Acute bacterial conjunctivitis.pdf

8/11/2019 Acute bacterial conjunctivitis.pdf

1/13

IntroductionBacterial conjunctivitis is usuallydivided according to its course andseverity into hyperacute, acute andchronic forms. Neisseria gonorrhoeaeis the most frequent cause of hyper-acute bacterial conjunctivitis, which isthen usually regarded as an oculogeni-tal disease, occurring in neonates andin sexually active adults. Hyperacutebacterial conjunctivitis is character-

ized by abrupt onset, profuse, thick,yellow-green purulent secretion, mixedocular injection and chemosis andsometimes the formation of an inam-matory membrane (Rubenstein 1999;Mannis & Plotnik 2005). If causedby N. gonorrhoeae , painful cornealinvolvement rapidly develops, andcorneal opacities and even cornealperforation and endophthalmitis mayoccur if the infection is not rapidlyrecognized and treated.

In chronic bacterial conjunctivitis,

ocular symptoms and signs last for atleast 4 weeks and relapses occur fre-quently. Conjunctival hyperaemia anddischarge are usually moderate ormild. Coagulase-positive and -negativestaphylococci are the organisms foundmost frequently. Exotoxins producedby staphylococci may cause punctateepithelial keratitis and marginal kera-titis. Moraxella lacunata is the speciesmost commonly found in chronicangular blepharoconjunctivitis (Ru-benstein 1999). Although such chronic

conjunctival bacterial infection doesoccur, other causes of the ocular com-plaints such as blepharitis, meibom-itis, acne rosacea, ocular allergy,dacryocystitis, ectropium, entropium,trichiasis and dry-eye disease shouldalways be excluded.

Worldwide, acute bacterial conjunc-tivitis represents the vast majority of all cases of bacterial conjunctivitis,and the rest of this article deals exclu-sively with this form of bacterial conjunctivitis.

EpidemiologyAcute conjunctivitis is commonlydened as conjunctivitis with symp-toms of less than 34 weeks duration,and is the eye disorder most com-monly seen by a general practitioner(GP) (Dart 1986; McDonnell 1988).According to the diagnoses made bythe GPs, the majority of cases arecaused by infections, followed byallergic conditions. Some seasonalvariations have been described; bacte-rial conjunctivitis shows a peak occur-rence during DecemberApril and

R eview Article

Acute bacterial conjunctivitisGunnar Hvding 1,2

1Department of Clinical Medicine, Section of Ophthalmology, University of Bergen,Bergen, Norway2Haukeland University Hospital, Bergen, Norway

ABSTRACT.

Acute bacterial conjunctivitis is the eye disease most commonly seen by gen-eral practitioners, and is estimated to represent approximately 1% of all con-sultations in primary care. This article gives a review of the epidemiology,aetiology, clinical picture, complications, differential diagnoses, in vitro exam-inations and therapy of acute bacterial conjunctivitis. Until now, topical anti-bacterial therapy has generally been preferred by both physicians and patientsbecause this will usually shorten the course of the disease slightly and allowthe early readmittance of children to their kindergarten or school. Recently,several reports from primary care have conrmed the well-known clinical expe-rience that the disorder has an excellent prognosis with a high frequency of spontaneous remission. In accordance, an expectant attitude or delayed pre-scription policy are now frequently strongly recommended. However, these

reports also emphasize the difculty in making a correct clinical distinctionbetween bacterial and viral conjunctivitis. The effect of a general non-prescrip-tion attitude on transmission rates of pathogens also remains to be claried.This must be born in mind when deciding how these patients should be han-dled. The socioeconomic and medical pros and cons of different treatment poli-cies are discussed, and a highly personal view on the optimal handling strategyfor these patients is also presented.

Key words: acute bacterial conjunctivitis aetiology clinical picture complications differen-tial diagnoses epidemiology in vitro examinations treatment

Acta Ophthalmol. 2008: 86: 517 2007 The Author

Journal compilation

2007 Acta Ophthalmol Scand

doi: 10.1111/j.1600-0420.2007.01006.x

Acta Ophthalmologica 2008

5


2/13

viral conjunctivitis peaks in the sum-mer, while allergic conjunctivitis isseen more frequently during the springand summer months (Gigliotti et al.1981; Schmitz et al. 1983; Fitch et al.1989; Block et al. 2000; Aoki & Taga-wa 2002). Worldwide, there are anestimated 5 million cases of neonatalinfectious conjunctivitis per year (Wil-helmus 2005). In the developed world,acute red eyes account for 14% of all GP consultations, and are mostfrequently diagnosed as acute bacte-rial conjunctivitis (Dart 1986; McDon-nell 1988; Sheikh & Hurwitz 2001;Everitt & Little 2002; Rietveld et al.2005). In Norway, it has been sug-gested that acute infective conjunctivi-tis is suspected in approximately 30

out of 1000 patients in a general medi-cal practice, and that this diagnosisappears to be correct in approximatelytwo out of three of these cases (Hvd-ing et al. 1991). In the UK, the pro-portion of patients who consulted forconjunctivitis increased from 284 per10 000 in 19811982 to 395 per 10 000in 19911992 (McCormick et al.1995). Acute infective conjunctivitisrepresents up to 1% of GP consulta-tions in the UK. Each year, one ineight children have signs and symp-

toms of acute conjunctivitis, andamong small children as much as 18%consulted their GP at least once ayear because of acute conjunctivitis(Dart 1986; McDonnell 1988; McCor-mick et al. 1995). However, it shouldalso be noted that reports by Dart(1986) and Buckley (1990) stronglyindicate that GPs tend to over-diag-nose microbial conjunctivitis.

Acute infective conjunctivitis is seenmost frequently in infants, school chil-dren and the elderly. It is usually theresult of bacterial or viral infections,the latter most often caused by adeno-viruses, Herpes simplex virus or picor-naviruses (Rubenstein 1999). Severalreports indicate that bacteria areresponsible for about 5075% of allcases of acute conjunctivitis in youngchildren (Gigliotti et al. 1981; Blocket al. 2000; Buznach et al. 2005). Inaccordance with this, a study by Roseet al. (2005) including 326 childrenwith a clinical diagnosis of conjuncti-vitis yielded bacterial pathogens in67% of the patients, viruses alonein 3% and both bacteria and virusesin 10%. On the other hand, an adeno-virus was isolated from 11 of 17

paediatric patients (65%) who simul-taneously had acute conjunctivitisand pharyngitis (Gigliotti et al. 1981).Furthermore, in 16 of 45 patients(36%) presenting to an ocular emer-gency room with symptoms and signsof acute conjunctivitis a viral aetiol-ogy was found (adenovirus in 15patients and H. simplex virus type I inone patient), while a bacterial aetiol-ogy was diagnosed in 18 cases (40%)and no cases of chlamydial conjuncti-vitis were found (Fitch et al. 1989).The difculty in making a correct clin-ical distinction between bacterial andviral conjunctivitis is illustrated by anarticle based on a questionnaire sentto 300 GPs in the UK (Everitt & Lit-tle 2002). The answers from the 236

responders (78%) showed that 92%felt condent or very condent of making the diagnosis of acute infec-tive conjunctivitis, but only 36% feltable to discriminate between acutebacterial and viral conjunctivitis. Asmuch as 95% of the responders usu-ally prescribed topical antibiotics topatients with suspected acute infectiveconjunctivitis and 87% used chloram-phenicol as their drug of choice, while13% primarily used fusidic acid.Features believed to increase the prob-

ability of a bacterial infection werehistory of a cold (85%) and the typeof discharge (87%). Although obvio-usly rare compared to bacterial andviral infections, chlamydial conjuncti-vitis is still an important differentialdiagnosis, particularly in newborns(Solberg et al. 1991; Dannevig et al.1992; Rubenstein 1999). In some com-munities, Chlamydia trachomatis hasbeen isolated from 4673% of neo-nates with purulent conjunctivitis,probably reecting a high prevalenceof chlamydial cervicitis in the localpopulation (Rapoza et al. 1986).

Based on a Medline literaturesearch, Richardson et al. (2001)reported only a low or moderate riskof transmission in infectious conjunc-tivitis, and they found no general rec-ommendation for exclusion frompreschool or school. However, theconclusions drawn were poorly evi-dence-based; infectivity and need forexclusion, etc. will obviously be differ-ent in adenoviral and bacteriologicalinfections. In epidemic keratoconjunc-tivitis caused by adenovirus type 8and type 19, the risk of infecting closesocial contacts has been reported to

be 1035% (McMinn et al. 1991; Wil-helmus 2005). There are also numer-ous reports on outbreaks of acutebacterial conjunctivitis in day-carecentres, boarding schools, militarycamps, nursing homes, intensive careunits, etc., but exact and evidence-based knowledge about the infectivityand community spread of acute bacte-rial conjunctivitis is generally lacking.A study including 20 children withacute conjunctivitis from ve differentday-care centres yielded Haemophilusinuenzae from eight eyes and Morax-ella from two eyes (Trottier et al.1991). Different H. inuenzae strainsoccurred within the same centre, butthe same strain of H. inuenzae wasalways found in both the eye and

nasopharynx. Thus, the eyes wereprobably infected from the nasophar-ynx. On the other hand, the authorssuggest that a viral infection spreadingfrom child to child may impair theresistance of the eye to bacterial colo-nization and thus permit bacterialsuperinfection. Reporting from sevennursing homes, Garibaldi et al. (1981)noted high prevalence and clusteringof infections such as upper respiratorytract infections, diarrhoea, conjuncti-vitis and infected decubital ulcers.

