232 and geographically.

Resistance to fluoroquinolones is mediatedby alterations in the target sites, topoisomerase2 and 4, which is a consequence of mutations inthe gyrA and parC genes, respectively. Whetherfuture respiratory quinolones will be lesssubject to such mutational resistance will haveto await their availability. By limiting recom-mendations for the empirical use of fluoroqui-nolones to alternative agents in selectedpatients managed in hospital, it is hoped thatthe potential risks of quinolone resistanceamong pneumococci can be minimised. Such astrategy of restricting use to prevent the emer-gence of resistance is also recommended byother recent guidelines.7

In considering their role in the managementof CAP, it is important to note the limitedclinical experience to date. There are relativelyfew studies that have recruited populations ofpatients representative of UK community andhospital treated patients with CAP.

Fluoroquinolones have also been associatedwith unexpected side eVects. Three fluoro-quinolones (sparfloxacin, trovafloxacin, andgrepafloxacin) have had their licensing ar-rangements altered or withdrawn after launch

because of the appearance of side eVects.Phototoxicity or interactions with theophyllinehave complicated the use of some of theseagents, while prolongation of the QTc intervaland dysrrhythmias have aVected others. Hepa-totoxicity and features of hypersensitivity havecomplicated the use of trovafloxacin.

For these reasons, as well as their likelygreater acquisition costs, the new fluoroqui-nolones are not recommended as first lineagents or for community use for pneumoniaand should be restricted in their use to specificsituations. They may provide a useful alterna-tive in selected hospitalised patients with CAPas indicated in table 9, or when there are con-cerns about the risk of C diYcile enteropathywith broad spectrum â-lactams.

Clarithromycin, which is available for oraland parenteral administration, has many fea-tures that are an improvement on erythromy-cin249 [Ib]. It is more active against Hinfluenzae, is administered twice daily, and isbetter absorbed with fewer gastrointestinal sideeVects. The relatively greater acquisition cost,particularly when administered orally, means itis oVered as an alternative rather than apreferred choice to erythromycin.

Azithromycin is licensed for the treatment ofCAP including pneumococcal pneumonia244

[Ib], 339 [Ib]. It has good activity againstLegionella species in vitro, but there is currentlyno parenteral formulation in the UK. Despitethe high tissue and lung concentrations, bloodlevels are very low and published experience intreating bacteraemic infections remains limitedwith occasional failures being reported.320 Forthese reasons, as well as cost, this drug has notbeen included in the recommendations.

Linezolid is the first member of a new classof antibiotic (the oxazolidinones) and waslicensed in the UK around the end of 2000. Itsmajor indications are for the treatment ofresistant staphylococcal (notably MRSA) andenterococcal infections. It is also active againstpenicillin susceptible and resistant pneumo-cocci. Although licensed for the treatment ofCAP, its use is limited to hospitalised patients.Its restricted spectrum of activity would alsorequire combined treatment. Although avail-able for intravenous and oral use, experienceremains limited and hence it is not recom-mended for empirical treatment. It may haveoccasional use in the management of docu-mented MRSA infections in those intolerant orunresponsive to other regimens.

Recommendations+ New fluoroquinolones are not recom-

mended as first line agents or for com-munity use for pneumonia but mayprovide a useful alternative in selectedhospitalised patients with CAP, as indi-cated in table 9 [C].

+ Oral and parenteral clarithromycin isoVered as an alternative to erythromy-cin [B+].


www.thoraxjnl.com



9 Complications and failure to improve

9.1 What factors and action should beconsidered in patients who fail toimprove in hospital?When a patient in hospital with CAP fails toimprove with initial management, they shouldbe carefully reviewed. Failure to improveshould lead to consideration of the variouspossibilities summarised in box 6. Elderlypatients often respond less rapidly.

In this situation there should be a carefulreview by an experienced clinician of the clini-cal history, examination, prescription chart,and results of initial available investigationresults. Further investigations, including arepeat chest radiograph, CRP and white cell

count, and further specimens for microbiologi-cal testing should be considered in the light ofany new information after the clinical review.Consideration should be given to referral to aphysician with an interest in respiratory medi-cine.

Recommendations+ For patients who fail to improve as

expected, there should be a carefulreview by an experienced clinician ofthe clinical history, examination, pre-scription chart, and results of all avail-able investigation results [D].

(A) Incorrect diagnosis or complicating conditionCOMMON REASONS

+ Pulmonary embolism/infarction+ Pulmonary oedema+ Bronchial carcinoma+ Bronchiectasis+ Slow response in the elderly patient

UNCOMMON REASONS

+ Pulmonary eosinophilia/eosinophilic pneumonia+ Cryptogenic organising pneumonia+ Pulmonary alveolar haemorrhage+ Foreign body+ Congenital pulmonary abnormality e.g. lobar sequestration

(B) Unexpected pathogen or pathogens not covered by antibiotic choice+ Pathogens always resistant to common antibiotics (e.g. an “atypical” pathogen not

responding to penicillin)+ Pathogens sometimes resistant to commonly used antibiotics (e.g. ampicillin resistant

H influenzae, penicillin resistant S pneumoniae, mycobacteria)

(C) Antibiotic ineVective or causing allergic reaction+ Poor absorption of oral antibiotic+ Inadequate dose+ Antibiotic hypersensitivity+ Patient not receiving or taking prescribed antibiotic

(D) Impaired local or systemic defences+ Local (e.g. bronchiectasis, endobronchial obstruction, aspiration)+ Systemic immune deficiency (e.g. HIV infection, hypogammaglobulinaemia, myeloma)

(E) Local or distant complications of CAPPULMONARY

+ Parapneumonic eVusion+ Empyema+ Lung abscess+ Adult respiratory distress syndrome (ARDS)

EXTRAPULMONARY

+ Phlebitis at intravenous cannula site+ Metastatic infection+ Septicaemia+ End organ sequelae of septicaemia (e.g. renal failure)

(F) Overwhelming infection

(G) Improvement expected too soon+ In the elderly, for example

Box 6 Reasons for failure to improve as expected.


www.thoraxjnl.com



+ Further investigations including a re-peat chest radiograph, CRP and whitecell count, and further specimens formicrobiological testing should be con-sidered in the light of any new infor-mation after the clinical review [D].

9.2 What are the common complicationsof CAP?A brief description of the common complica-tions of CAP is given below. Complicationsassociated with specific infections are summa-rised in table 12.

PLEURAL EFFUSION AND EMPYEMA

Parapneumonic eVusions develop in 36–57%of patients with bacterial pneumonia admittedto hospital and can be the cause of persistingpyrexia despite adequate antibiotic treatment340

[II]. The presence of bilateral pleural eVusionsin CAP is associated with increased mortality195

[II]. Although most eVusions will resolve withantibiotic treatment alone, it is recommendedthat thoracocentesis is performed promptly inpatients admitted to hospital with parapneu-monic eVusion. Those patients found to havean empyema (defined as the detection ofcloudy fluid, pus or organisms on Gram’s stainor culture341 [II]) or a complicated parapneu-monic eVusion (defined as clear pleural fluidwith a pH of <7.2342 [II]) should then haveearly and eVective pleural space drainage.Pleural fluid for measurement of pH should becollected anaerobically in a heparinised bloodgas syringe and measurement should beperformed in a blood gas analyser. Whileempyema has become relatively uncommon,delayed thoracocentesis and chest tube drain-age leads to longer and more costly hospitalisa-tion340 [III]. Further details of the managementof empyema will be available in the BTS guide-lines on the management of patients with pleu-ral disease which are in preparation.

Recommendations+ Early thoracocentesis is indicated for

all patients with parapneumonic eVu-sion [D].

+ Those found to have an empyema orclear pleural fluid with pH <7.2 shouldhave early and eVective pleural fluiddrainage [C].

+ The BTS guidelines on the managementof empyema should be followed [D].

LUNG ABSCESS

Lung abscess is a rare complication of CAP,being seen most commonly in the debilitated oralcoholic patient and following aspiration.Infection with anaerobic bacteria, S aureus,Gram negative enteric bacilli, or S milleri (inthe presence of poor dental hygiene) should beconsidered. Although most patients respond toappropriate antibiotics, early surgical drainagevia pneumonotomy may occasionally beneeded. Patients with lung abscess may requirea prolonged course of antibiotics.

Recommendations+ Less usual respiratory pathogens in-

cluding anaerobes, S aureus, Gramnegative enteric bacilli, and S millerishould be considered in the presence oflung abscess [D].

+ Prolonged antibiotic treatment and oc-casionally surgical drainage should beconsidered [D].

METASTATIC INFECTION

Patients with septicaemia associated withpneumonia can occasionally develop meta-static infection. Meningitis, peritonitis, endo-carditis, and septic arthritis have all beenreported. Purulent pericarditis can occur, usu-ally in direct relation to an empyema.

Most such complications can be detected bycareful history and examination.

Table 12 Some complications associated with specificinfections

Pathogen Complications

S pneumoniae SepticaemiaPyopneumothoraxPericarditis/endocarditisMeningitis/brain abscessPeritonitisArthritisHerpes labialis

M pneumoniae MeningoencephalitisAseptic meningitisGuillain-Barré syndromeTransverse myelitisCerebellar ataxiaAscending polyneuropathyPericarditisMyocarditisDiarrhoeaHaemolytic anaemiaSkin rashesPolyarthropathyHepatitisPancreatitisSplenomegalyAcute glomerulonephritisHaemorrhagic myringitis

Legionella spp ConfusionEncephalomyelitisGuillain-Barré syndromeCerebellar signsPericarditisHyponatraemiaRenal failureRhabdomyolysis and myositisDiarrhoeaPolyarthropathyJaundice/abnormal liver functionPancreatitisThrombocytopenia

C burnetii Optic neuritisHepatitisHaemolytic anaemiaOsteomyelitisEndocarditis with chronic infection

S aureus Pneumatoceles and/or pneumothorax(especially in children)

SepticaemiaLung abscessMetastatic infection


www.thoraxjnl.com



10 Prevention and vaccination strategies

10.1 IntroductionAlthough considerable progress has been madeover recent years in antimicrobial treatmentand supportive care for patients with CAP, theissue of prevention remains important espe-cially for those at “high risk”. Vaccines againstinfluenza A and against S pneumoniae are avail-able currently.

10.2 Influenza virus and vaccinationInfluenza is an acute viral infection of therespiratory tract aVecting all age groups. It isusually self-limiting with recovery in 2–7 days,but can be complicated by bronchitis, otitismedia in children, and secondary bacterialpneumonia notably due to S aureus. Primaryinfluenza pneumonia occurs but is rare andcarries a high mortality rate. The greatestmortality from influenza is in patients withunderlying disease such as severe chronicrespiratory, cardiac, or renal disease or diabetesmellitus343 344 [Ib], particularly in those agedover 65 years343 345 [III]. In patients over 65 yearsdeath from influenza occurs predominantly dueto secondary bacterial pneumonia and cardiacfailure and not primary infection346 [Ib].

INFLUENZA VIRUS

Influenza viruses are single stranded, seg-mented RNA viruses belonging to the familyOrthomyxoviridae. Two forms of influenzavirus are responsible for most clinical illness—influenza A and influenza B. Influenza C onlygives rise to an acute pharyngitis. Outbreaks ofinfluenza A occur most years and can causeepidemics (defined as >300 cases per 100 000patients at risk). Influenza B can also causeoutbreaks but these tend to be less extensiveand to be associated with less severe illness.Pandemics result from the emergence of newviruses in which there are major changes in thesurface proteins, mainly haemagglutinin (H)and neuraminidase (N). Influenza A viruses areantigenically labile and minor changes in thesesurface proteins (“antigenic drift”) occurprogressively from season to season. Majorchanges (“antigenic shift”) due to acquisitionof a “new” haemagglutinin results in the emer-gence of modified viruses against which thepopulation has little or no immunity. InfluenzaB viruses are also prone to antigenic drift butwith less frequent changes. The World HealthOrganisation monitors influenza virusesthroughout the world and makes recommenda-tions each year about the strains to be includedin vaccines for the forthcoming winter.

INFLUENZA VACCINE

Influenza vaccine is prepared each year usingviruses similar to those considered most likelyto be circulating in the forthcoming winter.The viruses are grown in the allantoic cavity ofchick embryos and are therefore contra-indicated in individuals with egg allergy. Theyare then chemically inactivated, treated, and

purified. Current vaccines are trivalent con-taining two type A and one type B subtypeviruses. Two types of vaccine are available—those prepared by disrupting whole virusesusing organic solvents for detergents (“splitvirus vaccines”) and those which contain puri-fied H and N surface antigens (“surfaceantigen vaccines”). Both vaccines are given as asingle 0.5 ml intramuscular injection and areequivalent in eYcacy and adverse reactions.Vaccines should be stored at 2–8°C andprotected from the light. They must not be fro-zen. They should be allowed to reach roomtemperature and shaken well before they aregiven.

In the UK immunisation is currently recom-mended for those of all ages with:+ chronic respiratory disease including asthma;+ chronic heart disease;+ chronic renal disease;+ immunosuppression due to disease or treat-

ment;+ diabetes mellitus;+ all those aged 65 years or older;+ those in long stay residential care.

ADVERSE EFFECTS

Uptake of influenza immunisation by those forwhom it is recommended is low in many coun-tries347 348 [III]. Patients are often concernedabout side eVects and both doctors andpatients may have doubts about the protectiveeYcacy of the vaccine. Reported side eVectsinclude fever, myalgia, and local and systemicallergic reactions but, in a randomised doubleblind controlled study of 1806 patients aged 60years or older, only discomfort at the injectionsite was more common in the vaccinatedgroup349 [Ib].

EVIDENCE OF EFFICACY

Evidence of the eYcacy of influenza vaccine inhumans has derived from three types of clinicalstudies: experimental studies in which volun-teers are challenged by live viruses understrictly controlled circumstances; field studieswhich register morbidity and mortality duringnaturally occurring influenza outbreaks; andimmune response studies which measure anti-body responses as a surrogate marker forprotection against influenza infection. Experi-mental studies are not applicable in popula-tions at risk of serious complications frominfluenza infection.

