Randomised Controlled Trial of Ultrasonography in Diagnosis of Acute Appendicitis_incorporating the Alvarado Score-1

8/12/2019 Randomised Controlled Trial of Ultrasonography in Diagnosis of Acute Appendicitis_incorporating the Alvarado Scor

1/6

Papers

Randomised controlled trial of ultrasonography indiagnosis of acute appendicitis, incorporating theAlvarado scoreCharles D Douglas, Neil E Macpherson, Patricia M Davidson, Jonathon S Gani

Abstract

ObjectivesTo determine whether diagnosis bygraded compression ultrasonography improvesclinical outcomes for patients with suspected

appendicitis.DesignA randomised controlled trial comparingclinical diagnosis (control) with a diagnostic protocolincorporating ultrasonography and the Alvaradoscore (intervention group).SettingSingle tertiary referral centre.Participants302 patients (age 5-82 years) referred tothe surgical service with suspected appendicitis. 160patients were randomised to the intervention group,of whom 129 underwent ultrasonography.Ultrasonography was omitted for patients withextreme Alvarado scores (1-3, 9, or 10) unlessrequested by the admitting surgical team.Main outcome measuresTime to operation, durationof hospital stay, and adverse outcomes, includingnon-therapeutic operations and delayed treatment inassociation with perforation.ResultsSensitivity and specificity of ultrasonographywere measured at 94.7% and 88.9%, respectively.Patients in the intervention group who underwenttherapeutic operation had a significantly shortermean time to operation than patients in the controlgroup (7.0v10.2 hours, P = 0.016). There were nodifferences between groups in mean duration ofhospital stay (53.4v54.5 hours, P = 0.84), proportionof patients undergoing a non-therapeutic operation(9%v11%, P = 0.59) or delayed treatment inassociation with perforation (3%v1%, P = 0.45).ConclusionGraded compression ultrasonography is

an accurate procedure that leads to the promptdiagnosis and early treatment of many cases ofappendicitis, although it does not prevent adverseoutcomes or reduce length of hospital stay.

Introduction

Acute appendicitis is one of the commonest surgicalemergencies. Simple appendicitis can progress to per-foration, which is associated with a much highermorbidity and mortality, and surgeons have thereforebeen inclined to operate when the diagnosis isprobable rather than wait until it is certain. 1 A clinical

decision to operate leads to the removal of a normalappendix in 15% to 30% of cases (although the figuremay be higher or lower in certain demographicgroups).1 This proportion may be reduced by

observing equivocal cases for a period of time, a prac-tice that seems to be safe for most patients.2 Some casesof appendicitis may resolve spontaneously.3 4 None theless, if a period of observation culminates in thediagnosis of a ruptured appendix, the patient may havesuffered a poor outcome that was avoidable. Reduc-tions in the number of unnecessary or non-therapeutic operations should not be achieved at theexpense of an increase in number of perforations.5

It has been claimed that diagnostic aids candramatically reduce the number of appendicectomiesin patients without appendicitis, the number ofperforations, and the time spent in hospital. 1 Methodsadvocated to assist in the diagnosis of appendicitisinclude laparoscopy,6 7 scoring systems,8 9 computer

programs,10 ultrasonography,11 computed tomogra-phy,12 and magnetic resonance imaging.13 Imagingtechniques have been shown to be particularlyaccurate.14 Graded compression ultrasonography is theleast expensive and least invasive of these and has beenreported to have an accuracy of 71% to 95%,14 butdoubts have been raised about the influence ofultrasonography on patient outcomes.15 Furthermore,it has been argued that findings at sonography shouldnot supercede clinical judgment in patients with a highprobability of appendicitis.16 This raises questionsabout whether sonography should be performed at allin patients at high risk and whether there is some reli-able means of selecting those who can benefit from

imaging.The Alvarado score is a 10 point scoring system forthe diagnosis of appendicitis based on clinical signsand symptoms and a differential leucocyte count (seetable 5). In his original paper Alvarado recommendedan operation for all patients with a score of 7 or moreand observation for patients with scores of 5 or 6.8