High patient-to-staff ratios, highemployee turnover, non-professionalpersonnel and increased susceptibilityof patients were identied as risk fac-tors for the spreading of infections.Patients in nursing homes for theelderly and long-term-care facilitieshave been reported to be highly sus-ceptible to conjunctivitis caused byStreptococcus pyogenes (Ruben et al.1984) and to methicillin-resistant Sta- phylococcus aureus (Brennen & Muder1990). According to King et al.(1988), conjunctivitis accounted for5% of nosocomial infections in a largeuniversity-afliated paediatric hospitalbetween January 1984 and April 1986.Pseudomonas aeruginosa was recov-ered from the conjunctiva of 30patients with chronic and debilitatingprimary disease. Compared to thetotal number of patients admitted tothe hospital, children under the age of 18 months were signicantly over-rep-resented. Seventy per cent of the casesoccurred in the intensive care unitpatients, and P. aeruginosa was recov-ered from 70% of patients who hadantecedent nasopharyngeal endotra-cheal cultures obtained. All patients


6


3/13

had received antibiotic treatment dur-ing hospitalization, and a need forrespiratory care characterized the pop-ulation. A cluster of 10 severe cases of nosocomial bacterial keratitis in threeintensive care units showed almostidentical bacterial growth in culturesfrom eye and sputum (Hilton et al.1983). All the patients had requiredfrequent tracheal suctioning. Only theleft eye was involved in nine of the10 cases; the ocular infections wereascribed to right-handed nurses habitof withdrawing the catheter diagonallyacross the patients face.

Studies on epidemic follicular conjunctivitis occurring in female teen-agers have shown a signicantlyincreased incidence of conjunctival

cultures yielding Moraxella lacunata(Ringvold et al. 1985; Schwartz et al.1989), implicating the use and sharingof mascara as a possible cause of theinfective conjunctivitis. During therst 6 months of 1980, large out-breaks of conjunctivitis were reportedfrom seven colleges and universities inNew York State; conjunctival swabsyielded non-typable Streptococcus pneumoniae alone or in combinationwith other organisms in 46% of thecases (Shayegani et al. 1982). Very

similar streptococcus strains were iso-lated during outbreaks of conjunctivi-tis in other US states.

Despite the lack of strictly evidence-based knowledge, conjunctival infec-tion by direct contact is generallybelieved to be particularly common inkindergartens and among school chil-dren and institutionalized patients. Inbacterial conjunctivitis, the suggestedperiods of incubation and communica-bility are 17 days and 27 days,respectively, while the correspondinggures for adenovirus conjunctivitisare 512 days and 1014 days (Wilhel-mus 2005).

AetiologyBacteria may frequently be isolatedfrom the conjunctiva of healthy subjects. Generally speaking, the normalconjunctival ora represents both bac-terial colonization and transient orrecurring bacterial contamination.Colonization implies a stable presenceof the micro-organism, balanced by thehost defence mechanisms, while con-tamination represents micro-organisms

introduced from sources outside theconjunctiva (Kowalski & Roat 2005).The majority of micro-organismspresent on the healthy conjunctivarepresent a continuous or recurringcontamination, while true colonizationis much more common on the eyelidmargins. Given the right conditions,any micro-organism can cause aninfection. The ocular pathogenicity of a micro-organism can be calculated bythe number of manifest ocular infec-tions divided by the number of eyesharbouring the same micro-organism(Wilhelmus 2005). A primary patho-gen will regularly cause infection,an opportunistic pathogen causesinfection in immunocompromizedindividuals, while normally occurring

micro-organisms may act as incidentalpathogens, replicating and causingdisease when host defence mecha-nisms have been impaired (Wilhelmus2005).

Coagulase-negative staphylococciand corynebacteria are frequentlypresent on the healthy conjunctiva,but more traditionally pathogenicorganisms, such as coagulase-positivestaphylococci, streptococci, haemophi-lus species, moraxellae and gram-negative coliform rods, are also

occasionally isolated from normal,non-inamed eyes (Locatcher-Khorazo & Seegal 1972; Fahmy et al.1975; Cagle & Abshire 1981; Hvding1981; Seibel & Ruprecht 1983; Olafsenet al. 1986; Weiss et al. 1993; Thiel &Schumacher 1994; Mannis & Plotnik2005). A greater number of bacterialspecies is generally recovered fromadults than from children. Facultativeanaerobic species, particularly coryne-bacteria, are found signicantly morefrequently in adults, while streptococciare obtained more frequently fromchildren (Singer et al. 1988). A similarbut more stable and abundant bacte-rial ora can usually be isolated fromthe eyelid margins of healthy subjects(Cagle & Abshire 1981). In such non-inamed eyes, an increased frequencyof coagulase-negative staphylococcihas been found on the conjunctiva of patients with diabetes mellitus com-pared to a non-diabetic control group(Martins et al. 2004), while a signi-cantly increased occurrence of S. aur-eus has been reported in patients withatopic dermatitis (Inoue 2002). Miller& Ellis (1977) found that thefrequency of positive conjunctival

bacterial cultures and the amount of lysozyme present in the tear uid werenot inuenced by the use of immuno-suppressive drugs, but there was apositive correlation between the num-ber of different bacterial species andan increasing dose of prednisolone.Franklin et al. (1977) and Friedlaenderet al. (1980) reported that immunode-cient patients had a higher incidenceof conjunctival and lid infections,while the microbiological ora wassimilar to that found in immunocom-petent patients. In accordance withthese reports, Gritz et al. (1997) andYamauchi et al. (2005) found a similarbacterial ora in AIDS patients, HIV-positive and HIV-negative patients.

The amount of bacteria present on

the healthy conjunctiva is usuallyrather small, commonly yielding lessthan 10 colonies on culture, whilecases of acute bacterial conjunctivitisoften yield a more conuent growth(Kowalski & Roat 2005). In additionto the sweeping and ushing action of eyelids and tear uid and the barrierfunction of the conjunctival and cor-neal epithelium, lysozyme, lactoferrin,beta-lysin, IgA and other immuno-globulins and complement systemcomponents in the lacrimal uid help

to prevent the normal bacterial orafrom causing conjunctivitis or keratitis(McClellan 1997; Kowalski & Roat2005; Mannis & Plotnik 2005). Inaddition, metabolic products made bythe normal conjunctival bacterial oraprobably also inhibit colonizationand multiplication of more virulentbacterial species (McClellan 1997).Mucosa-associated lymphoid tissue(MALT) is an important part of ourgeneral defence against pathogens.Eye-associated lymphoid tissue(EALT) appears to be a componentof this common mucosal immune sys-tem, but the exact mechanisms bywhich ocular mucosal immunologyacts to prevent eye infection are notfully understood (McClellan 1997; Jett2005).

Acute bacterial conjunctivitis is mostfrequently caused by S. aureus , Staphy-lococcus epidermidis , H. inuenzae ,Streptococcus pneumoniae , Streptococ-cus viridans, Moraxella catarrhalis andGram-negative intestinal bacteria (Sei-bel & Ruprecht 1983; Hrven 1993;Weiss et al. 1993; Block et al. 2000;Wald et al. 2001; Normann et al.2002; Buznach et al. 2005; Rose et al.


7


4/13

2005; Tarabishy et al. 2006). World-wide, both the frequency and the caus-ative micro-organisms vary accordingto climatic, social and hygienic condi-tions. Patient age is also an importantfactor; different micro-organisms pre-dominate the bacterial spectrum in thedifferent phases of life. Thus, in youngchildren H. inuenzae is the mostcommonly isolated bacteria in acuteconjunctivitis, followed by S. pneumo-niae and M. catarrhalis . In this agegroup, H. inuenzae conjunctivitismay often occur alongside recurringotitis media and respiratory infections(Block et al. 2000; Wald et al. 2001;Buznach et al. 2005; Rose et al. 2005).Conjunctivitis caused by S. aureus ismost frequently seen in neonates and

in older children, as well as in adultsand the elderly (Dannevig et al. 1992;Normann et al. 2002; Orden Martinezet al. 2004; Mannis & Plotnik 2005;Rietveld et al. 2005). Worldwide,S. aureus is the most frequently occur-ring cause of bacterial conjunctivitis(Mannis & Plotnik 2005).

Even in patients with so-called typi-cal clinical signs and symptoms of acute bacterial conjunctivitis, moststudies show a surprisingly high fre-quency (2268%) of negative bacterio-

logical cultures (Mahajan 1983;Olafsen et al. 1986; Hrven 1993;Block et al. 2000; Rietveld et al. 2004;Buznach et al. 2005; Kernt et al. 2005;Rose et al. 2005). The varying fre-quency of positive bacteriology indifferent studies may at least partly beexplained by different interpretationsof the culture results; some authorsreport only the frequency of tradition-ally pathogenic bacteria, while othersinclude all bacteria isolated and a fewmake a quantitative interpretation of the bacteriological ndings. Careshould also be taken to avoid contactwith the eyelid margins during conjunctival swabbing, because the eyelidscommonly harbour staphylococci,corynebacteria and alpha-haemolyticstreptococci (Weiss et al. 1993).

Contact with contaminated ngers isbelieved to be a common cause of acuteinfective conjunctivitis, but bacteriamay also reach the conjunctiva fromthe eyelid margins and adjacent skin,from the nasopharynx via the nasolacr-imal duct, from infected eye drops orcontact lenses, and more rarely fromthe genitals or via the bloodstream.Factors predisposing for infection

include ectropium and entropium, anobstructed nasolacrimal duct, abnor-mal lacrimal uid and injured conjunc-tival epithelium following trauma,dry-eye disease or previous infection(Hvding 2004; Mannis & Plotnik2005). Immunodeciency syndromesand systemic immunosuppression alsopredispose for acute bacterial conjunc-tivitis (Franklin et al. 1977; Friedlaend-er et al. 1980; Sharma et al. 2004). Invaginally delivered newborns, the pri-mary source of infective organisms isusually believed to be the birth canal of the mother (Isenberg et al. 1988; Nor-mann 2005), but this has been con-tradicted by Krohn et al. (1993), whosuggested that bacteria causing acuteconjunctivitis more commonly origi-

nate from the infants nasopharynx orfrom the infants care providers.

Clinical pictureAccording to most textbooks, acutebacterial conjunctivitis begins unilater-ally with a foreign body sensation,increased secretion and moderateconjunctival hyperaemia (Rubenstein1999; Mannis & Plotnik 2005). Thesecretion soon becomes mucopurulent

or purulent. Sticky eyelids in themorning is recognized as a very com-mon symptom, even in cases withouta distinctly abnormal secretion. Dis-tinct or severe pain is rare, althoughsome patients experience a strongburning or stinging sensation. Within12 days the fellow eye usuallybecomes involved. Although cliniciansgenerally agree that these symptomsand signs may be regarded as typicalfor acute bacterial conjunctivitis, anextensive literature search conductedby Rietveld et al. (2003) failed toprove that they can be classied asevidence-based diagnostic criteria forthe distinction between bacterial andviral conjunctivitis.