Several large well designed case-control fieldstudies from North America have focused onthe eYcacy of influenza vaccine in reducingadmissions for pneumonia and influenza inthose over 65 years of age and in the reductionof mortality346 350–352 [Ia]. They show that influ-enza vaccination reduces hospital deaths frompneumonia and influenza by about 65% andfrom all respiratory causes by 45%. Vaccinationis also associated with fewer hospital admis-sions for pneumonia and influenza and fewer


www.thoraxjnl.com



outpatient visits for all respiratory conditions inthose aged 65 years and over with chronic lungdisease353 [III].

In a further UK case-control study of 315deaths during the 1989–90 influenza epidemicand 770 matched controls, vaccination re-duced mortality by 41%344 [Ia]. In subjectswho received the vaccine for the first time in1989 vaccination reduced mortality by 9%,while in those who had been previouslyvaccinated mortality was reduced by 75%.Vaccination was also found to be equally eVec-tive in reducing mortality in nursing homes asin the community. Such studies may tend tounderestimate the protective eVect of influenzavaccination as some illnesses considered to beinfluenza are likely to be caused by other coin-cidental pathogens.

Possibly the most impressive evidence of theeYcacy of vaccination has come from theNetherlands where a randomised double blindcontrolled trial of influenza vaccine in subjectsover 65 years of age reduced the incidence ofserologically proven influenza by 50%354 [Ib].In a meta-analysis of 20 cohort studies ofpatients aged over 65 the pooled estimates ofthe eYcacy of vaccination were 56% forpreventing respiratory illness, 53% for prevent-ing pneumonia, 50% for preventing admissionto hospital, and 68% for preventing death355

[Ia].Immune response studies support these

findings and show that the protection rate frominfluenza vaccine judged by serological testingis over 75% for influenza A and 51–97% forinfluenza B356 [Ib]. A recent meta-analysis ofseven field studies (including 13 trials) and 12immune response studies (including 53 trials)confirmed that protection does not decreasewith annual repeated influenza vaccination357

[Ia].

EFFICACY IN LOW RISK GROUPS

While studies in “high risk” groups includingthose over 65 years of age have shown benefitfrom influenza vaccination, it remains unclearwhether healthy working adults and, particu-larly, those in the medical and nursingprofessions should be oVered immunisation. Ina randomised, double blind, placebo controlledtrial in Los Angeles, 179 hospital employeesshowed no clear reduction in influenza-like ill-ness, severity of illness or sick absenteeism358

[Ib]. In contrast, a study of 849 subjects inMinneapolis randomly assigned influenza vac-cine or placebo showed 25% fewer episodes ofupper respiratory illness, 43% fewer days ofsick leave, and 44% fewer visits to physicians’oYces for upper respiratory tract illness inthose receiving active vaccine346 [IIa]. Simi-larly, influenza vaccination prevented infectionby influenza A and B in 264 healthy hospitalbased healthcare professionals of mean age 28years and may reduce cumulative days of illnessand absence359 [Ib]. Although there is currentlyinsuYcient evidence on which to base a clearrecommendation about the routine immunisa-tion of healthcare workers, the Department ofHealth in the UK now advises National Health

Service employers to oVer influenza vaccina-tion to all staV.

FUTURE DEVELOPMENT

Intranasal vaccination appears to be highlyeVective and has been shown to reduce theincidence of laboratory confirmed influenza Ain nursing home residents when comparedwith inactivated vaccine alone.360 New forms ofdelivery system are also being developedincluding an oral vaccine using biodegradablemicrospheres designed to be taken up in thesmall bowel and reformulations of influenzavaccine with antigen delivery systems whichproduce significantly greater increases in anti-body response.

Recommendations (based on theDepartments of Health guidelines)+ Influenza vaccination is recommended

for those at “high risk” of mortalityfrom influenza or pneumonia [C].

+ These “high risk” groups include thosewith chronic lung, heart, renal and liverdisease, diabetes mellitus, immunosup-pression due to disease or treatment,and those aged over 65 years[C].

+ Influenza vaccine is contraindicated forthose with hypersensitivity to hens’ eggs[C].

10.3 Pneumococcal vaccinationS pneumoniae, the most common aetiologicalcause of CAP, is an encapsulated Gram positivecoccus. The capsule is composed of one of 90serologically distinct polysaccharides and viru-lence of the organism appears to be determinedby the nature of this capsule. Immunity topneumococci depends largely on production oftype specific anticapsular antibodies. Geo-graphical, temporal, and age diVerences in thedistribution of the 90 diVerent serotypes andthe ability of S pneumoniae to transfer capsulargenes from one strain to another361 362 all haveimplications for vaccination strategy. Thecurrent 23-valent vaccine includes serotypesthat cause 88% of the bacteraemic infections inthe USA and 96% of those in the UK.363

PNEUMOCOCCAL VACCINE

Pneumococcal vaccine is a polyvalent vaccinecontaining 25 mg purified capsular polysac-charide from each of 23 capsular types of Spneumoniae. It is supplied in single dose vialsand should be stored at 2–8°C. A single dose of0.5 ml is given subcutaneously or, preferably,intramuscularly into the deltoid muscle or lat-eral aspect of the mid thigh. Mild soreness andinduration at the site of injection is commonand a low grade fever occasionally occurs.Re-immunisation is not normally advised andis contraindicated within 3 years because of therisk of severe reactions. Such reactions appearto relate to high levels of circulating antibodies.Pneumococcal vaccine should not be givenduring acute infection and is not recom-mended during pregnancy or lactation. How-ever, in those aged over 65 years it can be givensafely at the same time as influenza vaccine at adiVerent site364 [Ib].


www.thoraxjnl.com



In the UK pneumococcal vaccine is recom-mended by the Department of Health for allthose aged 2 years or older in whom pneumo-coccal infection is likely to be more common ormore dangerous. Such groups include thosewith:+ asplenia;+ severe dysfunction of the spleen including

sickle cell disease and coeliac disease;+ chronic renal disease or nephrotic syn-

drome;+ chronic heart disease;+ chronic lung disease;+ chronic liver disease including cirrhosis;+ diabetes mellitus;+ immunodeficiency or immunosuppression

due to disease or treatment including HIVinfection.

ANTIBODY RESPONSE

Pneumococcal vaccination given to middleaged and elderly patients (age range 50–85years) at follow up 8 weeks after treatment inhospital for pneumonia has been shown to pro-duce an antibody response without severeadverse reactions365 [Ib]. A case-control studyof adults aged 33–85 years of age showed thatantibody levels appeared to wane 6 years afterimmunisation366 [IIa]. However, the level ofpneumococcal antibody required for protec-tion is not currently known. The newer CRM197

conjugated pneumococcal oligosaccharide vac-cines do not appear to oVer any advantagesover polysaccharide vaccines in those aged over60 years367 [IIb].

CLINICAL EFFICACY OF PNEUMOCOCCAL VACCINE

The results of numerous eYcacy studies havefound it diYcult to reach firm recommenda-tions. A meta-analysis of nine randomised con-trolled trials with 12 vaccine treated andcontrol study groups comprising 40 431 indi-viduals has shown that pneumococcal vaccina-tion oVers protection of around 66% againstdefinitive pneumococcal pneumonia with

bacteraemia for normal or “low risk” adults368

[Ia]. Surprisingly, this eVect could not beshown for the more heterogeneous group of“high risk” patients who are those for whomvaccination is recommended. It seems likelythat some of the trials included lackedsuYcient size to show protection in those agedover 65 years and other “high risk” groups.

Furthermore, the vaccine could not beshown to protect against pneumonia (allcauses) or mortality due to pneumococcalpneumonia in those aged 60–70 years368 [Ia].However, a recent 2 year retrospective cohortstudy of 1898 subjects aged over 65 years withchronic lung disease reported that pneumococ-cal vaccination was associated with fewerhospital admissions for pneumonia, fewerdeaths, and direct medical care cost savings369

[III].At the present time it seems reasonable to

conclude that, at least in immunocompetentpatients, the current 23-valent vaccine may beclinically eVective for preventing pneumococ-cal pneumonia with associated bacteraemia.EYcacy in immunocompromised patients orother “high risk” groups is much less certain.

Recommendations (based on theDepartments of Health guidelines)+ While pneumococcal vaccination is rec-

ommended by the Departments ofHealth for all those aged 2 years or olderin whom pneumococcal infection islikely to be more common or serious,there is no evidence that it is eVective insuch “at risk” groups [A+].

+ Pneumococcal vaccine should not begiven during acute infection and is notrecommended during pregnancy. Re-immunisation within 3 years is contra-indicated [C].

+ Pneumococcal and influenza vaccinescan be given together at diVerent sites[A–].


www.thoraxjnl.com



11 Acknowledgements and declaration ofinterest

11.1 AcknowledgementsMany people have helped with the preparationof these guidelines and our thanks go to them.In particular we thank Jenny Etches, Secretaryto the Committee; Sue Allen for her veryeYcient administrative help; Rosamund Mac-farlane for editorial help; Dr Hisham Ziglamfor coordinating the reference database; DrMartin Muers and Dr Bernard Higgins, Chair-men of the British Thoracic Society Standardsof Care Committee and Mrs Sheila Edwards,Chief Executive of the British Thoracic Societyfor support and advice; Dr John Winter, DrMike Pearson, Professor Ian Gilmore, Dr Mar-tin Connolly, Dr Mary Armitage, Dr DavidBlack, Dr Louise Restrick, Professor StephenSpiro, Dr Mark Britton, Professor RichardWise, Dr Martin Wood, Dr Cliodna McNultyand Dr Robert George for their comments,help in facilitating the process of peer reviewand endorsement by other professional groups.

11.2 Declaration of interests ofcommittee membersThe committee members fulfilled the require-ments of the British Thoracic Society regardingpersonal declaration of interests. Declaration ofInterest forms were updated annually by com-mittee members and contents shared withinthe committee and with the Secretary of theBritish Thoracic Society.

A summary of declarations of interests forthe life time of the committee is given below:

TB has received research funding from EisaiLtd, lecture fees from Aventis and support forattending conferences from Wyeth; GD hasreceived research funding from SmithKlineBeecham and Astra, lecture fees from Allenand Hanburys and support for attendingconferences from Allen and Hanburys and 3M;RGF has received consultancy fees from GlaxoWellcome, Nexstar, Bristol Myers Squibb,SmithKline Beecham, AstraZeneca, Pharma-cia, Parke Davis and Pantherix, research fund-ing from GlaxoWellcome and Pharmacia, andsupport for attending conferences from GlaxoWellcome, Aventis, Wyeth, and Biomerieux;WFH has received consultancy fees fromGlaxoWellcome, Schering Plough, BoehringerIngleheim, Hoechst Marion Roussel, Astra,3M, Zeneca and Rhone Poulenc Rorer,research funding from 3M and Rhone PoulencRorer, and support for attending conferencesor courses from GlaxoWellcome, ScheringPlough, Zeneca and 3M; DH has received con-sultancy fees from Bayer, GlaxoWellcome,SmithKline Beecham, Pfizer, Abbott and Bris-tol Myers Squibb, research funding fromHoechst Marion Roussel, SmithKline Bee-cham, Pharmacia/Upjohn, Grunenthal and

Abbott, lecturing fees from Key Med, HoechstMarion Roussel, GlaxoWellcome, Bayer andSchering Plough, and support for attendingconferences and meetings from Bayer, GlaxoWellcome, Schering Plough, SmithKline Bee-cham and Abbott; DH holds shares inSmithKline Beecham; WSL has received re-search funding from Hoechst Marion Rousseland Bayer and support for attending confer-ences from Bayer; JTM has received consul-tancy fees from Pfizer, Abbott, HoechstMarion Roussel, Trinity, GlaxoWellcome, re-search funding from Hoechst Marion Roussel,Rhone Poulenc Rorer and Bayer, lecture feesfrom AstraZeneca, Hoechst Marion Rousseland Pfizer, and support for attending confer-ences from Astra, Pfizer, Allen and Hanburysand 3M; RM—none declared; DN has receivedconsultancy fees from Pharmacia/Upjohn andBayer and research funding from HoechstMarion Roussel and Pharmacia/Upjohn; PS—none declared; MAW has received consultancyfees from Bayer, Glaxo, and Pfizer, lecture feesfrom Abbott, Pasteur Merieux, Bayer, Wyethand Glaxo, and support to attend conferencesfrom Schering Plough, Pasteur Merieux, Glaxoand Boehringer Ingelheim; JW—none de-clared.

11.3 AYliations and addresses ofcommittee membersPneumonia Guidelines Committee of the Brit-ish Thoracic Society Standards of Care Com-mittee: Dr J Macfarlane (Chairman andEditor), Consultant Physician, NottinghamCity Hospital; Dr T Boswell, ConsultantMicrobiologist, Nottingham City Hospital; DrG Douglas, Consultant Physician in Respira-tory Medicine and Infection, Aberdeen RoyalInfirmary; Professor R Finch, Professor ofInfectious Diseases, University of Nottingham,Nottingham City Hospital; Dr W Holmes,General Practitioner, Sherrington Park Medi-cal Practice, Nottingham; Dr D Honeybourne,Consultant Physician, Heartlands Hospital,Birmingham; Dr W S Lim, Specialist Registrarin General and Respiratory Medicine, Notting-ham City Hospital; Mr R Marriott, SeniorMedical Librarian, Nottingham City Hospital;Dr D Nathwani, Consultant Physician inInfectious Diseases, Kings Cross Hospital,Dundee; Dr P Saul, General Practitioner,Beech Avenue Health Centre, Wrexham; Dr MWoodhead, Consultant Physician, ManchesterRoyal Infirmary; Dr J Wyatt, Director ofKnowledge Management Centre, School ofPublic Policy, University College, London.


www.thoraxjnl.com



Appendix 1 Check list used by reviewers forappraising studies

Study: .............................................................. Reviewer: ...........................................