Subsequent prospective studies have suggested thatthe Alvarado score alone is inadequate as a diagnostictest,17 18 but it has been advocated as a means of select-ing patients who should undergo imaging.19

We designed a diagnostic protocol incorporatinggraded compression ultrasonography and the

Department ofSurgery, JohnHunter Hospital,NSW 2310,Australia

Charles D Douglas

surgical registrar

Faculty of Medicineand HealthSciences, Universityof Newcastle,Callaghan NSW2308, Australia

Neil E Macphersonmedical student

Patricia M Davidsonassociate professor of

paediatric surgery

Jonathon S Ganisenior lecturer

Correspondence to:CD [email protected]

BMJ 2000;321:17

1BMJ VOLUME 321 14 OCTOBER 2000 bmj.com


2/6

Alvarado score on the basis of work in our own institu-tion.20 We then undertook a randomised controlledtrial to assess whether the information provided by theprotocol improved clinical outcomes. We tested thehypotheses that compared with standard treatmentpatients assigned to the diagnostic protocol wouldhave a shorter mean duration of hospital stay; a lowerrate of unnecessary (non-therapeutic) operations; ashorter mean time to surgery for those undergoing

therapeutic operations; and an equal or lower rate ofdelayed treatment in association with perforation.

Methods

Ethics committee approval was obtained for this trial.Patients were considered for inclusion in the study ifthey were referred to the surgical service at JohnHunter Hospital and John Hunter Childrens Hospitalwith a provisional diagnosis of acute appendicitisbetween October 1997 and October 1998. All generalsurgeons (six), paediatric surgeons (three), and theirregistrars (seven) at the participating hospital wereinvolved in the study.

Patients were excluded from randomisation if theyfulfilled any of the following criteria: age less than 5years; evidence of generalised peritonitis; palpablemass in the right iliac fossa; evidence of acuteconfusional state or dementia; graded compressionultrasonography already performed. All other patientswere met by the project officer, a third year medicalstudent, who explained the study and obtainedconsent. The project officer randomly allocatedpatients by coin toss to control (standard treatment) ordiagnostic protocol (intervention) groups. He organ-ised a leucocyte count and performed a structuredclinical assessment from which he calculated theAlvarado score (modified only by using percussiontenderness in the place of rebound tenderness).

For patients in the control group, members of theadmitting surgical team were not informed of theAlvarado score. They proceeded with appropriateclinical assessment and management. They wererequested not to organise graded compressionultrasonography for 36 hours.

For patients in the intervention group, the projectofficer advised the admitting team of the Alvaradoscore. Ultrasonography was then organised if theAlvarado score was between 4 and 8, inclusive. AnAlvarado score of 9 or 10 was taken to be a relativeindication for surgery, but the admitting team wasgiven the option of organising graded compressionultrasonography; patients with an Alvarado of 3 or lesswere not eligible for ultrasonography. The admitting

team was advised of the result of ultrasonographywhen this was done.

UltrasonographyGraded compression ultrasonography results weredesignated positive, negative, or equivocal by theattending sonographer by using the following criteria:positiveappendix identified, tender and non-compressible or appendiceal phlegmon or abscessseen; negativeappendix not identified, no otherrelevant abnormality seen; equivocalappendix notidentified but abnormal amount of free fluid seen withthickened, dilated, or non-peristaltic bowel in the

region of the caecum. In our experience these latterfindings are often associated with perforation, and wesuggested to participating surgeons and registrars thatit was safest to consider such a report as a positiveresult.

In the few cases when the appendix was identifiedbut was compressible or not tender we asked thesonographer to make a judgment on the basis of his orher experience and any other sonographic infor-

mation, including appendiceal dimensions and bloodflow. This reporting system was based on the results ofa prospective study at our hospital (unpublished).