Conjunctival injection is the rule,i.e. the redness is most intense on theinside of the eyelids and peripherallyon the bulbar conjunctiva, while theeye is paler towards the corneosclerallimbal area (Fig. 1). In more severecases, the eye shows a mixed injection,i.e. the whole conjunctiva is red, andsmall haemorrhages and chemosismay also occur. Purulent discharge inthe palpebral opening (Fig. 2), or per-haps only in the inferior fornix or as

yellow incrustations on the eyelidmargins (Fig. 3), support the diagno-sis. In chronic conjunctivitis particu-larly if caused by S. aureus punctatekeratitis, blepharitis and marginalkeratitis are often seen.

Two studies conducted in generalpractice on patients with suspectedacute bacterial conjunctivitis (Carr1998; Wall et al. 1998) showed that apurulent secretion was present in8590% of patients, sticky eyelids(Fig. 4) in approximately 90% and aforeign body sensation and itching inapproximately 90%. More than 50%of the patients reported varyingdegrees of burning or stinging.

Fig. 1. Acute infective conjunctivitis withcharacteristic conjunctival injection and asmall amount of thick, yellow discharge inthe inferior fornix.

Fig. 2. Acute bacterial conjunctivitis withmixed injection and copious mucopurulentdischarge.

Fig. 3. Dry, yellow incrustations on the eye-lid margins, indicating an infective conjuncti-vitis.


8


5/13

Conjunctival or mixed injection wasnoted in almost all patients, whileabout 85% showed some erythema of the eyelid skin. On the other hand, arecent study including adult patientspresenting at 25 Dutch health centreswith a red eye and either (muco)puru-lent discharge or glued eyelids con-cluded that only early morning gluedeye(s) signicantly increased the prob-ability of a bacterial cause of thesymptoms, while itching and an earlierhistory of conjunctivitis decreased it(Rietveld et al. 2004). Therefore, nei-ther the type of injection nor theappearance of the secretion was signif-icantly associated with a positive bac-teriology obtained by conjunctivalswabbings. In a study reported by

Fitch et al. (1989), patient symptomsat presentation to an ocular emer-gency room were very similar in acutebacterial and viral conjunctivitis,although foreign body sensation wasrecorded signicantly more often inthose with a viral infection.

In younger children, particularly inkindergartens, acute conjunctivitismay occur in minor epidemics, and isthen commonly caused by H. inuen-zae or S. pneumoniae . As previouslymentioned, H. inuenzae conjunctivitis

is often associated with systemic infec-tions, particularly acute otitis mediaand infections of the upper respiratorytract. In a study by Buznach et al.(2005), the so-called conjunctivitis otitis syndrome was present in 32% of patients; in 82% of these cases theconjunctival culture yielded H. inuen-zae . Corresponding frequenciesreported by Block et al. (2000) were39% and 59%, respectively.

ComplicationsAcute bacterial conjunctivitis is almostinvariably a disease with a highly

favourable prognosis. Spontaneouscure is likely to occur within 1 2 weeks in at least 60% of cases(Sheikh & Hurwitz 2001; Rose et al.2005), and serious complications arevery seldomly seen (Schiebel 2003;Sheikh & Hurwitz 2006). However,large amounts of bacteria on the conjunctiva will imply a certain risk of keratitis, particularly in conditionspredisposing for corneal epithelialdefects. Accordingly, eyes with cornealepithelial disease and patients withdry eyes will be at greater risk of developing keratitis. An importantprecaution in the prevention of kerati-tis in patients with acute bacterialconjunctivitis is also to discontinuethe use of contact lenses and to omit

contact tonometry on such eyes(Hvding & Bertelsen 2004).

Differential diagnosesIn patients with an acute red eye, thepossibility of keratitis or iridocyclitismust always be considered. Earlierepisodes of acute uniocular redness inthe same eye should particularly raisethe suspicion of a recurrent H. simplex keratitis or a recurring acute

iridocyclitis. Previously known Mb.Bechterew also increases the probabil-ity of iridocyclitis. Simple techniques,such as uorescein staining and assess-ment of pupillary size and reactions,will in the absence of a slit lampgreatly help GPs to diagnose theseimportant diseases (Buckley 1990).

Viral conjunctivitis is undoubtedlyvery common, but its exact incidenceis difcult to assess because the symp-toms are usually mild and spontane-ous remission frequently occurs beforethe patients seek medical attention. Inaddition, many cases of viral conjunc-tivitis are misdiagnosed and mis-treated as bacterial conjunctivitis. Asin acute bacterial conjunctivitis, theviral conjunctivitis usually beginsacutely in one eye, while the secondeye is commonly involved within thenext week. The conjunctival hypera-emia is typically accompanied by anincreased watery secretion andenlarged preauricular lymph nodes.

Viral conjunctivitis is frequentlycaused by adenoviruses. In Japan,1 million cases of adenoviral conjunc-tivitis have been estimated to occureach year (Aoki & Tagawa 2002).

While most cases of adenovirus conjunctivitis only give mild, uncharacter-istic and transient symptoms, twowell-dened forms are described:pharyngoconjunctival fever (adenovi-rus type 3,4 and 7) and epidemic kera-toconjunctivitis (adenovirus type 8, 19and 37) (Rubenstein 1999; Aoki &Tagawa 2002). Both forms are trans-mitted via respiratory droplets ordirect ngereye contact. Multipleoutbreaks of adenovirus conjunctivitisare most frequently caused by adeno-virus type 8 or 19, and new cases of the disease commonly originate fromconsulting rooms or health institu-tions, where both the use of contami-nated tonometers and insufcienthand hygiene are important factors in

the spreading of the disease. Initially,the conjunctivitis is commonly accom-panied by a mild punctate epithelialkeratitis. These tiny epithelial inl-trates are not visible without a slitlamp. Subepithelial inltrates, whichpresumably are of immunological ori-gin, may appear in the second week of the disease (Dawson et al. 1972; Starr2004). These inltrates are not stainedby uorescein. Dense or conuatingcentral corneal inltrates may mark-edly reduce visual acuity. Tear uid is

usually watery, but mucopurulent oreven distinctly purulent secretion mayalso occur (Figs 5 and 6). Stingingand discomfort may be very pro-nounced, and an initial preauricularglandular enlargement is commonlyfound. In general practice (i.e. withoutthe use of a slit lamp), epidemic kera-toconjunctivitis should be suspectedon the basis of a careful case history,supported by the lack of improvementon topical antibacterial treatment. Thediagnosis is easily conrmed by conjunctival scrapings for viral culture,polymerase chain reaction (PCR) oradenovirus antigen.

Fig. 4. Sticky eyelids in acute bacterial conjunctivitis.

Fig. 5. Massive chemosis and thick, whitishdischarge in adenoviral epidemic keratoconjunctivitis.


9


6/13

Chlamydial conjunctivitis is primar-ily suspected in patients with typicalsymptoms and signs of acute bacterialconjunctivitis not satisfactorily res-ponding to standard antibacterialtreatment. The condition should par-

ticularly be considered in newbornswith prolonged conjunctivitis (Danne-vig et al. 1992; Rubenstein 1999). In astudy on neonatal conjunctivitis, signsof chlamydial conjunctivitis usuallypresented about 1 week postpartum,while signs of bacterial conjunctivitisappeared about 1 week later (Rapozaet al. 1986). PCR examination of conjunctival scrapings is diagnostic. Incases of neonatal chlamydial conjunc-tivitis, both the child and its parentsmust be treated simultaneously.

Allergic conjunctivitis is usuallyassociated with intense itching andwatery tear secretion. Chemosis is fre-quently present. In contact allergy,often caused by eye drops, the skinaround and especially below the eyebecomes erythematous or eczematousand itching. An exact record of thecase history is obviously of greatimportance in patients with suspectedallergic conjunctivitis.

A certain conjunctival follicular orpapillar hypertrophy is often seen inviral, chlamydial and allergic conjunc-tivitis, while this is not a frequent nd-ing in acute bacterial conjunctivitis.

In a contact lens wearer, thedevelopment of an acute red eyeshould primarily be suspected to becaused by the lens wear, and the useof lenses must be temporarily discon-tinued (Hvding & Bertelsen 2004).A strictly unilateral injection andpurulent discharge in a child shouldraise the suspicion of a retained for-eign body in the superior conjuncti-val fornix.

Because marked improvement orclinical cure of acute bacterialconjunctivitis usually occurs during

the rst few days of treatment, com-monly even without treatment, thecourse of the symptoms and signs aswell as the response to treatmentgive important diagnostic clues.Patients with suspected acute bacte-rial conjunctivitis should thereforealways be told to return for follow-up examinations if signicantimprovement does not occur withinthe next 12 days.

In vitro examinationsRoutine bacteriological examinationsare usually not indicated in the man-agement of clinically suspected acutebacterial conjunctivitis. According to

OBrien & Hahn (2005), microbiologi-cal evaluation should be consideredfor neonates and immunocompro-mised patients, as well as in allpatients with severe acute conjunctivi-tis. If treatment fails, a conjunctivalswab should be taken for bacteriologi-cal culture preferably a few daysafter the previously prescribed topicalmedication has been stopped (Mannis& Plotnik 2005). In addition, conjunc-tival scrapings should be taken foradenovirus, H. simplex virus and chla-

mydia cultures, PCR and or antigenanalyses. A general clue to the causeof a red eye may also be obtained byGiemsa-stained conjunctival scrapings,where neutrophils suggest a bacterialinfection, while lymphocytes andeosinophils indicate viral and allergicconjunctivitis, respectively (Weisset al. 1993; Cvenkel & Globocnik1997).