Please complete section 1 first. If study OK, complete one of sections 2a–d, as appropriate.1. General: is the study relevant to our question ?h Were the patients studied similar (in age, sex, disease severity...) to target patients?h Were the outcome measures of interest to us and our patients ?h Was the clinical setting (primary care, intensive care...) similar to our setting ?h Was the study carried out in a healthcare system similar to ours ?h Is the study design recognisable and appropriate, with clear methods described ?h Is the study recent enough to take account of any important advances ?h If negative, was this study large enough to provide useful information ?2a. Studies of cause and eVect (randomised trial of treatment)h Was assignment of patients to treatment truly randomised ?h Was the planned treatment concealed from those recruiting patients before enrolment ?h Were all patients who entered the study accounted for ?h Were patients analysed in the groups to which they were initially randomised ?h Were patients and doctors blind to the treatment given ?h Were groups treated the same way, apart from the treatment ?h Were the groups similar at the start of the trial ?2b. Studies of aetiology (case-control study of a harmful agent)h Were there two groups of cases, similar except for exposure to a harmful agent ?h Was occurrence of the outcome measured in the same way for both groups ?h Were enough patients followed up for long enough for the outcome to develop ?h Did exposure clearly precede the outcome ?h Was there a dose-response gradient ?h Was there a re-challenge, or improvement after the drug stopped?h Does the association make biological sense ?2c. Studies of diagnosis (evaluation of clinical findings or tests)h Was the finding or result compared with a 24 carat gold standard for diagnosis ?h Was the finding or result determined blind to the gold standard ?h Was the gold standard determined blind to the finding or test result ?h Was the gold standard determined in all cases, not just those with an abnormal result ?2d. Studies of prognosis, prognostic index (cohort studies)h Was a defined sample of patients assembled at an early stage of the disease ?h Were patients followed up long enough for the outcome to develop ?h Was the outcome clearly defined, objective and assessed blind to exposure in all cases?h Was the performance of any prognostic index tested on a fresh set of cases ?3. Comments:


www.thoraxjnl.com



Appendix 2 Additional checklist used forappraising studies to inform pneumonia aetiologyAbsolute requirements:+ Is this an original report ?+ Were patients with community acquired pneumonia separately identified?+ Was the study designed to assess community acquired pneumonia aetiology?+ Was the patient sample representative (e.g. suYcient numbers, consecutive cases, exclusions

clearly defined)?+ Was the study of suYcient duration to exclude seasonal bias?

Data qualification:+ Is the geographical area clear and relevant?+ Is the patient age group defined?+ Are microbial investigations clearly defined?+ Was the investigation biased toward a specific pathogen?+ Is the setting community, hospital, intensive care or a combination?


www.thoraxjnl.com



Appendix 3 Types of study and levels ofevidence used to illuminate specific clinicalquestions

Type of clinical questionEvidencelevel Brief definition Types of study providing this level of evidence

Treatment: Ia A good recent systematic review Systematic review of randomised trialsIb A rigorous study designed to answer the

questionA rigorous randomised trial comparing Twith best alternative

II One or more prospective clinical studies A cohort study or faulty randomised trial+ Is treatment (T) eVective? III One or more retrospective clinical studies A case control study

IVa Formal expert consensus Delphi study of expert practiceIVb Other information Study of pharmacology of T

Aetiology or harm: Ia A good recent systematic review Systematic review of cohort studiesIb A rigorous study designed to answer the

questionA large, well designed cohort study

+ Does the aetiological agent(A) cause disease?

II One or more prospective clinical studies A faulty cohort study

+ Does a certain drug (D)cause a specific side eVect?

III One or more retrospective clinical studies A case control study

IVa Formal expert consensus Delphi study of expert opinionIVb Other information Study of pathophysiology of D

Diagnosis or prognosis: Ia A good recent systematic review Systematic review of blind comparisons of Twith gold standard

+ Is the investigation (T) anaccurate test for diagnosis ofthe disease (D)?

Ib A rigorous study designed to answer thequestion

Blind prospective comparison of T, F or Mwith gold standard for D or E (eg. responseto specific therapy) with multivariate analysis

+ Is finding (F) an accuratepredictor of event or outcome(E)?

II One or more prospective clinical studies Analysis of prospective test results in patientsenrolled in an RCT of therapy for varyingstages of D. Prospective validation study withunivariate analysis

+ Does severity measureaccurately predict event oroutcome (E)?

III One or more retrospective clinical studies Retrospective study of test results or findingsin a database of patients with univariate ormultivariate analysis

IVa Formal expert consensus Delphi study of expert opinion about TIVb Other information Study of pathophysiology of D

Public health, health policy:+ Is policy (P) cost eVective in

the National Health Service?

Ia Economic and policy analysis based ongood recent systematic reviews

Economic and policy analysis with modellingand sensitivity analysis using data from SRsof eVectiveness and SRs of cost studies in thesame routine clinical settings

Ib Economic and policy analysis based on arigorous study designed to answer thequestion

Economic and policy analysis with modellingand sensitivity analysis using data from anRCT of eVectiveness and a cost study in thesame routine clinical setting

II Economic and policy analysis based onone or more prospective clinical studies

Economic and policy analysis with modellingand sensitivity analysis using otherprospective data in various settings

III Economic and policy analysis based onone or more retrospective clinical studies

Economic and policy analysis with modellingand sensitivity analysis using retrospectivedata

IVa Formal expert consensus Delphi study of national expert opinionabout P

IVb Other information Local opinion about P


www.thoraxjnl.com



Appendix 4 Generic levels of evidence andguideline statement grades, appropriate across alltypes of clinical questions

Evidence level DefinitionExample of study providing this level of evidencefor a therapy question

Guidelinestatement grade

Ia A good recent systematic review of studiesdesigned to answer the question of interest

Cochrane systematic review of randomisedcontrolled trials studying the eVectiveness offlu vaccines

A+

Ib One or more rigorous studies designed toanswer the question, but not formallycombined

Randomised controlled trial of eVectivenessof a flu vaccine

A–

II* One or more prospective clinical studieswhich illuminate, but do not rigorouslyanswer, the question

Prospective cohort study comparingpneumonia rates in patients who are and arenot vaccinated against flu; non-randomisedcontrolled trial

B+

III† One or more retrospective clinical studieswhich illuminate but do not rigorouslyanswer the question

Audit or retrospective case control study,comparing flu vaccination history in patientswho did and did not present with pneumonia

B–

IVa‡ Formal combination of expert views Delphi study of UK expert recommendationsfor flu vaccination

C

IVb Other information Expert opinion, informal consensus; in vitroor in vivo studies on related topics

D

*Hard to diVerentiate AHCPR’s “well designed controlled study without randomisation” (level IIa) from “other type of welldesigned experimental study” (level IIb).†Major criterion is retrospective versus prospective data collection since non-experimental designs are better suited than even RCTsfor answering certain questions.‡Distinguish formal consensus from informal consensus methods according to HTA systematic review (1998).


www.thoraxjnl.com



References

1 British Thoracic Society. British Thoracic Society guidelinesfor the management of community-acquired pneumonia inadults admitted to hospital. Br J Hosp Med 1993;49:346–50.

2 Bartlett JG, Dowell SF, Mandell LA, et al. Practiceguidelines for the management of community-acquiredpneumonia in adults. Infectious Diseases Society ofAmerica. Clin Infect Dis 2000;31:347–82.

3 Niederman MS, Bass JB Jr, Campbell GD, et al. Guidelinesfor the initial management of adults with community-acquired pneumonia: diagnosis, assessment of severity, andinitial antimicrobial therapy. American Thoracic Society.Am Rev Respir Dis 1993;148:1418–26.

4 Mandell LA, Marrie TJ, Grossman RF, et al. Canadianguidelines for the initial management of community-acquired pneumonia: an evidence-based update by theCanadian Infectious Diseases Society and the CanadianThoracic Society. The Canadian Community-AcquiredPneumonia Working Group. Clin Infect Dis 2000;31:383–421.

5 Conference Report. Infections des voies respiratoires. Con-ference de consensus en therpeutique anti-infectieuse. MedMal Infect 1991;21(Suppl):1–8.

6 Gialdroni GG, Bianchi. Guidelines for the management ofcommunity acquired pneumonia in adults. Monaldi ArchChest Dis 1995;50:21–7.

7 HeVelfinger JD, Dowell SF, Jorgensen JH, et al. Manage-ment of community acquired pneumonia in the era ofpneumococcal resistance: a report from the drug resistantStreptococcus pneumoniae therapeutics working group.Arch Intern Med 2000;160:1399–408.

8 Huchon GWM. Management of adult community acquiredlower respiratory infections. Eur Respir Rev 1998;8:391–426.

9 American Thoracic Society. Guidelines for the managementof adults with community acquired pneumonia. Am JRespir Crit Care Med 2001;163:1730–54.

10 Macfarlane J. An overview of community acquired pneumo-nia with lessons learned from the British Thoracic SocietyStudy. Semin Respir Infect 1994;9:153–65.

11 Macfarlane JT, Finch RG, Ward MJ, et al. Hospital study ofadult community-acquired pneumonia. Lancet 1982;ii:255–8.

12 White RJ, Blainey AD, Harrison KJ, et al. Causes ofpneumonia presenting to a district general hospital. Thorax1981;36:566–70.

13 Venkatesan P, Gladman J, Macfarlane JT, et al. A hospitalstudy of community acquired pneumonia in the elderly.Thorax 1990;45:254–8.

14 British Thoracic Society and the Public Health LaboratoryService. Community-acquired pneumonia in adults in UKhospitals in 1982–1983: a survey of aetiology, mortality,prognostic factors and outcome. Q J Med 1987;62:195–220.

15 McNabb W, Shanson D, Williams D, et al. Adultcommunity-acquired pneumonia in central London. J RSoc Med 1984;77:550–5.

16 Impallomeni M, Galletly NP, Wort SJ, et al. Increased risk ofdiarrhoea caused by Clostridium diYcile in elderly patientsreceiving cefotaxime. BMJ 1995;311:1345–6.

17 Starr JM, Rogers TR, Impallomeni M. Hospital-acquiredClostridium diYcile diarrhoea and herd immunity. Lancet1997;349:426–8.

18 McNulty C, Logan M, Donald IP, et al. Successful controlof Clostridium diYcile infection in an elderly care unitthrough use of a restrictive antibiotic policy. J AntimicrobChemother 1997;40:707–11.

19 Gladman J, Barer D, Venkatesan P, et al. Outcome of pneu-monia in the elderly: a hospital survey. Clin Rehab 1991;5:201–5.

20 Tang CM, Macfarlane JT. Early management of youngeradults dying of community acquired pneumonia. RespirMed 1993;87:289–94.

21 Woodhead MA, Macfarlane JT, Rodgers FG, et al. Aetiologyand outcome of severe community-acquired pneumonia. JInfect 1985;10:204–10.

22 Hirani NA, Macfarlane JT. Impact of management guide-lines on the outcome of severe community acquired pneu-monia. Thorax 1997;52:17–21.

23 Kinnear W, Finch RG, Pilkington R, et al. Nosocomial lowerrespiratory tract infections on surgical wards. Thorax 1990;45:187–9.

24 Bohte R, van Furth R, van den Broek PJ. Aetiology ofcommunity-acquired pneumonia: a prospective studyamong adults requiring admission to hospital. Thorax1995;50:543–7.

25 Almirall J, Morato I, Riera F, et al. Incidence of community-acquired pneumonia and Chlamydia pneumoniaeinfection: a prospective multicentre study. Eur Respir J1993;6:14–8.

26 SteinhoV D, Lode H, Ruckdeschel G, et al. Chlamydiapneumoniae as a cause of community-acquired pneumoniain hospitalized patients in Berlin. Clin Infect Dis 1996;22:958–64.

27 Lode H, Garau J, Grassi C, et al. Treatment of community-acquired pneumonia: a randomized comparison of spar-floxacin, amoxycillin-clavulanic acid and erythromycin. EurRespir J 1995;8:1999–2007.

28 Lieberman D, SchlaeVer F, Boldur I, et al. Multiplepathogens in adult patients admitted with community-acquired pneumonia: a one year prospective study of 346consecutive patients. Thorax 1996;51:179–84.

29 Marrie TJ, Peeling RW, Fine MJ, et al. Ambulatory patientswith community-acquired pneumonia: the frequency ofatypical agents and clinical course. Am J Med 1996;101:508–15.

30 Macfarlane JT. Guidelines and the management of commu-nity acquired pneumonia. In: Horizons in medicine. No 8.London: Royal College of Physicians, 1997.

31 Macfarlane JT, Holmes WF, Macfarlane RM, et al. Contem-porary use of antibiotics in 1089 adults presenting withacute lower respiratory tract illness in general practice inthe UK: implications for developing management guide-lines. Respir Med 1997;91:427–34.

32 Dewar JC, Macfarlane JT. Are junior doctors confidentabout prescribing for adults with lower respiratory tract ill-ness? Thorax 2000;54: A21.

33 A Working Party Report of the Federation of MedicalRoyal Colleges. Acute medicine: the physician’s roleproposals for the future. London: Royal College of Physi-cians, 2000.

34 Standing Medical Advisory Committee. SubGroup onAntimicrobial Resistance. The path of least resistance.London: Department of Health, 1998.

35 A prescription for improvement. Towards more rational prescrib-ing in general practice. London: HMSO, 1994.

36 Torres A, Ausina V. Empirical treatment of nonsevere com-munity acquired pneumonia: still a diYcult issue. EurRespir J 1995;8:1996–8.

37 Hart CA. Antibiotic resistance: an increasing problem? Italways has been there, but there are things we can do. BMJ1998;316:1255–6.

38 Macfarlane J. Lower respiratory tract infection and pneu-monia in the community. Semin Respir Infect 1999;14:151–62.

39 Chronic Obstructive Pulmonary Disease Guidelines Groupof the Standards of Care Committee of the BritishThoracic Society. Guidelines for the management ofchronic obstructive pulmonary disease. Thorax 1997;52(Suppl 5): S1–28.

40 Woodhead MA, Macfarlane JT, McCracken JS, et al.Prospective study of the aetiology and outcome ofpneumonia in the community. Lancet 1987;i:671–4.

41 Melbye H, Straume B, Aasebo U, et al. The diagnosis ofadult pneumonia in general practice. The diagnostic valueof history, physical examination and some blood tests.Scand J Prim Health Care 1988;6:111–7.