Ultrasonography was unavailable at this institutionbetween the hours of 10 pm and 8 am, and patientsentered in the study between these times had theirexamination at 8 am, unless the admitting surgicalteam deemed an immediate operation to be necessary.

SurgeryAll patients who underwent laparotomy or laparos-copy for suspected appendicitis had an appendicec-tomy. The diagnosis of appendicitis was made onhistological grounds on the basis of infiltration of themuscularis propria by neutrophil granulocytes.

A patient was considered to have had an operationif laparotomy or laparoscopy was performed. Opera-tions were considered to be therapeutic if disease wasfound, when the disease seemed to be the cause for thepatients pain, and when surgery was the appropriatetreatment for that disease. All other operations wereclassed as non-therapeutic operations. Operationswere considered to be non-therapeutic if the onlyabnormality was a non-inflamed appendix containinga faecolith.

PerforationsThe appendix (or bowel) was considered to beperforated if the surgeon clearly identified a perfora-tion or a peritoneal swab grew at least one definite

bowel organism or the histopathologist identified aperforation in association with gangrene or full-thickness necrosis.

Delayed treatment in association with perforationFor patients with perforation, treatment was consid-ered to be delayed if surgery had not started within 10hours of randomisation.

Follow upPatients were reviewed at one week and three monthswith a pro-forma assessment. When direct review wasnot possible, details were obtained from the patientsgeneral practitioner or surgeon.

Outcomes

We identified four outcome measures.Time to operation for therapeutic operations was

defined as the time in hours from randomisation toskin preparation.

Duration of stay was defined as the time in hoursfrom randomisation to discharge from hospital. Whena patient was discharged and then readmitted forongoing management of a complication of acuteappendicitis the duration of the subsequent admissionwas added to the first.

Rate of non-therapeutic operationsThis was thenumber of non-therapeutic operations (see above) as aproportion of the total number in each group.

Papers

2 BMJ VOLUME 321 14 OCTOBER 2000 bmj.com


3/6

Rate of delayed treatment in association with perforationwas the number of cases of delayed treatment inassociation with perforation (as defined above) dividedby the total number in each group.

PowerThis sample had a power of 80% to detect a differencebetween groups of 3.3 hours for mean time to theatre,15.2 hours for mean duration of stay, and a reduction

in the non-therapeutic operation rate from 11% to 2%.

Data analysisData were analysed on an intention to treat basis. Forcalculation of sensitivity and specificity of graded com-pression ultrasonography we included cases only if ahistological diagnosis was available. Diagnoses otherthan appendicitis were ignored. Equivocal ultrasonog-raphy reports were counted as positive. Thus if theresults of graded compression ultrasonography werereported as positive or equivocal for appendicitis but aperforated diverticulum and normal appendix werefound at operation, the test was counted as a false posi-tive, even though the operation was consideredtherapeutic.

Two by two contingency tables were analysed byPearsons 2 test (or Fishers exact test when stated), andcomparisons of means were analysed by a two tailedttest. Confidence intervals for single proportions werecalculated with the Wilson procedure without correc-tion for continuity.21

Results

Figure 1 shows the trial profile. In total 306 patientswere referred for inclusion; two patients failed to meetinclusion criteria and two patients refused consent,thus 302 patients were enrolled in the study, with 160in the intervention group and 142 in the controlgroup. The mean age was slightly lower in the

intervention group (20.2 v 23.5 years); 202 patientswere aged 14 years and over and 100 were aged under14 years. There was little difference between groupswith respect to sex, mean Alvarado score, orproportion with Alvarado score greater than 6. Figure2 shows the distribution of Alvarado scores. Table 1summarises the results. Subgroup analysis by age isshown in table 2.