A study by Block et al. (2000) onpaediatric acute conjunctivitis showedthat ciprooxacin, ooxacin andtetracycline had the greatest in vitroactivity, followed by gentamycine,tobramycine, polymyxin Btrimetho-prim and polymyxin Bneomycin.Papa et al. (2002) found netilmicin tohave a signicantly better in vitroeffect than gentamycin, giving abroad-spectrum in vitro coverage com-parable to that of ciprooxacin, oox-acin and noroxacin. Orden Martinezet al. (2004) reported that ciprooxa-cin, chloramphenicol and rifampinwere active against the commonlyoccurring bacteria.

However, it should be stressed thatthere is only limited value of in vitrosusceptibility testing of the bacteria

isolated in external ocular disorderssuch as acute bacterial conjunctivitis,because in vitro susceptibility is basedon minimum inhibitory concentration(MIC) values for serum concentra-tions, while breakpoints for in vitrosusceptibility of topical medicationshave not been determined (Blocket al. 2000). Thus, all currently avail-able susceptibility tests use muchlower concentrations of antibioticsthan that obtained in the conjunctivaand cornea by using the same antibi-otic agents as eye drops. Therefore,clinical and microbiological cure isoften obtained despite an in vitro sus-ceptibility test indicating bacterialresistance to the drug tested.

TreatmentGeneral aspects

Keratitis and other potential compli-cations of acute bacterial conjunctivi-tis are seen very seldomly in otherwisehealthy subjects. Extensive clinicalexperience has thus proven acute bac-terial conjunctivitis to be a diseasewith a highly favourable prognosisand a high frequency of spontaneouscure.

A meta-analysis published bySheikh & Hurwitz (2001), based onthe few placebo-controlled, random-ized, double-blind studies available atthat time, showed clinical remissionwithin 5 days in 64% of placebo-trea-ted patients. However, the same anal-ysis also showed that topicalantibiotic treatment resulted in asignicantly improved early clinicalremission, as well as improved earlyand late microbiological remission.Because this meta-analysis onlyincluded patients recruited from spe-cialist care populations, the results arenot necessarily representative for pri-mary care patients with suspectedacute bacterial conjunctivitis. There-fore, a well-designed and well-controlled study on acute infectiveconjunctivitis in children aged6 months to 12 years recruited from12 general medical practices in theUK was recently carried out (Roseet al. 2005). At baseline, conjunctivalswabs yielded one or more of thestudy-designed pathogenic bacteria(H. inuezae, S. pneumoniae andM. catarrhalis ) in 78% of the patients,while adenovirus or picornavirus was

Fig. 6. Yellow-green secretion in adenoviralepidemic keratoconjunctivitis.


10


7/13

recovered alone or together with theabove-mentioned bacteria in 13% of the cases. Clinical cure was recordedin 83% of the placebo group within7 days, compared with 86% of thosereceiving chloramphenicol eye drops.The so-called number needed to treat(NNT) to achieve one more clinicalcure than that occurring spontane-ously was calculated to be 25. Corre-sponding frequencies of clinical cureon day 7 in patients growing one ormore of the study-designed pathogenicbacteria ( H. inuenzae, S. pneumoniaeand M. catarrhalis ) were 80% and85%, respectively (NNT 22). Whenthe proportion of children achievingclinical cure each day in the chloram-phenicol and placebo groups were

compared, a signicantly betterresponse was obtained in the chloram-phenicol group at day 2 (26.4% versus15.9%); this difference remained sta-tistically signicant up to day 7,although the mean difference in thetime to cure in the two groups wasonly 0.3 days during this period. Bac-terial eradication at day 7 was alsoobtained signicantly more frequentlyin the chloramphenicol group (40.0%)than in the placebo group (23.2%),resulting in NNT 6. The frequency

of new conjunctivitis episodes duringthe following 6 weeks was almostidentical in the two study groups. Aweakness of this important study is apossible selection bias: only one thirdof the children presenting to the par-ticipating GPs with acute infectiveconjunctivitis during the trial periodwere included in the study. Thus, fam-ily doctors may conceivably haverecruited the less severe cases for thestudy, believing that patients withmore severe symptoms needed topicalantibiotics. In addition, patientrecruitment only took place duringofce hours, while one may speculateif patients presenting out of ofcehours generally will have more severesymptoms and a more abundant conjunctival ora of pathogens.

Another double-blind, randomizedand placebo-controlled study on infec-tious conjunctivitis in primary carewas published by Rietveld et al.(2005). Adult patients presenting at 25Dutch GP centres with a red eye andeither (muco)purulent discharge orsticky eyelids were invited to partici-pate. At baseline, only 50 of the 163patients (30.7%) completing the trial

were culture-positive. Based on boththe clinical examination by the GPsafter 7 days and the daily diary keptby each of the patients, the efcacy of fusidic acid gel compared to placebogel was evaluated. At this stage, clini-cal cure was recorded in 62% of thefusidic acid group and 59% of theplacebo group. Neither the severitynor the duration of symptoms differedsignicantly in the two patient groups.However, the treatment effectappeared stronger in culture-positivepatients, where bacterial eradicationin the fusidic acid and in the placebogroup was obtained in 76% and 41%,respectively, although more than 60%of the cultures obtained at baselineshowed in vitro resistance to fusidic

acid. Minor adverse events, mainly aburning sensation after instillation of the study medication, occurred in14% of the fusidic acid group and in3% of the placebo group. The mainweaknesses of the study were the lim-ited number of culture-proven cases atbaseline and the fact that patientrecruitment was performed duringofce hours only. The authors con-cluded that although the study lackedthe power to conclusively demonstrateequivalence between fusidic acid and

placebo, it did not support the currentGP prescription practice of using topi-cal antibiotics in the vast majority of patients with suspected infective conjunctivitis.

In another interesting, open, ran-domized and controlled study from 30general practices including 307 chil-dren and adults with acute, presum-ably infective conjunctivitis, theresults obtained by no treatment,delayed topical antibiotic treatment orimmediate topical chloramphenicoltreatment were compared (Everittet al. 2006). The different treatmentsdid not inuence the severity of symp-toms during the rst 3 days, but theduration of moderate symptoms wasshortest in the immediate treatmentgroup (3.3 days) and longest in the notreatment group (4.9 days). In thedelayed antibiotic treatment group,53% eventually received topical anti-biotics. When asked about their belief in the effectiveness of antibiotics foracute conjunctivitis, as well as theirintention to reattend for eye infec-tions, those receiving immediate topi-cal antibiotic therapy had the highestpositive score. In accordance with

this, the reattendance rate within2 weeks was lower in the delayedtreatment group than in those receiv-ing immediate treatment. The distribu-tion of an information leaet aboutacute conjunctivitis did not signi-cantly inuence the attitude of thepatients regarding acute conjunctivitis.The authors judged a delayed pre-scription attitude to be the mostfavourable alternative in manyrespects, reducing both the use of top-ical antibiotics and the frequency of early reattendance without increasingthe severity of the symptoms.

Despite the clinically experiencedand now also well-documented highfrequency of spontaneous cure (Riet-veld et al. 2005; Rose et al. 2005; Eve-

ritt et al. 2006), all questions about therisk of serious adverse events in thosenot receiving antibiotic treatment havenot been fully addressed (Sheikh &Hurwitz 2005). The effect of a generalnon-prescription attitude on transmis-sion rates of pathogens also remainsto be claried (Rose et al. 2005). Sofar, it has therefore been the prevailingopinion in both the medical professionand the public that acute bacterialconjunctivitis should preferably betreated with topical antibiotics (Hr-

ven 1994; Ehlers & Bek 2004). In addi-tion to the moderately improvedclinical response, topical antibiotictreatment may be socioeconomicallyfavourable, because the risk of spread-ing the infection is probably reducedand the course of the disease is oftenshortened. Therefore, absence fromwork due to illness is reduced, particu-larly because children with acute conjunctivitis may return to kindergartenafter only 12 days of treatment (inmany countries, children are deniedadmittance to kindergarten or juniorschool before they have been treatedwith antibiotics or the signs of conjunctivitis have disappeared) (Lie1994; Rose et al. 2005). On the otherhand, in countries where the nationalhealth system carries the expenses of both consultations and medications,the negative socioeconomic conse-quences of an active consultation andtreatment strategy are easily seen.Other negative sides of an active pre-scription policy include the patientsmedical expenses, increased medicalprofession workload, medicalization of a minor illness, the risk of adversereactions to the antibiotics used and


11


8/13

the risk of widespread microbiologicalresistance. The difculty in making acorrect clinical distinction between abacterial and a viral conjunctivitiswithout an available slit lamp andtime-consuming laboratory tests is alsoa point in favour of a more expectanttreatment attitude. A qualitative studyof patients perceptions of acute infec-tive conjunctivitis (Everitt et al. 2003)showed that patients generally regardacute conjunctivitis as a minor disease,but they still commonly believe that itwill not heal spontaneously, andaccordingly they seek medical help.However, when the self-limiting natureof the condition is explained, they oftenexpress a preference to wait a few daysbefore seeking medical attention or

being treated with topical antibiotics.As mentioned earlier, concurrent

otitis media occurs relatively fre-quently in paediatric patients withacute bacterial conjunctivitis causedby H. inuenzae (Block et al. 2000;Buznach et al. 2005). In a study com-paring the efcacy of short-term oralcexime therapy with topical use of polymyxin-bacitracin, the frequency of acute otitis media developing duringthe rst 15 days following topical orsystemic antibiotic treatment was

almost identical (Wald et al. 2001).The pros and cons of prescribingtopical antibiotics to patients with sus-pected acute bacterial conjunctivitismay be summarized as in Table 1. Ahighly personal view on the treatmentof this condition is presented inTable 2.

Choice of topical antibiotics

In a double-blind study of patientswith acute bacterial conjunctivitis,Papa et al. (2002) found netilmicin eyedrops to be signicantly more effectivethan gentamycin in both eradicatinginfection and ameliorating signs andsymptoms. On the other hand, numer-ous clinical studies show no signicantdifferences between the clinical effectof various antibiotic eye drops inpatients with suspected acute bacterialconjunctivitis (Leibowitz 1991; Milleret al. 1992a; Miller et al. 1992b; Hr-ven 1993; Carr 1998; Wall et al. 1998;Jackson et al. 2002; Normann et al.2002). The choice of treatment willtherefore be inuenced by the ease of medication, side-effects, general micro-biological considerations and price.