42 Diehr P, Wood RW, Bushyhead J, et al. Prediction of pneu-monia in outpatients with acute cough: a statisticalapproach. J Chronic Dis 1984;37:215–25.

43 Marrie TJ, Durant H, Yates L. Community-acquired pneu-monia requiring hospitalization: 5-year prospective study.Rev Infect Dis 1989;11:586–99.

44 Eddy DM. A manual for assessing health practices and design-ing practice policies. Philadelphia: American College of Phy-sicians, 1992.

45 Eccles M, Clapp Z, Grimshaw J, et al. North of England evi-dence based guidelines development project: methods ofguideline development. BMJ 1996;312:760–2.

46 Caine M, Lim WS, Macfarlane JT, et al. Keeping patientsinformed about their illness: community acquired pneumo-nia. Thorax 2000;55(Suppl 3):A71.

47 Woodhead M, Macfarlane J. Local antibiotic guidelines foradult community-acquired pneumonia: a survey of UKhospital practice in 1999. J Antimicrob Chemother 2000;46:141–3.

48 Sackett DL, Wennberg JE. Choosing the best researchdesign for each question. BMJ 1997;315:1636.

49 Sackett DL. Bias in analytic research. J Chronic Dis1979;32:51–63.

50 Jaeschke R, Guyatt GH, Sackett DL. User’s guides to themedical literature III. How to use an article about adiagnostic test: are the results of the study valid? JAMA1994;271:389–91.

51 Scottish Intercollegiate Guidelines Network. Clinicalguidelines: criteria for appraisal. Edinburgh: Royal Collegeof Physicians of Edinburgh, 1995.

52 Sackett DL, Richardson WS, Rosenberg WM, et al.Evidence-based medicine. London: Churchill Livingstone,1997.

53 Guyatt GH, Sackett DL, Sinclair JC, et al. User’s guides tothe medical literature. IX. A method for grading health carerecommendations. JAMA 1995;274:1800–4.

54 Gleason PP, Kapoor WN, Stone RA, et al. Medicaloutcomes and antimicrobial costs with the use ofthe American Thoracic Society guidelines for outpatientswith community-acquired pneumonia. JAMA 1997;278:32–9.

55 Gilbert K, Gleason PP, Singer DE, et al. Variations in anti-microbial use and cost in more than 2,000 patients withcommunity-acquired pneumonia. Am J Med 1998;104:17–27.

56 Nathwani D, Williams F, Winter J, et al. Use of indicators toevaluate the quality of community acquired pneumoniamanagement. Clin Infect Dis 2001;32:728–41.


www.thoraxjnl.com



57 Jokinen C, Heiskanen L, Juvonen H, et al. Incidence ofcommunity-acquired pneumonia in the population of fourmunicipalities in eastern Finland. Am J Epidemiol 1993;137:977–88.

58 Foy HM, Cooney MK, Allan I, et al. Rates of pneumoniaduring influenza epidemics in Seattle, 1964 to 1975. JAMA1979;241:253–8.

59 Macfarlane JT, Colville A, Guion A, et al. Prospective studyof aetiology and outcome of adult lower respiratory tractinfections in the community. Lancet 1993;341:511–4.

60 Marrie TJ. Epidemiology of community-acquired pneumo-nia in the elderly. Semin Respir Infect 1990;5:260–8.

61 Marston BJ, PlouVe JF, File TM Jr, et al. Incidence ofcommunity-acquired pneumonia requiring hospitalization.Results of a population-based active surveillance study inOhio. The Community-Based Pneumonia Incidence StudyGroup. Arch Intern Med 1997;157:1709–18.

62 Lave JR, Fine MJ, Sankey SS, et al. Hospitalizedpneumonia. Outcomes, treatment patterns, and costs inurban and rural areas. J Gen Intern Med 1996;11:415–21.

63 Guest JF, Morris A. Community-acquired pneumonia: theannual cost to the National Health Service in the UnitedKingdom. Eur Respir J 1997;10:1530–4.

64 Foy HM, Cooney MK, McMahan R, et al. Viral and myco-plasmal pneumonia in a prepaid medical care group duringan eight-year period. Am J Epidemiol 1973;97:93–102.

65 Minogue MF, Coley CM, Fine MJ, et al. Patientshospitalized after initial outpatient treatment for commu-nity acquired pneumonia. Ann Emerg Med 1998;31:376–80.

66 British Thoracic Society Research Committee and PublicHealth Laboratory Service. The aetiology, managementand outcome of severe community-acquired pneumonia onthe intensive care unit. Respir Med 1992;86:7–13.

67 Torres A, Serra-Batlles J, Ferrer A, et al. Severe community-acquired pneumonia. Epidemiology and prognostic fac-tors. Am Rev Respir Dis 1991;144:312–8.

68 Woodhead M. Management of pneumonia in the outpatientsetting. Semin Respir Infect 1998;13:8–16.

69 Lim WS, Macfarlane JT, Boswell TC, et al. SCAPA: Studyof Community Acquired Pneumonia Aetiology in adultsadmitted to hospital: implications for management guide-lines. Thorax 2001;56:296–301.

70 Karalus NC, Cursons RT, Leng RA, et al. Communityacquired pneumonia: aetiology and prognostic index evalu-ation. Thorax 1991;46:413–8.

71 Neill AM, Martin IR, Weir R, et al. Community acquiredpneumonia: aetiology and usefulness of severity criteria onadmission. Thorax 1996;51:1010–6.

72 Lim I, Shaw DR, Stanley DP, et al. A prospective hospitalstudy of the aetiology of community-acquired pneumonia.Med J Aust 1989;151:87–91.

73 Alkhayer M, Jenkins PF, Harrison BDW. The outcome ofcommunity acquired pneumonia treated on the intensivecare unit. Respir Med 1990;84:13–6.

74 Leroy O, Georges H, Beuscart C, et al. Severe community-acquired pneumonia in intensive care units: prospectivevalidation of a prognostic score. Intensive Care Med1996;22:1307–14.

75 Moine P, Vercken JB, Chevret S, et al. Severe community-acquired pneumonia. Etiology, epidemiology, and progno-sis factors. French Study Group for Community-AcquiredPneumonia in the Intensive Care Unit. Chest 1994;105:1487–95.

76 Niederman MS, McCombs JS, et al. The cost of treatingcommunity-acquired pneumonia. Clin Ther 1998;20:820–37.

77 Ruiz M, Ewig S, Marcos MA, et al. Etiology of community-acquired pneumonia: impact of age, comorbidity, andseverity. Am J Respir Crit Care Med 1999;160:397–405.

78 Ausina V, Coll P, Sambeat M, et al. Prospective study on theetiology of community-acquired pneumonia in childrenand adults in Spain. Eur J Clin Microbiol Infect Dis 1988;7:342–7.

79 Logroscino CD, Penza O, Locicero S, et al. Community-acquired pneumonia in adults: a multicentric observationalAIPO study. Monaldi Arch Chest Dis 1999;54:11–17.

80 Leroy O, Guilley J, Georges H, et al. EVect of hospital-acquired ventilator-associated pneumonia on mortality ofsevere community-acquired pneumonia. J Crit Care 1999;14:12–19.

81 Ostergaard L, Andersen PL. Etiology of community-acquired pneumonia: evaluation by transtracheal aspira-tion, blood culture, or serology. Chest 1993;104:1400–7.

82 Torres A, Dorca J, Zalacain R, et al. Community-acquiredpneumonia in chronic obstructive pulmonary disease: aSpanish multicenter study. Am J Respir Crit Care Med1996;154:1456–61.

83 Marrie TJ. Bacteraemic pneumococcal pneumonia: acontinuously evolving disease. J Infect 1992;24:247–55.

84 Marrie TJ, Durant H, Kwan C. Nursing home-acquiredpneumonia. A case-control study. J Am Geriatr Soc1986;34:697–702.

85 Garb JL, Brown RB, Garb JR, et al. DiVerences in etiology ofpneumonias in nursing home and community patients.JAMA 1978;240:2169–72.

86 Drinka PJ, Gauerke C, Voeks S, et al. Pneumonia in a nurs-ing home. J Gen Intern Med 1994;9:650–2.

87 Leroy O, Vandenbussche C, CoYnier C, et al. Community-acquired aspiration pneumonia in intensive care units. Epi-demiological and prognosis data. Am J Respir Crit Care Med1997;156:1922–9.

88 Sopena N, Sabria-Leal M, Pedro-Botet ML, et al. Compara-tive study of the clinical presentation of legionella pneumo-nia and other community-acquired pneumonias. Chest1998;113:1195–200.

89 Stout JE, Yu VL. Legionellosis. N Engl J Med 1997;337:682–7.

90 Mier L, Dreyfuss D, Darchy B, et al. Is penicillin G anadequate initial treatment for aspiration pneumonia? Aprospective evaluation using a protected specimen brushand quantitative cultures. Intensive Care Med 1993;19:279–84.

91 Kim PE, Musher DM, Glezen WP, et al. Association ofinvasive pneumococcal disease with season, atmosphericconditions, air pollution, and the isolation of respiratoryviruses. Clin Infect Dis 1996;22:100–6.

92 Mercat A, Nguyen J, Dautzenberg B. An outbreak of pneu-mococcal pneumonia in two men’s shelters. Chest 1991;99:147–51.

93 Hoge CW, Reichler MR, Dominguez EA, et al. An epidemicof pneumococcal disease in an overcrowded, inadequatelyventilated jail. N Engl J Med 1994;331:643–8.

94 Tan Y-K, Khoos K-L, Chin S-P, Ong Y-Y. Aetiology andoutcome of severe community-acquired pneumonia in Sin-gapore. Eur Respir J 1998;12:113–5.

95 Chan CHS, Cohen M, Pang J. A prospective study ofcommunity-acquired pneumonia in Hong Kong. Chest1992;101:442–6.

96 Joseph CA, Dedman D, Birtles R, et al. Legionnaires’disease surveillance: England and Wales, 1993. CommunDis Rep 1994;4:R109–11.

97 Joseph CA, Hutchinson EJ, Dedman D, et al. Legionnaires’disease surveillance: England and Wales 1994. Commun DisRep 1995;5:R180–3.

98 Newton LH, Joseph CA, Hutchinson EJ, et al. Legionnaires’disease surveillance: England and Wales, 1995. CommunDis Rep CDR Rev 1996;6:R151–5.

99 Joseph CA, Harrison TG, Ilijic-Car D, et al. Legionnaires’disease in residents of England and Wales: 1996. CommunDis Rep 1997;7:R153–9.

100 Joseph CA, Harrison TG, Ilijic-Car D, et al. Legionnaires’disease in residents of England and Wales: 1998. CommunDis Rep 1999;280–4.

101 Anon. Acute respiratory infections, and influenza vaccinefor 1999/2000. Commun Dis Rep 1999;9:87–90.

102 Saikku P, Wang SP, Kleemola M, et al. An epidemic of mildpneumonia due to an unusual strain of Chlamydia psittaci.J Infect Dis 1985;151:832–9.

103 Kleemola M, Saikku P, Visakorpi R, et al. Epidemics ofpneumonia caused by TWAR, a new Chlamydia organism,in military trainees in Finland. J Infect Dis 1988;157:230–6.

104 Troy CJ, Peeling RW, Ellis AG, et al. Chlamydia pneumo-niae as a new source of infectious outbreaks in nursinghomes. JAMA 1997;277:1214–8; published erraturmJAMA 1997;278:118.

105 Kauppinen MT, Saikku P, Kujala P, et al. Clinical pictureof community-acquired Chlamydia pneumoniae pneumo-nia requiring hospital treatment: a comparison betweenchlamydial and pneumococcal pneumonia. Thorax 1996;51:185–9.

106 Wreghitt T. Chlamydial infection of the respiratory tract.Commun Dis Rep 1993;3:R119–24.

107 Pebody RG, Wall PG, Ryan MJ, et al. Epidemiological fea-tures of Coxiella burnetii infection in England and Wales:1984 to 1994. Commun Dis Rep CDR Rev 1996;6:R128–32.

108 Hutchinson EJ, Joseph CA, Zambon M, et al. Influenzasurveillance in England and Wales: October 1995 to June1996. Commun Dis Rep 1996;6:R163–9.

109 Connolly AM, Salmon RL, Lervy B, Williams DH. Whatare the complications of influenza and can they beprevented? Experience from the 1989 epidemic of H3N2influenza A in general practice. BMJ 1993;306:1452–4.

110 Metlay JP, Kapoor WN, Fine MJ. Does this patient havecommunity-acquired pneumonia? Diagnosing pneumoniaby history and physical examination. JAMA 1997;278:1440–5.

111 Spiteri MA, Cook DG, Clarke SW. Reliability of elicitingphysical signs in examination of the chest. Lancet1988;i:873–5.

112 Marrie TJ. Pneumonia in the elderly. Curr Opin Pulm Med1996;2:192–7.

113 Fang GD, Fine M, OrloV J, et al. New and emergingetiologies for community-acquired pneumonia with impli-cations for therapy. A prospective multicenter study of 359cases. Medicine 1990;69:307–16.

114 Farr BM, Kaiser DL, Harrison BD, et al. Prediction ofmicrobial aetiology at admission to hospital for pneumoniafrom the presenting clinical features. Thorax 1989;44:1031–5.

115 Riquelme R, Torres A, El Ebiary M, et al. Community-acquired pneumonia in the elderly. Clinical and nutritionalaspects. Am J Respir Crit Care Med 1997;156:1908–14.

116 Bohte R, Hermans J, van den Broek PJ. Early recognitionof Streptococcus pneumoniae in patients with community-acquired pneumonia. Eur J Clin Microbiol Infect Dis1996;15:201–5.

117 Marrie TJ. Mycoplasma pneumoniae pneumonia requiringhospitalization, with emphasis on infection in the elderly.Arch Intern Med 1993;153:488–94.

118 Helms CM, Viner JP, Sturm RH, et al. Comparativefeatures of pneumococcal, mycoplasmal, and Legionnaires’disease pneumonias. Ann Intern Med 1979;90:543–7.

119 Helms CM, Wintermeyer LA, Zeitler RR, et al. Anoutbreak of community-acquired Legionnaires’ diseasepneumonia. Am J Public Health 1984;74:835–6.


www.thoraxjnl.com



120 Lieberman D, Porath A, SchlaeVer F, et al. Legionella spe-cies community-acquired pneumonia. A review of 56 hos-pitalized adult patients. Chest 1996;109:1243–9.