Sixteen patients were in breach of the trial protocolbecause the admitting surgeon in each case thoughtthat this was in the patients interests. All were includedin the reported analysis on an intention to treat basis.The results of a secondary analysis with these patientsexcluded were not substantially different with respectto adverse outcomes (intervention 17/154 (11%; 95%

confidence interval 7% to 17%) v control 16/132(12.1%; 8% to 19%); P = 0.8) or duration of stay (inter-vention 53.1 (46 to 60) hours vcontrol 53.0 (44 to 62)hours; P = 0.99).

UltrasonographyGraded compression ultrasonography was performedin 139 patients (see table 3). The sensitivity and specifi-city of ultrasonography for diagnosing appendicitiswas 94.7% and 88.9%, respectively. There were threefalse negative results. Six patients with a positive orequivocal result on ultrasonography recovered withoutsurgery.

SurgeryThere were 170 operations performed: 95 of the 160patients in the intervention group underwent surgerycompared with 75 of the 142 patients in the controlgroup (59.4%v52.8%, P = 0.25). Appendicitis was con-firmed histologically in 128 patients: 73 (45.6%; 38% to53%) in the intervention group and 55 (38.7%; 31% to47%) in the control group (P = 0.23). There were 13patients with other conditions that met the criteria fora therapeutic operation (see table 4). Twenty nine

operations were non-therapeutic: 14 (8.8%; 5.3% to14.2%) in the intervention group and 15 (10.6%; 6.5%to 16.7%) in the control group (P = 0.59).

PerforationsTwenty four patients had perforations, 19 had a perfo-rated appendicitis (14.8% (10% to 22%) of all cases of

Referred to surgical servicewith suspected appendicitis

n=306

Randomisedn=302

Intervention groupn=160

Alvarado 4-8n=106

(one incomplete score)

Alvarado 8n=50

Ultrasonographyn=101

(5 protocol breach)

2 patients lost to follow up at three months(both had had appendicectomy)

2 patients lost to followup at three months

(one appendicectomy,one negative result

on ultrasonography)

Ultrasonographyn=1

(protocol breach)

Ultrasonographyn=27

Ultrasonographyn=10

(all protocol breach)

Control groupn=142

Refusals n=2Exclusions n=2

Fig 1 Profile of trial of graded compression ultrasonography and Alvarado score comparedwith clinical methods in diagnosis of appendicitis

Table 1 Demographics and results according to allocation to diagnosis with graded

compression ultrasonography and Alvarado score or clinical diagnosis only. Results are

given with 95% confidence intervals unless stated otherwise

Intervention (n=160) Control (n=142)

No male 87 69

Mean age (years) 20.2 23.5

Alvarado score

Mean score 7.18 6.93

No (%) with score >6 99 (61.9) 86 (60.6)

Hospital data

Total No (%) of operations 95 (59.4; 52 to 67) 75 (52.8; 45 to 61)

No (%) of therapeutic operations 81 (50.6; 43 to 58) 60 (42. 3; 34 to 50)

Me an ti me t o t he ra pe uti c o pe ra ti on ( ho ur s) 7. 0 ( 5. 9 to 8. 1) 1 0. 2 ( 7. 9 t o 1 3)

Mean duration of stay (hours) 53.4 (47 to 60) 54.5 (46 to 63)

N o ( %) of no n- the ra pe uti c o pe ra ti on s 14 ( 8. 8; 5. 3 t o 14 ) 15 ( 10 .6; 6. 5 to 17 )

No (%) with delayed treatment in association withperforation

5 (3.1; 1.3 to 7.1) 2 (1.4; 0.4 to 5.0)

Total (%) adverse outcomes (delayed treatment inassociation with perforation and non-therapeuticoperations)

19 (11.9; 7.7 to 18) 17 (12.0; 7.6 to 18)

Papers



4/6

appendicitis) and five had other bowel perforations. Ofall perforations, 14 were in the intervention group and10 in the control group (perforations/number ingroup of 8.8% (5.3% to 14%) and 7.0% (3.9% to 12%),respectively, P = 0.58).