With the exception of fusidic acid,most commercially available antibioticeye drops should initially be appliedat least six times daily. Fucithalmicis an aqueous suspension of micro-crystalline fusidic acid formulatedas viscous eye drops. A protractedrelease is thereby obtained, giving anadequate concentration of fusidic acidfor more than 12 hrs in tear uid(Thorn & Johansen 1997). Therefore,Fucithalmic needs only to be appliedtwice daily, and in addition gives far

less initial blurring of vision thanthe use of eye ointments. A novelenhanced viscosity ophthalmic formu-lation of tobramycin given twice dailywas also shown by Kernt et al. (2005)to be as effective as ordinary tobra-mycin eye drops used four times dailyfor acute bacterial conjunctivitis.Recently, an ordinary ophthalmic for-mulation of gatioxacin administered

twice daily proved to be as effective asthe identical medication administeredfour times a day (Yee et al. 2005).

Acute bacterial conjunctivitis inyoung children is most frequentlycaused by H. inuenzae, which in vitrosusceptibility tests commonly classifyas resistant to fusidic acid. On theother hand, both clinical experienceand treatment studies show that evensuch cases respond well to Fucithal-mic eye drops (Hrven 1993; Jack-son et al. 2002). This is probably dueto the fact that conventional fucidicacid susceptibility tests generally use1 l g ml as the breakpoint for suscep-tibility (Hrven 1993), while concen-trations in conjunctival uid of about15 l g ml after 1 hr, 10 l g ml after6 hrs and 6 l g ml after 12 hrs areobtained after one drop of Fucithal-mic (Thorn & Johansen 1997).In vitro studies have shown H. inuen-zae to be susceptible to fusidic acidin concentrations of approximately8 l g ml (Hrven 1993). A generalobservation in medical treatment isthat the fewer the prescribed dailydoses, the better the patient compli-ance. In glaucoma treatment it has

Table 2. Management of suspected acute bac-terial conjunctivitis: a personal view.

Always prescribe topical antibioticsPurulent mucopurulent secretion and

distinct discomfort and ocular rednessPatients and staff in nursing homes,

neonatal units, intensive care units, etc.Children going to kindergartensContact lens wearersPatients with dry eyes or corneal epithelial

diseaseUsually prescribe topical antibioticsPurulent mucopurulent secretion and severe

ocular rednessPatients with previously known external

ocular diseaseDelayed prescription or no antibiotic treatmentPatients who do not want immediate

antibiotic treatmentPatients with moderate (muco)purulent

discharge and little or no discomfortCooperative and well-informed patients

Table 1. The pros and cons of topical anti-biotics in suspected acute bacterial conjunctivitis.

Immediate antibacterial treatmentPros:More rapidly reduced bacterial growth

Reduced transmission rates?Reduced risk of keratitis and other

complications?Increased early clinical remission

Reduced time out of work or educationEarlier restart of contact lens wearReduced symptoms and worries

Early return to kindergarten junior schoolLess parental time out of workSocioeconomically favourable

Cons:Increased burden on the healthcare systemSocioeconomically unfavourable (if society

pays medication and GP fees)Often unnecessary use of topical antibiotics

Patient adverse effectsNegative inuence on the normal ora of

both patient and milieuIncreased risk of bacterial resistance

Delayed antibacterial treatmentPros:Reduced use of topical antibiotics

(about 50%?)Reduced medicalization of an innocent

conditionImproved patient education and

responsibilityCons:Little or no reduction of health service

attendanceIncreased time out of work education

kindergartenNo antibacterial treatmentPros:Very high percentage of spontaneous clinical

cure 1 weekLess antibiotic load on patient and societyNo adverse events related to topical

antibioticsCons:Increased time out of work education

kindergartensIncreased risk of transmitting the

infection (?)At least a theoretically increased risk of

complications


12


9/13

been shown that compliance is signi-cantly improved if the treatment sche-dule is reduced from three or fourtimes a day to twice a day (Norell1981; MacKean & Elkington 1983;Kass et al. 1987) and that the middaydose is particularly likely to be omit-ted (Rotchford & Murphy 1998). Cor-respondingly, several studies haveshown a slight, but statistically signi-cant, improvement of treatment com-pliance by using Fucithalmic ratherthan chloramphenicol eye drops inacute bacterial conjunctivitis (Carr1998; Jackson et al. 2002; Normannet al. 2002). Fucithalmic thereforeappears to be an attractive therapeuticalternative in most cases of acute bac-terial conjunctivitis, particularly in

children and in the elderly, who needassistance to administer the eye drops.In addition to the antibiotic action,the lubricating effect of the viscousformulation probably contributes tothe subjectively improved condition of the patients.

Systemic use of fusidic acid is animportant therapeutic option inpatients with severe S. aureus infec-tions, particularly osteomyelitis. How-ever, it has been known for severaldecades that because of the high spon-

taneous mutation rate if fusidic acid isgiven as systemic monotherapy, thedrug should always be given in combi-nation with other antistaphylococcaldrugs (Brown & Thomas 2002; Dobie& Gray 2004; Howden & Grayson2006). During the last decade, a mark-edly increased incidence of fusidic-acid-resistant S. aureus has beenreported in many countries, while onlylow rates of fusidic acid resistancehave been found in the USA, wherefusidic acid is not generally availableas a therapeutic option (Ravenscroftet al. 2000; Rrtveit & Rrtveit 2003;Shah & Mohanraj 2003; Dobie &Gray 2004; El-Zimaity et al. 2004;Zinn et al. 2004). It is commonlybelieved that long-term topical fusidicacid monotherapy is the main causeof the increasing resistance seen inmany countries, particularly the wide-spread use of fusidic acid in chronicskin infections such as impetigo, fol-liculitis and crural ulcers (Ravenscroftet al. 2000; Brown & Thomas 2002;Mason & Howard 2004; Howden &Grayson 2006). The use of fusidic acidin patients with acute conjunctivitishas so far not been implicated as an

important source of increased staphy-lococcal resistance, most probablybecause Fucithalmic delivers a hightopical drug concentration and is usu-ally only prescribed for short-termuse.

Although the simple treatment sche-dule (morning and evening) is animportant argument in favour of Fucithalmic in the treatment of acutebacterial conjunctivitis, it should beemphasized that chloramphenicol eyedrops are also an excellent choice inthe treatment of this disease. Theeffect of chloramphenicol is as goodas that obtained with both Fucithal-mic and more recent, broad-spec-trum antibiotics such as ciprooxacin,noroxacin and tobramycin (Miller

et al. 1992b; Hrven 1993; Carr 1998;Orden Martinez et al. 2004). Thegreatest disadvantage of chloramphe-nicol eye drops is the need for fre-quent application, which may increasethe risk of suboptimal patient compli-ance. However, Lrum et al. (1994)reported a satisfactory clinical effecteven with a simplied regimen (fourtimes daily), although the success ratewas slightly lower than that obtainedby the ordinary chloramphenicoltreatment regime.

Chloramphenicol eye drops are themedication of choice in acute bacterialconjunctivitis in many countries, hav-ing more than a 90% market share inAustralia and almost a 70% marketshare in the UK (Sheikh & Hurwitz2001). Chloramphenicol also repre-sents more than 50% of the topicalocular antibiotics prescribed in Ire-land, and is the topical medicationused in about 55% of patients treatedfor an acute red eye in England(Doona & Walsh 1995). On the otherhand, as a result of the fear of serioussystemic side-effects, particularlyaplastic anaemia and grey baby syn-drome, topical treatment with chl-oramphenicol is rarely used in theUSA and some other countries (Ray-ner & Buckley 1996; Doona & Walsh1998). The relationship between oraluse of chloramphenicol and severe,often fatal, aplastic anaemia has beenknown for more than 50 years (Ritchet al. 1950). Later, the incidence of aplastic anaemia following oral chl-oramphenicol therapy has been esti-mated to be about one in 36 000patients (Wallerstein et al. 1969),which is approximately 13 times

greater than the risk of idiopathicaplastic anaemia in the whole popula-tion (Fraunfelder et al. 1982; Doona& Walsh 1995). A reversible bonemarrow disease mainly suppressingthe red blood cell line is caused by adose-related response, and has neverbeen suspected to be caused by topicalchloramphenicol use. On the otherhand, a progressive and often fatalbone marrow aplasia affecting allthree haematopoetic lines is an idio-syncratic reaction, which may conceiv-ably be precipitated even by topicaluse (Stern & Killingsworth 1989). Ithas been proposed that aplastic anae-mia triggered by topical chloramphe-nicol medication may be the result of an individual metabolic predisposition

and that the idiosyncratic reactiontherefore only occurs in geneticallypredisposed subjects (Yunis 1973;Fraunfelder et al. 1993). Severalreports have suggested a relationshipbetween aplastic anaemia and the useof eye drops or ointment containingchloramphenicol (Rosenthal & Black-man 1965; Carpenter 1975; Abramset al. 1980; Fraunfelder et al. 1982).However, the aetiological role of chl-oramphenicol in these published caseshas been questioned, particularly

because many of the patients had alsoused several systemic medications, orhad extraocular diseases or a familyhistory of blood dyscrasias (Besamus-ca & Bastiaensen 1986; Buckley et al.1995; Rayner & Buckley 1996; Lan-caster et al. 1998; Wiholm et al. 1998).Two population studies published byWiholm et al. (1998) did not supportthe claim that chloramphenicol eyedrops increase the risk of aplasticanaemia. Furthermore, in a surveyincluding about 400 general practicesin the UK between 1988 and 1995, atotal of 442 543 patients received674 148 prescriptions for chloramphe-nicol eye drops (Lancaster et al.1998). Among these, three patientswith serious haematological suppres-sion were detected. The authors con-clude that even in the unlikely eventthat all these three cases were causedby chloramphenicol eye drops, therisk of serious haematological toxicityis small. A Dutch regional retrospec-tive casecontrol study identied 12patients with aplastic anaemia and190 patients with other cytopenic dys-crasias during a 4 year study period(Besamusca & Bastiaensen 1986).