121 Granados A, Podzamczer D, Gudiol F, et al. Pneumoniadue to Legionella pneumophila and pneumococcalpneumonia: similarities and diVerences on presentation.Eur Respir J 1989;2:130–4.

122 Falco V, Fernandez dS, Alegre J, et al. Legionellapneumophila. A cause of severe community-acquiredpneumonia. Chest 1991;100:1007–11.

123 Lieberman D, Ben Yaakov M, Lazarovich Z, et al. Chlamy-dia pneumoniae community-acquired pneumonia: a reviewof 62 hospitalized adult patients. Infection 1996;24:109–14.

124 File TM Jr, PlouVe JF Jr, Breiman RF, et al. Clinical char-acteristics of Chlamydia pneumoniae infection as the solecause of community-acquired pneumonia. Clin Infect Dis1999;29:426–8.

125 Lieberman D, Lieberman D, Boldur I, et al. Q-fever pneu-monia in the Negev region of Israel: a review of 20 patientshospitalised over a period of one year. J Infect 1995;30:135–40.

126 Goyette M, Bouchard J, Poirier A, et al. Importance of Qfever in community acquired pneumonia. Can J Infect Dis1996;7:370–3.

127 Caron F, Meurice JC, Ingrand P, et al. Acute Q feverpneumonia: a review of 80 hospitalized patients. Chest1998;114:808–13.

128 Sobradillo V, Ansola P, Baranda F, et al. Q feverpneumonia: a review of 164 community-acquired cases inthe Basque country. Eur Respir J 1989;2:263–6.

129 Feldman C, Kallenbach JM, Levy H, et al. Comparison ofbacteraemic community-acquired lobar pneumonia due toStreptococcus pneumoniae and Klebsiella pneumoniae inan intensive care unit. Respiration 1991;58:265–70.

130 Rudin ML, Michael JR, Huxley EJ. Community-acquiredacinetobacter pneumonia. Am J Med 1979;67:39–43.

131 Henderson A, Wall D. Streptococcus milleri liver abscesspresenting as fulminant pneumonia. Aust NZ J Surg 1993;63:237–40.

132 Goolam MA, Feldman C, Smith C, et al. Does primaryStreptococcus viridans pneumonia exist? S Afr Med J1992;82:432–4.

133 Metlay JP, Schulz R, Li YH, et al. Influence of age onsymptoms at presentation in patients with community-acquired pneumonia. Arch Intern Med 1997;157:1453–9.

134 Marrie TJ, Haldane EV, Faulkner RS, et al. Community-acquired pneumonia requiring hospitalization. Is it diVer-ent in the elderly? J Am Geriatr Soc 1985;33:671–80.

135 Marrie TJ, Blanchard W. A comparison of nursing home-acquired pneumonia patients with patients withcommunity-acquired pneumonia and nursing home pa-tients without pneumonia. J Am Geriatr Soc 1997;45:50–5.

136 Macfarlane JT, Miller AC, Roderick Smith WH, et al.Comparative radiographic features of community acquiredLegionnaires’ disease, pneumococcal pneumonia, myco-plasma pneumonia, and psittacosis. Thorax 1984;39:28–33.

137 Prout S, Potgieter PD, Forder AA, et al. Acute community-acquired pneumonias. S Afr Med J 1983;64:443–6.

138 Kauppinen MT, Lahde S, Syrjala H. Roentgenographicfindings of pneumonia caused by Chlamydia pneumoniae.A comparison with streptococcus pneumonia. Arch InternMed 1996;156:1851–6.

139 Gikas A, Kofteridis D, Bouros D, et al. Q fever pneumonia:appearance on chest radiographs. Radiology 1999;210:339–43.

140 Macfarlane J, Rose D. Radiographic features of staphylo-coccal pneumonia in adults and children. Thorax 1996;51:539–40.

141 Korvick JA, Hackett AK, Yu VL, et al. Klebsiella pneumo-nia in the modern era: clinicoradiographic correlations.South Med J 1991;84:200–4.

142 Tanaka N, Matsumoto T, Kuramitsu T, et al. High resolu-tion CT findings in community-acquired pneumonia. JComput Assist Tomogr 1996;20:600–8.

143 Syrjala H, Broas M, Suramo I, et al. High-resolution com-puted tomography for the diagnosis of community-acquired pneumonia. Clin Infect Dis 1998;27:358–63.

144 Reittner P, Muller NL, Heyneman L, et al. Mycoplasmapneumoniae pneumonia: radiographic and high-resolutionCT features in 28 patients. AJR 2000;174:37–41.

145 Beall DP, Scott WW Jr, Kuhlman JE, et al. Utilization ofcomputed tomography in patients hospitalized withcommunity-acquired pneumonia. Md Med J 1998;47:182–7.

146 Mittl RL Jr, Schwab RJ, Duchin JS, et al. Radiographicresolution of community-acquired pneumonia. Am J RespirCrit Care Med 1994;149:630–5.

147 Jay SJ, Johanson WG Jr, Pierce AK. The radiographicresolution of Streptococcus pneumoniae pneumonia. NEngl J Med 1975;293:798–801.

148 Granet KM, Wallach SL, Horvath K, et al. Chestradiographs in patients with community-acquired pneumo-nia. N Engl J Med 1996;93:37–41.

149 Holmberg H, Kragsbjerg P. Association of pneumonia andlung cancer: the value of convalescent chest radiographyand follow-up. Scand J Infect Dis 1993;25:93–100.

150 Gibson SP, Weir DC, Burge PS. A prospective audit of thevalue of fibreoptic bronchoscopy in adults admitted withcommunity acquired pneumonia. Respir Med 1993;87:105–9.

151 Holmberg H, Bodin L, Jonsson I, et al. Rapid aetiologicaldiagnosis of pneumonia based on routine laboratoryfeatures. Scand J Infect Dis 1990;5:537–45.

152 Smith RP, Lipworth BJ, Cree IA, et al. C-reactive protein.A clinical marker in community-acquired pneumonia.Chest 1995;108:1288–91.

153 Hansson LO, Hedlund JU, Ortqvist A. Sequential changesof inflammatory and nutritional markers in patients withcommunity-acquired pneumonia. Scand J Clin Lab Invest1997;57:111–8.

154 Ortqvist A, Hedlund J, Wretlind B, et al. Diagnostic andprognostic value of interleukin-6 and C-reactive protein incommunity-acquired pneumonia. Scand J Infect Dis 1995;27:457–62.

155 Sanyal S, Smith PR, Saha AC, et al. Initial microbiologicstudies did not aVect outcome in adults hospitalized withcommunity-acquired pneumonia. Am J Respir Crit CareMed 1999;160:346–8.

156 Menendez R, Cordoba J, DeLa CP, et al. Value of thepolymerase chain reaction assay in noninvasive respiratorysamples for diagnosis of community-acquired pneumonia.Am J Respir Crit Care Med 1999;159:1868–73.

157 Ewig S, Bauer T, Hasper E, et al. Value of routine microbialinvestigation in community-acquired pneumonia treated ina tertiary care center. Respiration 1996;63:169.

158 Woodhead MA, Arrowsmith J, Chamberlain-Webber R, etal. The value of routine microbial investigation incommunity-acquired pneumonia. Respir Med 1991;85:313–7.

159 Taylor EL, Marrie TJ, Fine MJ, et al. Observations from amulticentre study on the use of the sputum specimen inpatients hospitalized with community-acquired pneumo-nia. Can J Infect Dis 1999;10:39–46.

160 Reed WW, Byrd GS, Gates RH Jr, et al. Sputum Gram’sstain in community-acquired pneumococcal pneumonia. Ameta-analysis. Western J Med 1996;165:197–204.

161 Rein MF, Gwaltney JM Jr, O’Brien WM, et al. Accuracy ofGram’s stain in identifying pneumococci in sputum. JAMA1978;239:2671–3.

162 Fine MJ, OrloV JJ, Rihs JD, et al. Evaluation of housestaVphysicians’ preparation and interpretation of sputum Gramstains for community-acquired pneumonia. J Gen InternMed 1991;6:189–98.

163 Croce MA, Fabian TC, Waddle-Smith L, et al. Utility ofGram’s stain and eYcacy of quantitative cultures for post-traumatic pneumonia: a prospective study. Ann Surg 1998;227:743–51.

164 Venkatesan P, Macfarlane JT. Role of pneumococcalantigen in the diagnosis of pneumococcal pneumonia. Tho-rax 1992;47:329–31.

165 Farrington M, Rubenstein D. Antigen detection inpneumococcal pneumonia. J Infect 1991;23:109–16.

166 Jalonen E, Paton JC, Koskela M, et al. Measurement ofantibody responses to pneumolysin: a promising methodfor the presumptive aetiological diagnosis of pneumococcalpneumonia. J Infect 1989;19:127–34.

167 Birtles RJ, Harrison TG, Samuel D, et al. Evaluation ofurinary antigen ELISA for diagnosing Legionella pneu-mophila serogroup 1 infection. J Clin Pathol 1990;43:685–90.

168 Boswell TC, Marshall LE, Kudesia G. False-positivelegionella titres in routine clinical serology testing detectedby absorption with campylobacter: implications for theserological diagnosis of legionnaires’ disease. J Infect 1996;32:23–6.

169 Medical Devices Agency. Mycoplasma pneumoniae anti-body assays. Medical Devices Agency Evaluation ReportMDA/98/62 1998.

170 Sillis M, White P, Caul EO, et al. The diVerentiation ofChlamydia species by antigen detection in sputumspecimens from patients with community-acquired acuterespiratory infections. J Infect 1992;25(Suppl 1):77–86.

171 Garnett P, Brogan O, Lafong C, et al. Comparison of throatswabs with sputum specimens for the detection of Chlamy-dia pneumoniae antigen by direct immunofluorescence. JClin Pathol 1998;51:309–11.

172 Boman J, Allard A, Persson K, et al. Rapid diagnosis of res-piratory Chlamydia pneumoniae infection by nestedtouchdown polymerase chain reaction compared with cul-ture and antigen detection by EIA. J Infect Dis 1997;175:1523–6.

173 Sillis M, Wreghitt TG, Paul ID, et al. Chlamydial respira-tory infections. Don’t get bogged down by diVerentiatingspecies. BMJ 1993;307:62–3.

174 Kutlin A, Tsumura N, Emre U, et al. Evaluation ofChlamydia immunoglobulin M (IgM), IgG, and IgA rELI-SAs Medac for diagnosis of Chlamydia pneumoniae infec-tion. Clin Diag Lab Immun 1997;4:213–6.

175 Ewig S, Ruiz M, Mensa J, et al. Severe community-acquired pneumonia. Assessment of severity criteria. Am JRespir Crit Care Med 1998;158:1102–8.

176 Meyer RJ, Town GI, Harre E, et al. An audit of the assess-ment and management of adults admitted to ChristchurchHospital with community acquired pneumonia. NZ Med J1997;110:349–52.

177 Farr BM, Sloman AJ, Fisch MJ. Predicting death inpatients hospitalized for community-acquired pneumonia.Ann Intern Med 1991;115:428–36.

178 Ewig S, Bauer T, Hasper E, et al. Prognostic analysis andpredictive rule for outcome of hospital treated community-acquired pneumonia . Eur Respir J 1995;8:392–7.

179 Fine MJ, Smith MA, Carson CA, et al. Prognosis and out-comes of patients with community-acquired pneumonia. Ameta-analysis. JAMA 1996;275:134–41.

180 Fine MJ, Singer DE, Hanusa BH, et al. Validation of apneumonia prognostic index using the MedisGroupsComparative Hospital Database. Am J Med 1993;94:153–9.

181 Hedlund J, Kalin M, Ortqvist A. Recurrence of pneumoniain middle-aged and elderly adults after hospital-treatedpneumonia: aetiology and predisposing conditions. Scand JInfect Dis 1997;29:387–92.


www.thoraxjnl.com



182 Conte HA, Chen YT, Mehal W, et al. A prognostic rule forelderly patients admitted with community-acquired pneu-monia. Am J Med 1999;106:20–8.

183 Janssens JP, Gauthey L, Herrmann F, et al. Community-acquired pneumonia in older patients. J Am Geriatr Soc1996;44:539–44.

184 Riquelme R, Torres A, El-Ebiary M, et al. Community-acquired pneumonia in the elderly: a multivariate analysisof risk and prognostic factors. Am J Respir Crit Care Med1996;154:1450–5.

185 Leroy O, Devos P, Guery B, et al. Simplified prediction rulefor prognosis of patients with severe community acquiredpneumonia in intensive care units. Chest 1999;116:157–65.

186 Fine MJ, Hanusa BH, Lave JR, et al. Comparison of adisease-specific and a generic severity of illness measure forpatients with community-acquired pneumonia. J GenIntern Med 1995;10:359–68.

187 Pachon J, Prados MD, Capote F, et al. Severe community-acquired pneumonia. Etiology, prognosis, and treatment.Am Rev Respir Dis 1990;142:369–73.

188 Leroy O, Santre C, Beuscart C, et al. A five-year study ofsevere community-acquired pneumonia with emphasis onprognosis in patients admitted to an intensive care unit.Intensive Care Med 1995;21:24–31.

189 Fine MJ, Auble TE, Yealy DM, et al. A prediction rule toidentify low-risk patients with community-acquired pneu-monia. N Engl J Med 1997;336:243–50.

190 Ortqvist A, Hedlund J, Grillner L, et al. Aetiology, outcomeand prognostic factors in community-acquired pneumoniarequiring hospitalization. Eur Respir J 1990;3:1105–13.

191 van Eeden SF, Coetzee AR, Joubert JR. Community-acquired pneumonia: factors influencing intensive careadmission. S Afr Med J 1988;73:77–81.

192 Qureshi KN, Hodkinson HM. Evaluation of a ten-question mental test in the institutionalized elderly. AgeAging 1974;3:152–7.

193 Gomez J, Banos V, Gomez JR, et al. Prospective study ofepidemiology and prognostic factors in community-acquired pneumonia. Eur J Clin Microbiol Infect Dis1996;15:556–60.