Delayed treatment in association with perforationThere were seven cases of delayed treatment in associ-ation with perforation (six cases of appendicitis andone of perforation of a caecal carcinoma). Five of thesewere in the intervention group and two were in the

control group (3.1% v1.4%, P = 0.45, Fishers exacttest).

Follow upThere were no readmissions with appendicitis duringthe follow up period. Two patients required readmis-sion for complications: one in the intervention groupfor drainage of an abscess and one in the controlgroup for an early small bowel obstruction. Ninepatients had minor wound infections diagnosed oneweek after discharge (4v5, P = 0.51, Fishers exact test).Four patients were lost to follow up at three months,three of whom had had their appendix removedduring their admission (and were therefore able to beanalysed for all end points); the fourth, in the control

group, had had a negative result on ultrasonographybefore discharge.

Discussion

We have confirmed the high sensitivity and specificityof graded compression ultrasonography in thediagnosis of appendicitis. All our patients whounderwent surgery after a positive result on ultra-sonography proved to have appendicitis. Patients withequivocal signs of appendicitis are usually admitted tohospital for a day or night of observation. If the resulton graded compression ultrasonography is positive,however, the surgeon can operate immediately. In ourstudy, this lead to a significant reduction in mean time

to therapeutic operation.As some cases of appendicitis seem to resolve with-

out surgery,3 4 however, graded compression ultra-sonography could lead to an increase in therapeuticoperations (by correctly diagnosing appendicitis inpatients who may have recovered during a period ofobservation). Our results are consistent with this, with ahigher proportion of therapeutic operations occurringin the intervention group, although the difference wasnot significant. This is despite the fact that Alvaradoscores were similarly distributed between groups.

The reduced time to operation in the inter ventiongroup did not result in a reduced duration of hospital

25

20

15

10

5

01

0.7 0.60.7

1.9

3.5 3.8

7.7

16.9

14.8 14.4

12.0

14.4

13.4

16.3

22.5 22.5

16.9

8.8

7.7

2 3 4 5 6 7 8 9 10

Alvarado score

Proportionofgroup(%)

Intervention

Control

Fig 2 Distribution of Alvarado scores in patients with suspectedappendicitis allocated to diagnosis with graded compressionultrasonography and Alvarado score (intervention group) or clinicaldiagnosis only (control)

Table 2 Primary outcomes in adults (>14 years) and children (14 years)

No of patients 94 108

Mean t ime to therapeut ic operation (hours) 7.1 (5.8 to 8.4) 9.7 (7.2 to 12) 0.068

Mean duration of stay (hours) 59.3 (49 to 69) 59.2 (49 to 69) 0.99No (%) of non-therapeutic operations 11 (11.7; 6.7 to 20) 12 (11.1; 6.5 to 18) 0.90

No (%) with delayed treatment in associationwith perforation

2 (2.1; 0.6 to 7.4) 1 (0.93; 0.2 to 5.1) 0.60*

Total (%) adverse outcomes (delayedtreatment in association with perforationand non-therapeutic operations)

13 (13.8; 8.3 to 22) 13 (12.0; 7.2 to 20) 0.70

Children (


5/6

stay. There was a trend towards shorter stays in theintervention group for patients undergoing therapeu-tic operations but because a larger proportion of thecontrol group was managed non-operatively (andtherefore discharged early) there was no differenceoverall.

Adverse outcomesThere are two outcomes that surgeons seek to avoid incases of suspected appendicitis. The first is anon-therapeutic operation. The second is delayed

treatment in a patient who is subsequently found tohave perforation (delayed treatment in association withperforation). In this study, the proportion of patients ineach group who had an adverse outcome (either anon-therapeutic operation or delayed treatment) wassimilar. The occurrence of a number of cases of delaywith perforation, despite a low rate of perforatedappendicitis (14.8%), suggests that rate of delayedtreatment in association with perforation is a moreappropriate measure of the consequences of delayeddiagnosis than overall perforation rate.