13


10/13

None of these cases were unequivo-cally related to the topical use of chl-oramphenicol, which on a yearly basiswas used by one out of 29 patientsincluded in the survey; the frequencyof topical chloramphenicol use wasapproximately the same in patientswith and without blood dyscrasias.Therefore, although the occurrence of non-dose-related cases of idiosyncraticaplastic anaemia can not be totallyexcluded, it must evidently be extre-mely rare (Walker et al. 1998). Inconclusion, topical ophthalmic useof chloramphenicol appears to be asafe, effective and cheap treatment inpatients with external ocular infec-tions, but the medication shouldnot be given to individuals with a per-

sonal or family history indicatinghaematological disease (Buckley et al.1995).

In the USA and many other coun-tries, eye drops containing cipro-oxacin, noroxacin, gentamycin,tobramycin and other broad-spectrumantibacterial agents are frequentlyused in acute bacterial conjunctivitis.Because these medications are oftenused in topical or systemic treatmentof more serious infections, the unnec-essary development of bacterial resis-

tance is highly undesirable. Increasingoccurrence of ciprooxacin-resistantP. aeruginosa has recently beenreported by several authors. Chaudhryet al. (1999) found that while only oneof 227 (0.44%) ocular isolates wasresistant to ciprooxacin from 1991 to1994, eight of 196 (4.1%) ocular iso-lates showed in vitro resistance from1995 to 1998. Resistance to ciproox-acin implied resistance also to mostother (at that time) readily availableuoroquinolones. In the same study,all isolates were sensitive to genta-mycin.

A 5 year review of the in vitro ef-cacy of uoroquinolones on 1053 ocu-lar isolates obtained in 19931997from 825 corneal ulcers (Goldsteinet al. 1999) showed a signicantlyincreased resistance of S. aureus toboth ciprooxacin and ooxacin(from about 5% in 1993 to 35% in1997). In contrast to the ndings of Chaudhry et al. (1999), none of theP. aeruginosa isolates were classiedas resistant to uoroquinolones, butresistance among other pseudomonasspecies increased from zero to 28.6%during the same period. Considerable

gaps in the coverage against coagu-lase-negative staphylococci, strepto-coccus species, enterococci species andanaerobs were also found, but thesegaps did not appear to change overthe years included in the study. Later,levooxacin eye drops have beenreported to give a higher rate of microbial eradication than ooxacineye drops in patients with bacterialconjunctivitis (Schwab et al. 2003),mainly because of improved eradica-tion of S. pneumoniae and H. inuen-zae . However, clinical cure rates didnot differ in the two patient groups,illustrating that clinical cure andmicrobiological eradication are notsynonyms. Contrasting the report byGoldstein et al. (1999), both drugs

were highly effective against S. aureus ,eradicating all strains present at base-line. Recent in vitro studies of bacte-rial isolates from patients with acuteconjunctivitis showed that the fourth-generation uoroquinolones (gatiox-acin and moxioxacin) had somewhatbetter in vitro efcacy than earlier u-oroquinolones against G+ isolates,but not against H. inuenzae (Kowal-ski et al. 2005).

Several other studies have alsoshown alarmingly high and increasing

frequencies of antibiotic resistanceamong bacteria isolated from patientswith presumed bacterial conjunctivitis.Block et al. (2000) reported that theoccurrence of beta-lactamase-pro-ducing H. inuenzae isolated fromchildren with acute conjunctivitisincreased from 44% in the late 1980sto 69% at the end of their study, andthat the frequency of penicillin-non-susceptible S. pneumoniae (PNSP)then was three times higher than thatfound 5 years earlier (Doern et al.1996). Their study also showed dimin-ished antibacterial activity of gentami-cin, tobramycin and polymyxin B. Inconjunctival cultures from 428 chil-dren aged 236 months with suspectedacute bacterial conjunctivitis, Buznachet al. (2005) found beta-lactamaseproduction in 29% of H. inuenzaeisolates, while penicillin non-suscepti-bility was observed in 60% of S. pneu-moniae isolates. As much as 96% of the H. inuenzae isolates and 97% of the S. pneumoniae isolates showedin vitro susceptibility to chlorampheni-col.

Every antibacterial agent will tosome extent inuence the ecological

balance both in the patients normalbacterial ora and in the surroundingmilieu. Broad-spectrum medicationswith a large ecological shadow, suchas uoroquinolons and tetracyclines,have a greater tendency than agentswill a smaller ecoshadow to causebacterial resistance as well as alteringthe normal bacterial ora (Midtvedt2004). Among others, OBrien &Hahn (2005) have suggested that u-oroquinolones should not be usedroutinely for bacterial conjunctivitis,but should be reserved for particularlysevere cases. In my opinion, neitheruoroquinolones nor aminoglycosidesshould be used in the treatment of uncomplicated acute bacterial conjunctivitis, and topical administration

of these medications should berestricted to more serious eye infec-tions and to cases of conjunctivitis inpatients not tolerating or improvingon fusidic acid and chloramphenicoltreatment.

ReferencesAbrams SM, Degnan TJ & Vinciguerra V

(1980): Marrow aplasia following topicalapplication of chloramphenicol eye oint-ment. Arch Intern Med 140 : 576577.

Aoki K & Tagawa Y (2002): A twenty-oneyear surveillance of adenoviral conjunctivi-tis in Sapporo, Japan. Int Ophthalmol Clin42 : 4954.

Besamusca FW & Bastiaensen LA (1986):Blood dyscrasias and topically applied chl-oramphenicol in ophthalmology. Doc Oph-thalmol 64 : 8795.

Block SL, Hedrick J, Tyler R, Smith A, Find-lay R, Keegan E & Stroman DW (2000):Increasing bacterial resistance in pediatricacute conjunctivitis (19971998). Antimic-rob Agents Chemother 44: 16501654.

Brennen C & Muder RR (1990): Conjunctivi-

tis associated with methicillin resistantStaphylococcus aureus in a long-term-carefacility. Am J Med 88 : 1417.

Brown EM & Thomas P (2002): Fusidic acidresistance in Staphylococcus aureus iso-lates. Lancet 359 : 803.

Buckley SA (1990): Survey of patients takingtopical medication at their rst presenta-tion to eye care casualty. Br Med J 300:14971498.

Buckley RJ, Kirkness CM, Kanski JJ, Ridg-way AE, Tullo AB & Watson PG (1995):Safe in patients with no history of blooddyscrasia. Br Med J 311 : 451.

Buznach N, Dagan R & Greenberg D (2005):

Clinical and bacterial characteristics of acute bacterial conjunctivitis in children inthe antibiotic resistance era. Pediatr InfectDis J 24 : 823828.


14


11/13

Cagle CD & Abshire RL (1981): Quantitativeocular bacteriology. Invest Ophth Vis Sci20 : 751757.

Carpenter G (1975): Chloramphenicol eye-drops and marrow aplasia. Lancet Aug 16:326327.

Carr WD (1998): Comparison of Fucithal-mic (fusidic acid viscous eye drops 1%)and Chloromycetin Redidrops (chloram-phenicol eye drops 0.5%) in the treatmentof acute bacterial conjunctivitis. J Clin Res1: 403411.

Chaudhry NA, Flynn HW, Murray TG,Tabandeh H, Mello MJ & Miller D(1999): Emerging ciprooxacin-resistantpseudomonas aeruginosa. Am J Ophhtal-mol 128 : 509510.

Cvenkel B & Globocnik M (1997): Conjuncti-val scrapings and impression cytology inchronic conjunctivitis. Correlation withmicrobiology. Eur J Ophthalmol 7: 1923.

Dannevig L, Straume B & Melby K (1992):Ophthalmia neonatorum in Northern Nor-way. II. Microbiology with emphasis onChlamydia trachomatis. Acta Ophthalmol(Copenh) 70 : 1925.

Dart JKG (1986): Eye disease at a commu-nity health centre. Br Med J 293: 1477 1480.

Dawson CR, Hanna L & Togni B (1972):Adenovirus type 8 infections in the UnitedStates. IV. Observations on the pathogene-sis of lesions in severe eye disease. ArchOphthalmol 87 : 258268.

Dobie D & Gray J (2004): Fusidic acid resis-tance in Staph. Aureus. Arch Dis Child 89 :7477.

Doern GV, Brueggemann A, Holley HP &Rauch AM (1996): Antimicrobial resistanceof Streptococcus pneumoniae recoveredfrom outpatients in the United States dur-ing the winter months of 1994 to 1995:results of a 30-center national surveillancestudy. Antimicrobial Agents Chemotherapy40 : 12081213.

Doona M & Walsh JB (1995): Use of chl-oramphenicol as topical eye medication:time to cry halt? Br Med J 310 : 12171218.

Doona M & Walsh JB (1998): Topical chl-oramphenicol is an outmoded treatment.Br Med J 316 : 1903.

Ehlers N & Bek T (2004): Akut purulent conjunctivitis. In: Hvding G (ed.) Oftalmo-logi. Nordisk lrebok og atlas. Bergen:Studia, 129132.

El-Zimaity D, Kearns AM, Dawson SJ, PriceS & Harrison GA (2004): Survey, charac-terization and susceptibility to fusidic acidof Staphylococcus aureus in the Camarthenarea. J Antimicrob Chemother 54: 441 446.

Everitt H, Kumar S & Little P (2003): Aqualitative study of patients perceptions of acute infective conjunctivitis. Br J GeneralPract 53 : 3641.

Everitt H & Little P (2002): How do GPs

diagnose and manage acute infective conjunctivitis? A GP survey. Fam Pract 19:658660.

Everitt HA, Little PS & Smith PW (2006): Arandomised controlled trial of managementstrategies for acute infective conjunctivitisin general practice. Br Med J 333 : 321324.