194 Feldman C, Kallenbach JM, Levy H, et al. Community-acquired pneumonia of diverse aetiology: prognosticfeatures in patients admitted to an intensive care unit and a“severity of illness” core. Intensive Care Med 1989;15:302–7.

195 Hasley PB, Albaum MN, Li Y-H, et al. Do pulmonaryradiographic findings at presentation predict mortality inpatients with community-acquired pneumonia? Arch InternMed 1996;156:2206–12.

196 Tapson VF. Risk factors in community-acquiredpneumonia: can they be quantified? Eur Respir Rev 1998;8:291–4.

197 Marras TK, Chan CK. Use of guidelines in treatingcommunity-acquired pneumonia. Chest 1998;113:1689–94.

198 Atlas SJ, Benzer TI, Borowsky LH, et al. Safely increasingthe proportion of patients with community-acquired pneu-monia treated as outpatients: an interventional trial. ArchIntern Med 1998;158:1350–6.

199 Marrie TJ, Lau CY, Wheeler SL, et al. A controlled trial ofa critical pathway for treatment of community-acquiredpneumonia. CAPITAL Study Investigators. Community-Acquired Pneumonia Intervention Trial Assessing Levo-floxacin. JAMA 2000;283:749–55.

200 Farr BM, Woodhead MA, Macfarlane JT, et al. Risk factorsfor community-acquired pneumonia diagnosed by generalpractitioners in the community. Respir Med 2000;94:422–7.

201 Tydeman DE, Chandler AR, Graveling BM, et al. Aninvestigation into the eVects of exercise tolerance trainingon patients with chronic airways obstruction. Physiotherapy1984;70:261–4.

202 LaCroix AZ, Lipson S, Miles TP, et al. Prospective study ofpneumonia hospitalizations and mortality of U.S. olderpeople: the role of chronic conditions, health behaviors,and nutritional status. Public Health Rep 1989;104:350–60.

203 Fulmer JD, Snider GL. American College of ChestPhysicians/National Heart, Lung, and Blood InstituteNational Conference on Oxygen Therapy. Heart Lung1984;13:550–62.

204 JeVrey AA, Warren PM, Flenley DC. Acute hypercapnicrespiratory failure in patients with chronic obstructive lungdisease: risk factors and use of guidelines for management.Thorax 1992;47:34–40.

205 Tydeman D. An investigation into the eVectiveness ofphysiotherapy in the treatment of patients with communityacquired pneumonia. Physio Pract 1989;5:75–8.

206 Bjorkqvist M, Wiberg B, Bodin L, et al. Bottle-blowing inhospital-treated patients with community-acquired pneu-monia. Scand J Infect Dis 1997;29:77–82.

207 Confalonieri M, Potena A, Carbone G, et al. Acute respira-tory failure in patients with severe community-acquiredpneumonia. A prospective randomized evaluation ofnoninvasive ventilation. Am J Respir Crit Care Med1999;160:1585–91.

208 Bone RC, Balk RA, Cerra FB, et al. Definitions for sepsisand organ failure and guidelines for the use of innovativetherapies in sepsis. The American College of ChestPhysicians/Society of Critical Care Medicine (ACCP/SCCM) Consensus Conference Committee. Chest 1992;101:1644–55.

209 Marik P, Kraus P, Sribante J, et al. Hydrocortisone andtumor necrosis factor in severe community-acquired pneu-monia. A randomized controlled study. Chest 1993;104:389–92.

210 Riley TV. Clostridium diYcile: a pathogen of the nineties.Eur J Clin Microbiol Infect Dis 1998;17:137–41.

211 Zadik PM, Moore AP. Antimicrobial associations of anoutbreak of diarrhoea due to Clostridium diYcile. J HospInfect 1998;39:189–93.

212 MacGowan AP, Feeney R, Brown I, et al. Health careresource utilization and antimicrobial use in elderlypatients with community-acquired lower respiratory tractinfection who develop Clostridium diYcile-associateddiarrhoea. J Antimicrob Chemother 1997;39:537–41.

213 Settle CD, Wilcox MH, Fawley WN, et al. Prospectivestudy of the risk of Clostridium diYcile diarrhoea in elderlypatients following treatment with cefotaxime orpiperacillin-tazobactam. Aliment Pharmacol Ther 1998;12:1217–23.

214 Spencer RC. The role of antimicrobial agents in theaetiology of Clostridium diYcile-associated disease. J Anti-microb Chemother 1998;41(Suppl C):21–7.

215 Ludlum H, Brown N, Sule O, et al. An antibiotic policyassociated with reduced risk of Clostridium diYcile-associated diarrhoea. Age Aging 1999;28:578–80.

216 Bignardi GE. Risk factor for Clostridium diYcile infection.J Hosp Infect 1998;40:1–15.

217 Department of Health and Public Health LaboratoryService Joint Working Group. The prevention and manage-ment of Clostridium diYcile infection. London: Department ofHealth, 1994.

218 Wilcox MH. Respiratory antibiotics use and ClostridiumdiYcile infection: is it the drugs or is it the doctors? Thorax2000;55:633–4.

219 Boswell TC, Nye KJ, Smith EG. Increased incidence ofClostridium diYcile infection. J Hosp Infect 1998;39:78–9.

220 Johnson RH, Levine S, Traub SL, et al. Pneumonia StudyGroup. Sequential intravenous/oral ciprofloxacin com-pared with parenteral ceftriaxone in the treatment of hospi-talised patients with community acquired pneumonia.Infect Dis Clin Pract 1996;5:265–72.

221 Rizzato G, Allegra L. EYcacy and tolerability of ateicoplanin-ciprofloxacin combination in severecommunity-acquired pneumonia. Clin Drug Invest 1997;14:337–45.

222 Powell M, Yeo SF, Seymour A, et al. Antimicrobialresistance in Haemophilus influenzae from England andScotland in 1991. J Antimicrob Chemother 1992;29:547–54.

223 Felmingham D, Washington J. Trends in the antimicrobialsusceptibility of bacterial respiratory tract pathogens: find-ings of the Alexander Project 1992–1996. J Chemother1999;11(Suppl 1):5–21.

224 Andrews J, Ashby J, Jevons G, et al. A comparison of anti-microbial resistance rates in Gram-positive pathogensisolated in the UK from October 1996 to January 1997 andOctober 1997 to January 1998. J Antimicrob Chemother2000;45:285–93.

225 Johnson AP. Antibiotic resistance among clinically impor-tant gram-positive bacteria in the UK. J Hosp Infect1998;40:17–26.

226 Pallares R, Linares J, Vadillo M, et al. Resistance to penicil-lin and cephalosporin and mortality from severe pneumo-coccal pneumonia in Barcelona, Spain. N Engl J Med 1995;333:474–80.

227 Einarsson S, Kristjansson M, Kristinsson KG, et al. Pneu-monia caused by penicillin-non-susceptible and penicillin-susceptible pneumococci in adults: a case-control study.Scand J Infect Dis 1998;30:253–6.

228 Klugman K, Feldman C. The clinical relevance ofantibiotic resistance in the management of pneumococcalpneumonia. Infect Dis Clin Pract 1998;7:180–4.

229 Pallares R, Viladrich PF, Linares J, et al. Impact ofantibiotic resistance on chemotherapy for pneumococcalinfections. Microb Drug Resist 1998;4:339–47.

230 Garau J, Aguilar L, Rodriguez-Creixems M, et al. Influenceof comorbidity and severity on the clinical outcome of bac-teremic pneumococcal pneumonia treated with beta-lactam monotherapy. J Chemother 1999;11:266–72.

231 Ewig S, Kleinfeld T, Bauer T, et al. Comparative validationof prognostic rules for community acquired pneumonia inan elderly population. Eur Respir J 1999;14:370–5.

232 Chen DK, McGeer A, de Azavedo JC, et al. Decreasedsusceptibility of Streptococcus pneumoniae to fluoroqui-nolones in Canada. Canadian Bacterial SurveillanceNetwork. N Engl J Med 1999;341:233–9.

233 Nielsen K, Bangsborg JM, Hoiby N. Susceptibility ofLegionella species to five antibiotics and development ofresistance by exposure to erythromycin, ciprofloxacin, andrifampicin. Diagn Microbiol Infect Dis 2000;36:43–8.

234 House of Lords Select Committee on Science andTechnology. Resistance to antibiotics and other antimicrobialagents. London: HMSO, 1998.

235 Honeybourne D. Antibiotic penetration in the respiratorytract and implications for the selection of antimicrobialtherapy. Curr Opin Pulm Med 1997;3:170–4.

236 Wise R, Honeybourne D. Pharmacokinetics and pharma-codynamics of fluoroquinolones in the respiratory tract.Eur Respir J 1999;14:221–9.

237 Drusano GL, Craig WA. Relevance of pharmacokineticsand pharmacodynamics in the selection of antibiotics forrespiratory tract infections. J Chemother 1997;9(Suppl3):38–44.

238 Craig WA. Pharmacokinetic/pharmacodynamic parameters:rationale for antibacterial dosing of mice and men. ClinInfect Dis 1998;26:1–10.

239 Preston SL, Drusano GL, Berman AL, et al. Pharmacody-namics of levofloxacin: a new paradigm for early clinicaltrials. JAMA 1998;279:125–9.


www.thoraxjnl.com



240 Preston SL, Drusano GL, Berman AL, et al. Levofloxacinpopulation pharmacokinetics and creation of a demo-graphic model for prediction of individual drug clearancein patients with serious community-acquired infection.Antimicrob Agents Chemother 1998;42:1098–104.

241 Anderson G, Esmonde TS, Coles S, et al. A comparativesafety and eYcacy study of clarithromycin and erythromy-cin stearate in community-acquired pneumonia. J Antimi-crob Chemother 1991;27(Suppl A):117–24.

242 Chien SM, Pichotta P, Siepman N, et al. Treatment ofcommunity-acquired pneumonia. A multicenter, double-blind, randomized study comparing clarithromycin witherythromycin. Canada-Sweden Clarithromycin-PneumoniaStudy Group. Chest 1993;103:697–701.

243 Green JA, Butler T, Todd WM. Randomized double-blindtrial of the comparative eVecacy and safety of cefpodoximeproxetil and cefaclor in the treatment of acute community-acquired pneumonia. Curr Ther Res 1994;55:1003–15.

244 Bohte R, van’t Wout JW, Lobatto S, et al. EYcacy andsafety of azithromycin versus benzylpenicillin or erythro-mycin in community-acquired pneumonia. Eur J ClinMicrobiol Infect Dis 1995;14:182–7.

245 Chan R, Hemeryck L, O’Regan M, et al. Oral versus intra-venous antibiotics for community acquired lower respira-tory tract infection in a general hospital: open, randomisedcontrolled trial. BMJ 1995;310:1360–2.

246 PlouVe JF, Herbert MT, File TM, et al. Ofloxacin versusstandard therapy in treatment of community-acquiredpneumonia requiring hospitalization. Pneumonia StudyGroup. Antimicrob Agents Chemother 1996;40:1175–9.

247 File TM Jr, Segreti J, Dunbar L, et al. A multicenter, rand-omized study comparing the eYcacy and safety ofintravenous and/or oral levofloxacin versus ceftriaxoneand/or cefuroxime axetil in treatment of adults withcommunity-acquired pneumonia. Antimicrob Agents Chemo-ther 1997;41:1965–72.

248 Van den Brande P, Vondra V, Vogel F, et al. Sequentialtherapy with cefuroxime followed by cefuroxime axetil incommunity-acquired pneumonia. Chest 1997;112:406–15.

249 Vetter N, Stamler D, O’Neil S, et al. Clarithromycin vscombined cefuroxime and erythromycin in the treatment ofhospitalized community-acquired pneumonia patients:intravenous followed by oral therapy. Clin Drug Invest1997;14:439–49.

250 O’Doherty B, Muller O. Randomized, multicentre study ofthe eYcacy and tolerance of azithromycin versus clarithro-mycin in the treatment of adults with mild to moderatecommunity-acquired pneumonia. Azithromycin StudyGroup. Eur J Clin Microbiol Infect Dis 1998;17:828–33.

251 Carbon C, Ariza H, Rabie WJ, et al. Comparative study oflevofloxacin and amoxycillin/clavulanic acid in adults withmild-to-moderate community-acquired pneumonia. ClinMicrobiol Infect 1999;5:724–32.

252 Rovira E, Martinez-Moragon E, Belda A, et al. Treatmentof community-acquired pneumonia in outpatients: rand-omized study of clarithromycin alone versus clarithromycinand cefuroxime. Respiration 1999;66:413–8.

253 Johnson RH, Levine S, Traub SL. Sequential intravenous/oral ciprofloxacin compared with parenteral ceftriaxone inthe treatment of hospitalized patients with community-acquired pneumonia. Infect Dis Clin Pract 1996;5:265–72.

254 Bartoloni A, Strohmeyer M, Corti G, et al. Multicenterrandomized trial comparing meropenem (1.5 g daily) andimipenem/cilastatin (2 g daily) in the hospital treatment ofcommunity-acquired pneumonia. Drugs Exp Clin Res1999;25:243–52.

255 Baumgartner JD, Glauser MP. Tolerance study of ceftriax-one compared with amoxicillin in patients with pneumo-nia. Am J Med 1984;77:54–8.

256 Fredlund H, Bodin L, Back E, et al. Antibiotic therapy inpneumonia: a comparative study of parenteral and oraladministration of penicillin. Scand J Infect Dis 1987;19:459–66.

257 Zuck P, Rio Y, Ichou F. EYcacy and tolerance ofcefpodoxime proxetil compared with ceftriaxone in vulner-able patients with bronchopneumonia. J Antimicrob Chem-other 1990;26(Suppl E):71–7.

258 Poirier R. Comparative study of clarithromycin androxithromycin in the treatment of community-acquiredpneumonia. J Antimicrob Chemother 1991;27(Suppl A):109–16.

259 Koulla-Shiro S, Kuaban C, Auckenthaler R, et al. Adultresponse to initial treatment with ampicillin in communityacquired pneumonia in Yaounde, Cameroon. Cent Afr JMed 1993;39:188–92.