In the intervention group, many adverse outcomeswere associated with a clinical management decisionthat was at odds with a correct diagnosis under theintervention protocol.Of particular note is the fact that

11 patients had a non-therapeutic operation after anegative result on graded compression ultrasonogra-phy. If surgeons had relied on the results of the proto-col these unnecessary operations would have beenavoided. But what would have happened to otherpatients in the study?

The protocol, if allowed to override clinicaljudgment, would have led to operations in at least sixpatients who recovered without surgery and in twopatients who did in fact undergo non-therapeuticoperations (one with an Alvarado score of 10 and onewith an incorrect sonographic diagnosis of ovariancyst). More importantly, rigid adherence to themanagement indicated by the protocol would haverequired that no patient with a negative result could

have an operation. There were three patients withappendicitis who had a negative result on ultrasonog-raphy, and in each case the appendix was gangrenousor perforated. Had graded compression ultrasonogra-phy been relied on, these patients would have had anindefinite delay in treatment.

Use of the Alvarado score to select patients forimagingWe used the Alvarado score as an objective means ofstratifying patients according to risk so that those witha high or low probability of appendicitis need not haveunnecessary imaging. The Alvarado score (table 5) is

based on a simple and largely objective assessment thatrequires minimal clinical experience, neverthelessNEM had an intensive period of training before thestudy began, and we ensured that he was elicitingsymptoms and signs in a consistent and objective way.We believe that having one person perform the sameobjective assessment on all occasions was a reliable andreproducible method of risk stratification.

Whether the Alvarado score or some other form of

risk stratification is used, selection of patients for imag-ing is an issue that cannot be ignored. Had weperformed graded compression ultrasonography onall patients in the intervention group the results wouldprobably have been worse. We would still have hadthree false negative tests, all in patients with a gangre-nous or perforated appendicitis, and possibly more. Ofthe 31 patients in the intervention group who did notundergo graded compression ultrasonography therewere 23 with Alvarado scores of 9 or 10, three withscores of 3 or less, and five breaches of protocol(because the surgeon was not prepared to wait forultrasonography). Of these 31 patients, 27 underwentsurgery and 25 (93%) had a therapeutic operation, witha mean time to operation of 5.6 hours. Our diagnosticprotocol incorporating the Alvarado score was, if any-thing, safer, faster, and more accurate than gradedcompression ultrasonography alone,but it still failed toproduce better outcomes than unaided clinicaldiagnosis.

Possible biasesDistributions of Alvarado scores in each group weresimilar, and the proportions of each group with anAlvarado score of greater than 6 (that is, patients whowould be predicted to have appendicitis on the basis ofAlvarados original paper) were almost identical.Therefore the disparity between groups in number of

Table 5 Alvarado score for diagnosis of acute appendicitis

Symptom/sign/test Score

Migration of pain 1

Anorexia 1

Nausea-vomiting 1

Tenderness in right iliac fossa 2

Rebound pain 1

Raised temperature (>37.3C) 1

Leucocyte count>

10

10

9

/l 2Differential white cell count with neutrophils >75% 1

Total 10

What is already known on this topicUltrasonography is an accurate test for thediagnosis of acute appendicitis

Few studies have examined the effect of diagnosticultrasonography on clinical outcomes, and therehave been no randomised controlled trials

What this study adds

This study confirmed the accuracy ofultrasonography and found a reduction in meantime to operation for patients undergoingtherapeutic operation

There was no benefit of ultrasonography in termsof length of hospital stay, rate of non-therapeuticoperations, or rate of delayed treatment inassociation with perforation

False negative tests occurred in patients withgangrenous and perforated appendixes

Ultrasonography remains a test of unprovedbenefit and should not be used by those who areinexperienced in the clinical diagnosis ofappendicitis

Papers



6/6

therapeutic operations performed is unlikely to reflecta difference in disease prevalence.

To ensure that our results were not biased by thenon-availability of the imaging service to after 10 pm,we performed a secondary analysis on the 231 patientswho were enrolled in the study before this time. Thedifference in mean time to theatre was marginallyshort of significance and there was still no differencebetween groups in duration of hospital stay or in

adverse outcomes.