Fahmy JA, Mller S & Weiss Bentzon M(1975): Bacterial ora of the normal conjunctiva II. Methods of obtaining cultures.Acta Ophthalmol (Copenh) 53 : 237253.

Fitch CP, Rapoza PA, Owens S, Murillo-Lo-pez F, Johnson RA, Quinn TC, Pepose JS& Taylor HR (1989): Epidemiology anddiagnosis of acute conjunctivitis at aninner-city hospital. Ophthalmology 96:12151220.

Franklin RM, Winkelstein JA & Seto DSY(1977): Conjunctivitis and keratoconjuncti-vitis associated with primary immunode-ciency diseases. Am J Ophthalmol 84: 563 566.

Fraunfelder FT, Bagby GC & Kelly DJ(1982): Fatal aplastic anemia following top-ical administration of ophthalmic chloram-phenicol. Am J Ophthalmol 93 : 356360.

Fraunfelder FT, Morgan RL & Yunis AA(1993): Blood dyscrasias and topical oph-thalmic chloramphenicol. Am J Ophthal-mol 115 : 812813.

Friedlaender MH, Masi RJ, Osumoto M,Smolin G & Ammann AJ (1980): Ocularmicrobial ora in immunodecient patients.Arch Ophthalmol 98 : 289294.

Garibaldi RA, Brodine S & Matsumiya S(1981): Infections among patients in nurs-ing homes. N Engl J Med 305 : 731735.

Gigliotti F, Williams WT, Hayden FG &Hendley JO (1981): Etiology of acute conjunctivitis in children. J Pediatr 98: 531536.

Goldstein MH, Kowalski RP & Gordon YJ(1999): Emerging uoroquinolone resis-tance in bacterial keratitis. Ophthalmology106 : 13131318.

Gritz DC, Scott TJ, Sedo SF, Cevallos AV,Margolis TP & Whitcher JP (1997): Ocularora of patients with AIDS compared withthose of HIV-negative patients. Cornea 16:400405.

Hilton E, Adams AA, Uliss A, Lesser ML,Samuels S & Lowy FD (1983): Nosocomialbacterial eye infections in intensive-careunits. Lancet 11 : 13181320.

Hrven I (1993): Acute conjunctivitis. A com-parison of fusidic acid viscous eye dropsand chloramphenicol. Acta Ophthalmol(Copenh) 71 : 165168.

Hrven I (1994): Bakteriell konjunktivitt[Bacterial conjunctivitis]. Tidsskr NorLgeforen 114 : 661.

Hvding G (1981): The conjunctival and con-tact lens bacterial ora during lens wear.Acta Ophthalmol (Copenh) 59 : 387401.

Hvding G (2004): Akutt bakteriell konjunktivitt [Acute bacterial conjunctivitis].Tidsskr Nor Lgeforen 124 : 15181520.

Hvding G & Bertelsen T (2004): Kontaktlin-ser. In: Hvding G (ed.) Oftalmologi. Nor-disk lrebok og atlas. Bergen: Studia, 103

108.Hvding G, Bratland SZ & Digranes A(1991): Rdt ye praktisk veiledning i

allmenn praksis. Oslo: Lvens KemiskeFabrik.

Howden BP & Grayson ML (2006): Dumberand dumber the potential waste of a use-ful antistaphylococcal agent: emerging fusi-dic acid resistance in Staphylococcusaureus. Clin Infect Dis 42 : 394400.

Inoue Y (2002): Ocular infections in patientswith atopic dermatitis. Int Ophthalmol Clin42 : 5569.

Isenberg SJ, Apt L, Yoshimori R, McCartyJW & Alvarez RN (1988): Source of theconjunctival bacterial ora at birth andimplications for ophthalmia neonatorumprophylaxis. Am J Ophthalmol 106: 458 462.

Jackson WB, Low DE, Dattani D, WhitsittPF, Leeder RG & MacDougall R (2002):Treatment of acute bacterial conjunctivitis:1% fusidic acid viscous drops vs. 0.3%tobramycin drops. Can J Ophthalmol 37:228237.

Jett BD (2005): Host defence against ocularinfection. In: Tasman M & Jaeger EA (eds)Duanes clinical ophthalmology, vol. 2.Philadelphia: Lippincott, Williams &Wilkins, 120.

Kass MA, Gordon M, Morley RE, MeltzerDW & Goldberg JJ (1987): Compliancewith topical timolol treatment. Am J Oph-thalmol 103 : 188193.

Kernt K, Martinez M, Bertin D, Stroman D,Cupp G, Martinez C, Tirado M & GuaschJ (2005): A clinical comparison of two for-mulations of tobramyin 0.3% eyedrops inthe treatment of acute bacterial conjuncti-vitis. Eur J Ophthalmol 15 : 541549.

King S, Devi SP, Mindorff C, Patrick ML,Gold R & Ford-Jones EL (1988): Nosoco-mial pseudomonas aeruginosa conjunctivi-tis in a pediatric hospital. Infect ControlHosp Epidemiol 9: 7780.

Kowalski RP & Roat MI (2005): Normalora of the human conjunctiva and eyelid.In: Tasman M & Jaeger EA (eds) Duanesclinical ophthalmology, vol. 2. Philadelphia:Lippincott, Williams & Wilkins, 111.

Kowalski RP, Yates KA, Romanowski EG,Karenchak LM, Mah ES & Gordon YJ(2005): An ophthalmologists guide tounderstanding antibiotic susceptibility andminimum inhibitory concentration data.Ophthalmology 112 : 19871991.

Krohn MA, Hillier SL, Bell TA, KronmalRA & Grayston JT (1993): The bacterialetiology of conjunctivitis in early infancy.Am J Epidemiol 138 : 326332.

Lrum E, Fiskaadal HJ, Erdal JE & LbergRM (1994): Kloramfenikol yedra per vedakutt bakteriell konjunktivitt [Chloramphe-nicol eyedrops in acute bacterial conjuncti-vitis. A comparison of 2 dosage regimes ingeneral practice]. Tidsskr Nor Lgefor 114 :671673.

Lancaster T, Swart AM & Jick H (1998):Risk of serious haematolocical toxicity with

use of chloramphenicol eye drops in a Brit-ish general practice database. Br Med J316 : 667.


15


12/13

Leibowitz HM (1991): Antibacterial effective-ness of ciprooxacin 0.3% ophthalmicsolution in the treatment of bacterialconjunctivitis. Am J Ophthalmol 112 (Sup-pl): 2933.

Lie KK (1994): Konjunktivitt og barnehager.Statens Institutt for Folkehelse MSIS-Rap-port 22:6. Oslo: Statens Institutt for Folk-ehelse.

Locatcher-Khorazo D & Seegal B (1972):Microbiology of the eye. St Louis: Mosby.

MacKean JM & Elkington AR (1983): Com-pliance with treatment of patients withchronic open-angle glaucoma. Br J Ophtal-mol 67 : 4649.

Mahajan VM (1983): Acute bacterial infec-tions of the eye: their aetiology and treat-ment. Br J Ophthalmol 67 : 191194.

Mannis MJ & Plotnik RD (2005): Bacterialconjunctivitis. In: Tasman M & Jaeger EA(eds) Duanes clinical ophthalmology, vol.4. Philadelphia: Lippincott, Williams &Wilkins, 111.

Martins EN, Alvarenga LS, Ho ing-LimaAL, Freitas D, Zorat-Yu MC, Farah ME& Mannis MJ (2004): Aerobic bacterialconjunctival ora in diabetic patients. Cor-nea 23 : 136142.

Mason BW & Howard AJ (2004): Fusidicacid resistance in community isolates of methicillin susceptible Staphylococcus aur-eus and the use of topical fusidic acid: aretrospective casecontrol study. Int JAntimicrob Agents 23 : 300303.

McClellan KA (1997): Mucosal defence of theouter eye. Surv Ophthalmol 42 : 233245.

McCormick M, Fleming D & Charlton J(1995): Morbidity statistics from generalpractice: Fourth National Survey 199192.London. HMSO.

McDonnell PJ (1988): How do general practi-tioners manage eye disease in the commu-nity? Br J Ophthalmol 72 : 733736.

McMinn PC, Stewart J & Burrell CJ (1991):A community outbreak of epidemic kerato-conjunctivitis in Central Australia due toadenovirus type 8. J Infect Dis 164 : 1113 1118.

Midtvedt T (2004): Om bruken av antimikro-bielle midler. In: Vilberg A (ed.) Norsklegemiddelha ndbok for helsepersonell.Oslo: Foreningen for utgivelse av Norsklegemiddelha ndbok, 602606.

Miller B & Ellis P (1977): Conjunctival orain patients receiving immunosuppressivedrugs. Arch Ophthalmol 95 : 20122014.

Miller IM, Vogel R, Cook TJ & WittreichJM (1992a): Topically administered nor-oxacin compared with topically adminis-tered gentamicin for the treatment of external ocular bacterial infections. Am JOphthalmol 113 : 638644.

Miller IM, Wittreich JM, Cook T & Vogel R(1992b): The safety and efcacy of topicalnoroxacin compared with chlorampheni-col for the treatment of external ocular

bacterial infections. Eye 6: 111114.Norell SE (1981): Monitoring compliancewith pilocarpine therapy. Am J Ophthal-mol 92 : 727731.

Normann EK (2005): Conjunctivitis in chil-dren. Lancet 366 : 67.

Normann EK, Bakken O, Peltola J, Andreas-son B, Buhl S, Sigg P & Nielsen K (2002):Treatment of acute neonatal bacterial conjunctivitis: a comparison of fusidic acid tochloramphenicol eye drops. Acta Ophthal-mol Scand 80 : 183187.

OBrien TP & Hahn TW (2005): Therapy of bacterial infections of the eye. In: TasmanM & Jaeger EA (eds) Duanes clinical oph-thalmology, vol. 3. Philadelphia: Lippin-cott, Williams & Wilkins, 69.

Olafsen LD, Strvold G & Melby K (1986):A microbiological study of conjunctivitiswith emphasis on Chlamydia trachomatisin Northern Norway. Acta Ophthalmol(Copenh) 64 : 463470.