260 Gaillat J, Bru JP, Sedallian A. Penicillin G/ofloxacin versuserythromycin/amoxicillin-clavulanate in the treatment ofsevere community-acquired pneumonia. Eur J Clin Micro-biol Infect Dis 1994;13:639–44.

261 Low DE, Mandell LA. A prospective open-label multi-centre trial on the use of 1g, once daily ceftriaxone in lowerrespiratory tract infections. Can J Infect Dis 1994;5(SupplC):3–8C.

262 Uzun O, Hayran M, Akova M, et al. EYcacy of a three-daycourse of azithromycin in the treatment of community-acquired pneumococcal pneumonia. Preliminary report. JChemother 1994;6:53–7.

263 Jang TN, Liu CY, Wang FD, et al. A randomized compara-tive study on the safety and eYcacy of clarithromycin anderythromycin in treating community-acquired pneumonia.Chung Hua I Hsueh Tsa Chih (Taipei) 1995;55:302–6.

264 Rizzato G, Montemurro L, Fraioli P, et al. EYcacy of athree day course of azithromycin in moderately severecommunity-acquired pneumonia. Eur Respir J 1995;8:398–402.

265 Oh HM, Ng AW, Lee SK. Cefuroxime compared toamoxicillin-clavulanic acid in the treatment of community-acquired pneumonia. Singapore Med J 1996;37:255–7.

266 Siegel RE, Halpern NA, AlmenoV PL, et al. A prospectiverandomized study of inpatient iv. antibiotics forcommunity-acquired pneumonia. The optimal duration oftherapy. Chest 1996;110:965–71.

267 Genne D, Siegrist HH, Humair L, et al. Clarithromycinversus amoxicillin-clavulanic acid in the treatment ofcommunity-acquired pneumonia. Eur J Clin Microbiol InfectDis 1997;16:783–8.

268 Ho A, Leung R, Lai CK, et al. Hospitalized patients withcommunity-acquired pneumonia in Hong Kong: a rand-omized study comparing imipenem/cilastatin and ceftazi-dime. Respiration 1997;64:224–8.

269 Cabellos C, Ariza J, Barreiro B, et al. Current usefulness ofprocaine penicillin in the treatment of pneumococcalpneumonia. Eur J Clin Microbiol Infect Dis 1998;17:265–8.

270 Fogarty CM, Sullivan JG, Chattman MS, et al. Once a daylevofloxacin in the treatment of mild to moderate andsevere community-acquired pnenmonia in adults. Infect DisClin Pract 1998;7:400–7.

271 Rhew DC, Hackner D, Henderson L, et al. The clinicalbenefit of in-hospital observation in ‘low-risk’ pneumoniapatients after conversion from parenteral to oral anti-microbial therapy. Chest 1998;113:142–6.

272 Speich R, Imhof E, Vogt M, et al. EYcacy, safety, and toler-ance of piperacillin/tazobactam compared to co-amoxiclavplus an aminoglycoside in the treatment of severe pneumo-nia. Eur J Clin Microbiol Infect Dis 1998;17:313–7.

273 Wallace RJ Jr, Martin RR, Quinones FJ, et al. Ceforamideand cefazolin therapy of pneumonia: comparative clinicaltrial. Antimicrob Agents Chemother 1981;20:648–52.

274 Grieco MM, Lange M, Daniels JA, et al. Single-blind con-trolled study of ceftizoxime and cefamandole in thetreatment of community-acquired pneumonia. J AntimicrobChemother 1982;10(Suppl C):223–5.

275 Wallace RJ Jr, Niefield SL, Waters S, et al. Comparativetrial of cefonicid and cefamandole in the therapy ofcommunity-acquired pneumonia. Antimicrob Agents Chem-other 1982;21:231–5.

276 Keeton GR, Kehoe B, Phillips SW, et al. Ceftazidime andcefamandole in the treatment of pneumonia. J AntimicrobChemother 1983;12(Suppl A):27–30.

277 Geckler RW, McCormack GD, Goodman JS. Comparisonof cefonicid and cefamandole for the treatment ofcommunity-acquired infections of the lower respiratorytract. Rev Infect Dis 1984;6(Suppl 4):S847–52.

278 Weber DJ, Calderwood SB, Karchmer AW, et al. Ampicillinversus cefamandole as initial therapy for community-acquired pneumonia. Antimicrob Agents Chemother 1987;31:876–82.

279 Carbon C, Leophonte P, Petitpretz P, et al. EYcacy andsafety of temafloxacin versus those of amoxicillin in hospi-talized adults with community-acquired pneumonia. Anti-microb Agents Chemother 1992;36:833–9.

280 Kissling M. Cefetamet pivoxil in community-acquiredpneumonia: an overview. Curr Med Res Opin 1992;12:631–9.

281 Wiesner B, Wilen-Rosenqvist G, Lehtonen L. Twice dailydosing of erythromycin acistrate in the treatment of acutebronchitis and pneumonia. Arzneimittelforschung 1993;43:1014–7.

282 Aquino R, Lansang MA, Salvador A, et al. TheeVectiveness and safety of ceftriaxone to cefetamet pivoxilvs ceftazidime to cefetamet pivoxil vs cefoperazone to cefe-tamet pivoxil in the treatment of community-acquiredpneumonia in adult filipinos. Phil J Intern Med 1994;31:237–40.

283 Liippo K, Tala E, Puolijoki H, et al. A comparative study ofdirithromycin and erythromycin in bacterial pneumonia. JInfect 1994;28:131–9.

284 Williams D, Perri M, Zervos MJ. Randomized comparativetrial with ampicillin/sulbactam versus cefamandole in thetherapy of community acquired pneumonia. Eur J ClinMicrobiol Infect Dis 1994;13:293–8.

285 De Palma M, Rocchi D, Canepa G, et al. Single daily doseof cefodizime in patients with community-acquiredpneumonia: an open-label, controlled, randomized study.The Italian Multicentre Community-Acquired PneumoniaGroup. Clin Ther 1995;17:413–24.

286 Hopkins S, Williams D. Five-day azithromycin in the treat-ment of patients with community-acquired pneumonia.Curr Ther Res 1995;56:915–25.

287 Aubier M, Lode H, Gialdroni-Grassi G, et al. Sparfloxacinfor the treatment of community-acquired pneumonia: apooled data analysis of two studies. J Antimicrob Chemother1996;37(Suppl A):73–82.

288 Hernandez JM, Sides GD, Conforti PM, et al. Clinical eY-cacy of dirithromycin in patients with bacteremic pneumo-nia. Clin Ther 1996;18:1128–38.

289 Ortqvist A, Valtonen M, Cars O, et al. Oral empiric treat-ment of community-acquired pneumonia. A multicenter,double-blind, randomized study comparing sparfloxacinwith roxithromycin. The Scandinavian Sparfloxacin StudyGroup. Chest 1996;110:1499–506.

290 Portier H, May T, Proust A. Comparative eYcacy of spar-floxacin in comparison with amoxycillin plus ofloxacin inthe treatment of community-acquired pneumonia. FrenchStudy Group. J Antimicrob Chemother 1996;37(SupplA):83–91.


www.thoraxjnl.com



291 Rubinstein E. Safety profile of sparfloxacin in thetreatment of respiratory tract infections. J AntimicrobChemother 1996;37(Suppl A):145–60.

292 Donowitz GR, Brandon ML, Salisbury JP, et al. Spar-floxacin versus cefaclor in the treatment of patients withcommunity-acquired pneumonia: a randomized, double-masked, comparative, multicenter study. Clin Ther 1997;19:936–53.

293 Drehobl M, Bianchi P, Keyserling CH, et al. Comparisonof cefdinir and cefaclor in treatment of community-acquired pneumonia. Antimicrob Agents Chemother 1997;41:1579–83.

294 Ragnar NS. Atypical pneumonia in the Nordic countries:aetiology and clinical results of a trial comparing fleroxacinand doxycycline. Nordic Atypical Pneumonia StudyGroup. J Antimicrob Chemother 1997;39:499–508.

295 O’Doherty B, Dutchman DA, Pettit R, et al. Randomized,double-blind, comparative study of grepafloxacin andamoxycillin in the treatment of patients with community-acquired pneumonia. J Antimicrob Chemother 1997;40(Suppl A):73–81.

296 Aubier M, Verster R, Regamey C, et al. Once-dailysparfloxacin versus high-dosage amoxicillin in the treat-ment of community-acquired, suspected pneumococcalpneumonia in adults. Sparfloxacin European Study Group.Clin Infect Dis 1998;26:1312–20.

297 Salvarezza CR, Mingrone H, Fachinelli H, et al. Compari-son of roxithromycin with cefixime in the treatment ofadults with community-acquired pneumonia. J AntimicrobChemother 1998;41(Suppl B):75–80.

298 Tremolieres F, de Kock F, Pluck N, et al. Trovafloxacinversus high-dose amoxicillin (1 g three times daily) in thetreatment of community-acquired bacterial pneumonia.Eur J Clin Microbiol Infect Dis 1998;17:447–53.

299 Zervos M, Nelson M. Cefepime versus ceftriaxone forempiric treatment of hospitalized patients withcommunity-acquired pneumonia. The Cefepime StudyGroup. Antimicrob Agents Chemother 1998;42:729–33.

300 Daniel R. Trovafloxacin vs high dose amoxicillin in thetreatment of community-acquired bacterial pneumonia.Drugs 1999;58(Suppl 2):304–5.

301 Daniel R. Oral trovafloxacin compared with amoxicillin(plus optional erythromycin) for the treatment of mild tomoderate community-acquired pneumonia. Drugs 1999;58(Suppl 2):320–2.

302 Grossman RF, Campbell DA, Landis SJ, et al. Treatmentof community-acquired pneumonia in the elderly: the roleof cefepime, a fourth-generation cephalosporin. J Antimi-crob Chemother 1999;43:549–54.

303 Moola S, Hagberg L, Churchyard GA, et al. A multicenterstudy of grepafloxacin and clarithromycin in the treatmentof patients with community-acquired pneumonia. Chest1999;116:974–83.

304 Peterson J. Oral trovafloxacin vs oral cefuroxime +/– eryth-romycin in community-acquired pneumonia. Drugs 1999;58(Suppl 2):301–3.

305 Ramirez J, Unowsky J, Talbot GH, et al. Sparfloxacin ver-sus clarithromycin in the treatment of community-acquiredpneumonia. Clin Ther 1999;21:103–17.

306 Fogarty C, Grossman C, Williams J, et al. EYcacy andsafety of moxifloxacin vs clarithromycin for community-acquired pneumonia. Infections in Medicine 1999;748–63.

307 Blaser MJ, Klaus BD, Jacobson JA, et al. Comparison ofcefadroxil and cephalexin in the treatment of community-acquired pneumonia. Antimicrob Agents Chemother 1983;24:163–7.

308 Cassell GH, Drnec J, Waites KB, et al. EYcacy ofclarithromycin against Mycoplasma pneumoniae. J Antimi-crob Chemother 1991;27(Suppl A):47–59.

309 Morales JO, Snead H. EYcacy and safety of intravenouscefotaxime for treating pneumonia in outpatients. Am JMed 1994;97:28–33.

310 Ramirez JA. Switch therapy in community-acquired pneu-monia. Diagn Microbiol Infect Dis 1995;22:219–23.

311 Higuera F, Hidalgo H, Feris J, et al. Comparison of oralcefuroxime axetil and oral amoxycillin/clavulanate in thetreatment of community-acquired pneumonia. J AntimicrobChemother 1996;37:555–64.

312 Berman SJ, Sieger B, Geckler W, et al. A comparative studyof meropenem and ceftazidime in the treatment of patientshospitalized with community-acquired pneumonia. CurrTher Res 1997;58:903–16.

313 Ailani RK, Agastya G, Ailani RK, et al. Doxycycline is acost-eVective therapy for hospitalized patients withcommunity-acquired pneumonia. Arch Intern Med 1999;159:266–70.

314 Schonwald S, Kuzman I, Oreskovic K, et al. Azithromycin:single 1.5 g dose in the treatment of patients with atypicalpneumonia syndrome: a randomized study. Infection 1999;27:198–202.

315 Sherer Y, Bakshi E, Rotman P, et al. Comparative clinicalstudy of cefonicid, chloramphenicol, and penicillin incommunity-acquired pneumonia. Int J Mol Med 1998;2:343–8.

316 Stocks JM, Wallace RJ Jr, GriYth DE, et al. Ofloxacin incommunity-acquired lower respiratory infections. A com-parison with amoxicillin or erythromycin. Am J Med 1989;87:52–6S.

317 Morandini G, Perduca M, Zannini G, et al. ClinicaleYcacy of azithromycin in lower respiratory tract infec-tions. J Chemother 1993;5:32–6.

318 Finch RG, Woodhead MA. Practical considerations andguidelines for the management of community-acquiredpneumonia. Drugs 1998;55:31–45.

319 Moreno S, Garcia-Leoni ME, Cercenado E, et al.Infections caused by erythromycin-resistant Streptococcuspneumoniae: incidence, risk factors, and response totherapy in a prospective study. Clin Infect Dis 1995;20:1195–200.

320 Jackson MA, Burry VF, Olson LC, et al. Breakthrough sep-sis in macrolide-resistant pneumococcal infection. PediatrInfect Dis J 1996;15:1049–51.

321 Amsden GW. Pneumococcal macrolide resistanc: myth orreality? J Antimicrob Chemother 1999;44:1–6.

322 Perez-Trallero E. Pneumococcal macrolide resistance: nota myth. J Antimicrob Chemother 2000;45:401–3.

323 Carbon C, Poole MD. The role of newer macrolides in thetreatment of community-acquired respiratory tract infec-tion. A review of experimental and clinical data. JChemother 1999;11:107–18.

324 Moss PJ, Finch RG. The next generation: fluoroquinolo-nes in the management of acute lower respiratory infectionin adults. Thorax 2000;55:83–5.

325 Sportel JH, Koeter GH, van Altena R, et al. Relationbetween beta-lactamase producing bacteria and patientcharacteristics in chronic obstructive pulmonary disease(COPD). Thorax 1995;50:249–53.