Orden Martinez B, Martinez Ruiz R & Mil-lan PR (2004): Conjunctivitis bacteriana:pato genos ma s prevalentes y sensibilidadantibio tica [Bacterial conjunctivitis: mostprevalent pathogens and their antibioticsensitivity]. An Pediatr (Barc) 61 : 3236.

Papa V, Aragona P, Scuderi AC, Blanco AR,Di Zola PBA, Santocono M & Milazzo G(2002): Treatment of acute bacterial conjunctivitis with topical netilmicin. Cornea21 : 4347.

Rapoza PA, Quinn TC, Kiesling LA & Tay-lor HR (1986): Epidemiology of neonatalconjunctivitis. Ophthalmology 93 : 456461.

Ravenscroft JC, Layton A & Barnham M(2000): Observations on high levels of fusi-dic acid resistant Staphylococcus aureus inHarrogate, North Yorkshire, UK. ClinExp Dermatol 25 : 327330.

Rayner SA & Buckley RJ (1996): Ocular chl-oramphenicol and aplastic anemia. Is therea link? Drug Safety 14 : 273276.

Richardson M, Elliman D, Maguire H, Simp-son J & Nicoll A (2001): Evidence base of incubation periods, periods of infectious-ness and exclusion policies for the controlof communicable diseases in schools andpreschools. Pediatr Infect Dis J 20: 380 391.

Rietveld RP, ter Riet G, Bindels PJE, BinkD, Sloos JH & van Weert HCPM (2005):The treatment of acute infectious conjunc-tivitis with fusidic acid: a randomised con-trolled trial. Br J General Pract 55: 924 930.

Rietveld RP, ter Riet G, Bindels PJE, SloosJH & van Weert HCPM (2004): Predictingbacterial cause in infectious conjunctivitis:cohort study on informativeness of combi-nations of signs and symptoms. Br Med J329 : 206210.

Rietveld RP, van Wert HCMP, Ter Riet G &Bindels PJE (2003): Diagnostic impact of signs and symptoms in acute infectiousconjunctivitis: systematic literature search.Br Med J 327 : 789.

Ringvold A, Vik E & Bevanger LS (1985):Moraxella lacunata isolated from epidemic

conjunctivitis among teen-aged females.Acta Ophthalmol (Copenh) 63 : 427431.Ritch ML, Ritterhof RJ & Hoffman RJ

(1950): A fatal case of aplastic anemia

following chloramphenicol therapy. AnnIntern Med 33 : 14591467.

Rrtveit S & Rrtveit G (2003): Epidemi avbulls impetigo i Austevoll kommune a r2002 [An epidemic of bullous impetigo inthe municipality of Austevoll in the year of 2002]. Tidsskr Nor Lgeforen 123: 2557 2560.

Rose PW, Harnden A, Brueggemann AB,Perera R, Sheikh A, Crook D & Mant D(2005): Chloramphenicol treatment foracute infective conjunctivitis in children inprimary care: a randomised double-blindplacebo-controlled study. Lancet 366: 37 43.

Rosenthal RL & Blackman A (1965): Bonemarrow hypoplasia following use of chl-oramphenicol eye drops. JAMA 191: 148 149.

Rotchford AP & Murphy KM (1998): Com-pliance with timolol treatment in glaucoma.Eye 12 : 234236.

Ruben FL, Norden CW, Heisler B & KoricaY (1984): An outbreak of Streptococcuspyogenes infections in a nursing home.Ann Intern Med 101 : 494496.

Rubenstein JB (1999): Disorders of the conjunctiva and limbus. In: Yanoff M & Du-ker JS (eds) Ophthalmology. St Louis:Mosby, 1218.

Schiebel NEE (2003): Use of antibiotics inpatients with acute bacterial conjunctivitis.Ann Emerg Med 41 : 407409.

Schmitz H, Wigand R & Heinrich W (1983):Worldwide epidemiology of human adeno-virus infections. Am KJ Epidemiol 117 :455466.

Schwab IR, Friedlaender M, McCulley J,Lichtenstein SJ & Moran CT (2003): Aphase III clinical trial of 0.5% levooxacinophthalmic solution versus 0.3% ooxacinophthalmic solution for the treatment of bacterial conjunctivitis. Ophthalmology110 : 457465.

Schwartz B, Harrison LH, Motter JS, MotterRN, Hightower AW & Broome CV (1989):Investigation of an outbreak of Moraxellaconjunctivitis at a Navajo boarding school.Am J Ophthalmol 107 : 341347.

Seibel W & Ruprecht KW (1983): Bacterio-logische Befunde in Bindehautabstrichen[Bacteriological ndings in conjunctivalsmears]. Klin Mbl Augenheilk 183 : 6062.

Shah M & Mohanraj M (2003): High levelsof fusidic acid-resistant Staphylococcusaureus in dermatology patients. Br J Der-matol 148 : 10181020.

Sharma T, Grewal J, Gupta S & Murray PI(2004): Ophthalmic manifestations of acuteleucemias: the ophthalmologists role. Eye18 : 663672.

Shayegani M, Parsons LM, Gibbons WE Jr& Cambell D (1982): Characterization of nontypable Streptococcus pneumoniae-likeorganisms isolated from outbreaks of conjunctivitis. J Clin Microbiol 16 : 814.

Sheikh A & Hurwitz B (2001): Topical antibi-otics for acute bacterial conjunctivitis: asystematic review. Br J General Prac 51:473477.


16


13/13

Sheikh A & Hurwitz B (2005): Topical anti-biotics for acute bacterial conjunctivitis:Cochrane systematic review and meta-analysis update. Br J General Prac 55:962964.

Sheikh A & Hurwitz B (2006): Antibioticsversus placebo for acute bacterial conjunc-tivitis. Cochrane Database Syst Rev 2:CD001211.

Singer TR, Isenberg SJ & Apt L (1988): Con- junctival anaerobic and aerobic bacterialora in paediatric versus adult subjects. BrJ Ophthalmol 72 : 448451.

Solberg R, Meberg A & Schyen R (1991):Neonatal konjunktivitt i en barselavdelingog en neonatalenhet [Neonatal conjunctivi-tis in a nursery and a neonatal unit].Tidsskr Nor Lgefor 111 : 12301232.

Starr MB (2004): How best to treat adenovi-ral corneal opacities? Cornea 23 : 217219.

Stern GA & Killingsworth DW (1989): Com-plications of topical antimicrobial agents.Int Ophthalmol Clin 29 : 137142.

Tarabishy AB, Hall GS, Procop GW & JengBH (2006): Bacterial culture isolates fromhospitalized pediatric patients with conjunctivitis. Am J Ophthalmol 142 : 678 680.

Thiel HJ & Schumacher U (1994): U ber dieStandortora der menschlichen Bindehaut.Untersuchungen von 135 Personenunterschiedlichen Alters [Normal ora of the human conjunctiva: a study of 135 per-sons of various ages]. Klin MonatblAugenheilk 205 : 348357.

Thorn P & Johansen S (1997): Pharmacoki-netic investigation of fusidic acid 1% vis-cous eye drops in healthy volunteers. Eur JOphthalmol 7: 912.

Trottier S, Stenberg K, von Rosen IA &Svanborg C (1991): Haemophilus inuen-zae causing conjunctivitis in day-care chil-dren. Pediatr Infect Dis J 10 : 578584.

Wald ER, Greenberg D & Hoberman A(2001): Short term oral cexime therapyfor treatment of bacterial conjunctivitis.Pediatr Infect Dis J 21 : 10391042.

Walker S, Diaper CJM, Bowman R, SweeneyG, Seal DV & Kirkness CM (1998): Lack of evidence for systemic toxicity following top-ical chloramphenicol use. Eye 12: 875879.

Wall AR, Sinclair N & Adenis JP (1998):Comparison of Fucithalmic (fusidic acidviscous eye drops 1%) and Noroxin (nor-oxacin ophthalmic solution 0.3%) in thetreatment of acute bacterial conjunctivitis.J Drug Assess 1: 549558.

Wallerstein RO, Condit PK, Kasper CK,Brown JW & Morrison FR (1969): State-wide study of chloramphenicol therapy andfatal aplastic anemia. JAMA 208: 2045 2050.

Weiss A, Brinser JH & Nazar-Stewart V(1993): Acute conjunctivitis in childhood. JPediatr 122 : 1014.

Wiholm B-E, Kelly JP, Kaufman D, Issarag-risil S, Levy M, Anderson T & Shapiro S(1998): Relation of aplastic anaemia to useof chloramphenicol eye drops in two inter-national casecontrol studies. Br Med J328 : 666.

Wilhelmus KR (2005): Epidemiology of ocu-lar infections. In: Tasman M & Jaeger EA(eds) Duanes clinical ophthalmology, vol.2. Town: Lippincott, Williams & Wilkins,164.

Yamauchi Y, Minoda H, Yokoi K, Maruy-ama K, Kumakura S, Usui M, Cruz JM &

Fukutake K (2005): Conjunctival ora inpatients with human immunodeciencyvirus infection. Ocul Immunol Inamm 13:301304.

Yee RW, Tepedino M, Bernstein P, JensenH, Schiffmann R & Whitcup SM (2005): Arandimized, investigator-masked clinicaltrial comparing the efcacy and safety of gatioxacin 0.3% administered BID versusQID for the treatment of acute bacterialconjunctivitis. Curr Med Res Opin 21:425431.

Yunis A (1973): Chloramphenicol inducedbone marrow suppression. Semin Hematol10 : 225234.

Zinn CS, Westh H & Rosdahl VT (2004): Aninternational multicenter study of antimi-crobial resistance and typing of hospitalStaphylococcus aureus isolates from 21 lab-oratories in 19 countries or states. MicrobDrug Resist 10 : 160168.

Received on February 2nd, 2007.Accepted on June 10th, 2007.

Correspondence:Gunnar HvdingDepartment of Clinical MedicineSection of OphthalmologyUniversity of BergenHaukeland University HospitalN-5021 BergenNorwayTel: + 47 55 97 41 08Fax: + 47 55 97 41 43Email: [email protected]


17

Documents

Acute bacterial conjunctivitis.pdf