326 Meehan TP, Fine MJ, Krumholz HM, et al. Quality ofcare, process, and outcomes in elderly patients with pneu-monia. JAMA 1997;278:2080–4.

327 Simpson JC, Macfarlane JT, Watson J, et al. A nationalconfidential enquiry into community acquired pneumoniadeaths in young adults in England and Wales. Thorax 2000;55:1040–5.

328 Schulin T, Wennersten CB, Ferraro MJ, et al. Susceptibili-ties of Legionella spp. to newer antimicrobials in vitro.Antimicrob Agents Chemother 1998;42:1520–3.

329 Burgess DS, Lewis II JS. EVect of macrolides as part ofinitial empiric therapy on medical outcomes for hospital-ised patients with community acquired pneumonia. ClinTher 2000;22:872–8.

330 Edelstein PH. Antimicrobial chemotherapy for Legion-naires disease: time for a change. Ann Intern Med1998;129:328–30.

331 Wallace RJ Jr, Brown BA, GriYth DE, et al. Reducedserum levels of clarithromycin in patients treated withmultidrug regimens including rifampin or rifabutin forMycobacterium avium-M intracellulare infection. J InfectDis 1995;171:747–50.

332 Plemmons RM, McAllister CK, Garces MC, et al.Osteomyelitis due to Mycobacterium haemophilum in acardiac transplant patient: case report and analysis of inter-actions among clarithromycin, rifampin, and cyclosporine.Clin Infect Dis 1997;24:995–7.

333 Leophonte P, Veyssier P. Levofloxacin in the treatment ofcommunity-acquired pneumococcal pneumonia. PresseMed 1999;28:1975–9.

334 Norrby SR, Petermann W, Willcox PA, et al. A comparativestudy of levofloxacin and ceftriaxone in the treatment ofhospitalized patients with pneumonia. Scand J Infect Dis1998;30:397–404.

335 MacGregor RR, Graziani AL. Oral administration ofantibiotics: a rational alternative to the parenteral route.Clin Infect Dis 1997;24:457–67.

336 Mandell LA, Bergeron MG, Gribble MJ, et al. Sequentialantibiotic therapy: eVective cost management and patientcare. Can J Infect Dis 1995;6:306–15.

337 Siegel RE, Alicea M, Lee A, et al. Comparison of 7 versus10 days of antibiotic therapy for hospitalized patients withuncomplicated community-acquired pneumonia: a pro-spective, randomized, double-blind study. Am J Ther 1999;6:217–22.

338 Nathwani D. Sequential switch therapy for lower respira-tory tract infections: a European perspective. Chest1998;113:211–8S.

339 Kinasewitz G, Wood RG. Azithromycin versus cefaclor inthe treatment of acute bacterial pneumonia. Eur J ClinMicrobiol Infect Dis 1991;10:872–7.

340 HeVner JE, McDonald J, Barbieri C, et al. Management ofparapneumonic eVusions. An analysis of physician practicepatterns. Arch Surg 1995;130:433–8.

341 Light RW. A new classification of parapneumonic eVusionsand empyema. Chest 1995;108:299–301.

342 HeVner JE, Brown LK, Barbieri C, et al. Pleural fluidchemical analysis in parapneumonic eVusions. A meta-analysis. Am J Respir Crit Care Med 1995;151:1700–8; pub-lished erraturm 1995;152:823.

343 Nguyen-van-Tam JS, Nicholson KG. Influenza deaths inLeicestershire during the 1989–90 epidemic: implicationsfor prevention. Epidemiol Infect 1992;108:537–45.

344 Ahmed AE, Nicholson KG, Nguyen-van-Tam JS. Reduc-tion in mortality associated with influenza vaccine during1989–90 epidemic. Lancet 1995;346:591–5.

345 Ashley J, Smith T, Dunnell K. Deaths in Great Britainassociated with the influenza epidenic of 1989–90. Popula-tion Trends 1991;65:16–20.

346 Nichol KL, Margolis KL, Wuorenma J, et al. The eYcacyand cost eVectiveness of vaccination against influenzaamong elderly persons living in the community. N Engl JMed 1994;331:778–84.

347 Centers for Disease Control. Influenza vaccination levels.JAMA 1992;268:2360.


www.thoraxjnl.com



348 Nguyen-van-Tam JS, Nicholson KG. Influenzaimmunization; vaccine oVer, request and uptake inhigh-risk patients during the 1991/2 season. Epidemiol Infect1993;111:347–55.

349 Govaert ThME, Dinant GJ, Aretz K, et al. Adversereactions to influenza vaccine in elderly people: ran-domised double blind placebo controlled trial. BMJ 1993;307:988–90.

350 Fedson DS, Wajda A, Nicol JP, et al. Clinical eVectivenessof influenza vaccination in Manitoba. JAMA 1993;270:1956–61.

351 Foster DA, Talsma A, Furumoto-Dawson A, et al.Influenza vaccine eVectiveness in preventing hospitaliza-tion for pneumonia in the elderly. Am J Epidemiol1992;136:296–307.

352 Nichol KL. Complications of influenza and benefits ofvaccination. Vaccine 1999;17: S47–52.

353 Nichol KL, Baken L, Nelson A. Relation betweeninfluenza vaccination and outpatient visits, hospitalization,and mortality in elderly persons with chronic lung disease.Ann Intern Med 1999;130:397–403.

354 Govaert TME, Thijs CTMC, Masurel N, et al. TheeYcacy of influenza vaccination in elderly individuals: arandomized double-blind placebo-controlled trial. JAMA1994;272:1661–5.

355 Gross PA, Hermogenes AW, Sacks HS, et al. The eYcacyof influenza vaccine in elderly persons. A meta-analysis andreview of the literature. Ann Intern Med 1995;123:518–27.

356 Palache AM, Beyer WEP, Luchters G, et al. Influenzavaccines: the eVect of vaccine dose on antibody response inprimed populations during the ongoing interpandemicperiod. A review of the literature. Vaccine 1993;11:892–908.

357 Beyer WEP, De Bruijn IA, Palache AM, et al. Protectionagainst influenza after annually repeated vaccination: ameta-analysis of serologic and field studies. Arch Intern Med1999;159:182–8.

358 Weingarten S, StaniloV H, Ault M, et al. Do hospitalemployees benefit from the influenza vaccine? A placebo-controlled clinical trial. J Gen Intern Med 1988;3:32–7.

359 Wilde JA, McMillan JA, Serwint J, et al. EVectiveness ofinfluenza vaccine in health care professionals: a rand-omized trial. JAMA 1999;281:908–13.

360 Treanor JJ, Mattison HR, Dumyati G, et al. Protective eY-cacy of combined live intranasal and inactivated influenzaA virus vaccines in the elderly. Ann Intern Med 1992;117:625–33.

361 Dillard JP, Vandersea MW, Yother J. Characterisation ofthe cassette containing genes for type 3 capsular polysac-charides. Microb Drug Resist 1997;3:73–88.

362 Sniadack DH, Schwartz B, Lipman H, et al. Potentialinterventions for the prevention of childhood pneumonia:geographic and temporal diVerences in serotype and sero-group distribution of sterile site pneumococcal isolatesfrom children. Implications for vaccine strategies. PediatrInfect Dis J 1995;14:503–10.

363 George RC. The epidemiology of pneumococcal disease.London: Royal Society of Medicine, 1995: 1–7.

364 Honkanen PO, Keistinen T, Kivela S L. Reactions follow-ing administration of influenza vaccine alone or with pneu-mococcal vaccine to the elderly. Arch Intern Med 1996;156:205–8.

365 Hedlund JU, Kalin ME, Ortqvist AB, et al. Antibodyresponse to pneumococcal vaccine in middle-aged and eld-erly patients recently treated for pneumonia. Arch InternMed 1994;154:1961–5.

366 Davidson M, Bulkow LR, Grabman J, et al. Immunogenic-ity of pneumococcal revaccination in patients with chronicdisease. Arch Intern Med 1994;154:2209–14.

367 Shelly MA, Jacoby H, Riley GJ, et al. Comparison of pneu-mococcal polysaccharide and CRM197-conjugated pneu-mococcal oligosaccharide vaccines in young and elderlyadults. Infect Immun 1997;65:242–7.

368 Fine MJ, Smith MA, Carson CA, et al. EYcacy ofpneumococcal vaccination in adults: a meta-analysis ofrandomized controlled trials. Arch Intern Med 1994;154:2666–77.

369 Nichol KL, Baken L, Wuorenma J, Nelson A. The healthand economic benefits associated with pneumococcal vac-cination of elderly persons with chronic lung disease. ArchIntern Med 1999;159:2437–42.

370 Melbye H, Berdal BP, Straume B, et al. Pneumonia: aclinical or radiographic diagnosis? Etiology and clinicalfeatures of lower respiratory tract infection in adults in gen-eral practice. Scand J Infect Dis 1992;24:647–55.

371 Berntsson E, Lagergard T, Strannegard O, et al. Etiology ofcommunity-acquired pneumonia in out-patients. Eur J ClinMicrobiol 1986;5:446–7.

372 Michetti G, Pugliese C, Bamberga M, et al. Community-acquired pneumonia: is there diVerence in etiologybetween hospitalized and out-patients? Minerva Medica1995;86:341–51.

373 Blanquer J, Blanquer R, Borras R, et al. Aetiology of com-munity acquired pneumonia in Valencia, Spain: a multi-centre prospective study. Thorax 1991;46:508–11.

374 Almirall J, Bolibar I, Vidal J, et al. Epidemiology ofcommunity-acquired pneumonia in adults: a population-based study. Eur Respir J 2000;15:757–63.

375 Berntsson E, Blomberg J, Lagergard T, et al. Etiology ofcommunity-acquired pneumonia in patients requiring hos-pitalization. Eur J Clin Microbiol 1985;4:268–72.

376 Levy M, Dromer F, Brion N, et al. Community-acquiredpneumonia. Importance of initial noninvasive bacteriologicand radiographic investigations. Chest 1988;93:43–8.

377 Burman LA, Trollfors B, Andersson B, et al. Diagnosis ofpneumonia by cultures, bacterial and viral antigendetection tests, and serology with special reference to anti-bodies against pneumococcal antigens. J Infect Dis1991;163:1087–93.

378 Hone R, Haugh C, O’Connor B, et al. Legionella: an infre-quent cause of adult community acquired pneumonia inDublin. Ir J Med Sci 1989;158:230–2.

379 Ruf B, Schurmann D, Horbach I, et al. Incidence andclinical features of community-acquired legionellosis inhospitalized patients. Eur Respir J 1989;2:257–62.

380 Pareja A, Bernal C, Leyva A, et al. Etiologic study ofpatients with community-acquired pneumonia. Chest 1992;101:1207–210.

381 Holmberg H. Aetiology of community-acquired pneumo-nia in hospital treated patients. Scand J Infect Dis 1987;19:491–501.

382 Aubertin J, Dabis F, Fleurette J, et al. Prevalence oflegionellosis among adults: a study of community-acquiredpneumonia in France. Infection 1987;15:328–31.

383 Blasi F, Cosentini R, Raccanelli R, et al. Emergingpathogens of community-acquired pneumonia: a two-yearprospective study. J Chemother 1995;7(Suppl 4):115–6.

384 Socan M, Marinic-Fiser N, Kraigher A, et al. Microbialaetiology of community-acquired pneumonia in hospital-ised patients. Eur J Clin Microbiol Infect Dis 1999;18:777–82.

385 Lorente ML, Falguera M, Nogues A, et al. Diagnosis ofpneumococcal pneumonia by polymerase chain reaction(PCR) in whole blood: a prospective clinical study. Thorax2000;55:133–7.

386 Mundy LM, Auwaerter PG, Oldach D, et al. Community-acquired pneumonia: impact of immune status. Am J RespirCrit Care Med 1995;152:1309–15.

387 Bates JH, Campbell GD, Barron AL, et al. Microbial etiol-ogy of acute pneumonia in hospitalized patients. Chest1992;101:1005–12.

388 Olaechea PM, Quintana JM, Gallardo MS, et al. A predic-tive model for the treatment approach to community-acquired pneumonia in patients needing intensive care unitadmission. Intensive Care Med 1996;22:1294–300.

389 Ortqvist A, Sterner G, Nilsson JA. Severe community-acquired pneumonia: factors influencing need of intensivecare treatment and prognosis. Scand J Infect Dis 1985;17:377–86.

390 Rello J, Quintana E, Ausina V, et al. A three-year study ofsevere community-acquired pneumonia with emphasis onoutcome. Chest 1993;103:232–5.

391 Almirall J, Mesalles E, Klamburg J, et al. Prognostic factorsof pneumonia requiring admission to the intensive careunit. Chest 1995;107:511–6.

392 Sorensen J, Forsberg P, Hakanson E, et al. A new diagnos-tic approach to the patient with severe pneumonia. Scand JInfect Dis 1989;21:33–41.

393 Baranda MM, Carranceja JC, Errasti CA. Q fever in theBasque Country: 1981–1984. Rev Infect Dis 1985;7:700–1.

394 Marrie TJ, Haldane EV, Faulkner RS, et al. Theimportance of Coxiella burnetii as a cause of pneumonia inNova Scotia. Can J Public Health 1985;76:233–6.

395 Feldman C, Ross S, Mahomed AG, et al. The aetiology ofsevere community-acquired pneumonia and its impact oninitial, empiric, antimicrobial chemotherapy. Respir Med1995;89:187–92.

396 Potgieter PD, Hammond JM. Etiology and diagnosis ofpneumonia requiring Intensive Care Unit admission. Chest1992;101:199–203.

397 Thompson JE. Community acquired pneumonia in northeastern Australia: a hospital based study of Aboriginal andnon-Aboriginal patients. Aust NZ J Med 1997;27:59–61.

398 Boonsawat W, Boonma P, Tangdajahiran T, et al.Community-acquired pneumonia in adults at SrinagarindHospital. J Med Assoc Thai 1990;73:345–52.

399 Lee KH, Hui KP, Tan WC, et al. Severe community-acquired pneumonia in Singapore. Singapore Med J1996;37:374–7.


www.thoraxjnl.com



Documents

Guia de Manejo de Nac en Adultos 65 Pags