General commentsWhen performed by experienced sonographers,graded compression ultrasonography is an accuratetest. In this trial the accuracy was over 93%, equal tothat of computed tomography without coloniccontrast.14 False negative reports, however, do occur: inour study 5% of negative results were incorrect. Thereis no certain way of determining which negative resultis a false negative, and the consequences of not operat-ing may be serious. Patients cannot be safely sent homeafter a negative result unless there are also clinicalgrounds for their discharge. It is therefore inappropri-ate for graded compression ultrasonography to be

used by those who lack experience in the clinical diag-nosis of appendicitis.

ConclusionThe diagnosis of acute appendicitis aided by gradedcompression ultrasonography has not been shown toproduce better outcomes than clinical diagnosis alone.Further studies of graded compression ultrasonogra-phy and other diagnostic methods in suspected appen-dicitis should be randomised trials.

We thank Dr John Bear and Dr Jan Bishop for assistance withsonographic and histopathological diagnosis. We also thanksonographers Sue Mullen, Jenny Gosling, Warren Jones, BrettRoworth, and Darrin Gray for their excellent technicalassistance. In addition we thank Professor Michael Hensley, whoreviewed the proposal for study design and gave advice about

data interpretation and analysis, and Dr Brian Draganic, whogave statistical advice and reviewed and edited the drafted paper.

Contributors: CDD initiated the study, formulated the studyhypotheses, proposed the study design, analysed the data, wasthe principal author of the paper, and is the guarantor. NEMmanaged the running of the trial, contributed to study design,

collected the data, initiated and participated in data analysis, andhelped to write the paper. PMD initiated the research inultrasonography, supervised the running of the trial, facilitatedand coordinated involvement of different departments, contrib-uted to data interpretation, and helped to edit the paper. JSGcontributed to study design, supervised the running of the trial,contributed to data interpretation and analysis, and helped towrite and edit the paper.

Funding: None.Competing interests: None declared.

1 Hoffmann J, Rasmussen OO. Aids in the diagnosis of acute appendicitis.Br J Surg1989;76:774-9.

2 Jones PF. Active observation in the management of acute abdominal painin childhood.BMJ1976;ii:551-3.

3 Ooms HW, Koumans RK, Ho-Kang-You PJ, Puylaert JB. Ultrasono-graphy in the diagnosis of acute appendicitis. Br J Surg1991;78:315-8.

4 Heller MB, Skolnick ML. Ultrasound documentation of spontaneouslyresolving appendicitis.Am J Emerg Med1993;11:51-3.

5 Velanovich V, Savata R. Balancing the normal appendectomy rate withthe perforated appendicitis rate: implications for quality assurance. AmSurg1992;58:264-9.

6 Olsen JB, Myren CJ, Haahr PE. Randomized study of the value of lapar-oscopy before appendicectomy.Br J Surg1993;80:822-923.

7 Moberg AC, Ahlberg G, Leijonmarck CE, Montgomery A, Reiertsen O,Rosseland AR, et al. Diagnostic laparoscopy in 1043 patients withsuspected appendicitis.Eur J Surg1998;164:833-40.

8 Alvarado A. A practical score for the early diagnosis of acute append-icitis.Ann Emerg Med1986;15:557-64.

9 Teicher I, Landa B, Cohen M, Cabnick LS, Wise L. Scoring system to aidin the diagnosis of appendicitis. Ann Surg1983;198:753-9.

10 de Dombal FT, Leaper DJ, Staniland JR, McCann AP, Horrocks JC.Computer-aided diagnosis of acute abdominal pain. BMJ1972;ii:9-13.

11 Puylaert JBCM. Acute appendicitis: US evaluation using gradedcompression.Radiology1986;158:355-60.

12 Balthazar EJ. Appendicitis: prospective evaluation with high-resolutionCT.Radiology1991;180:21-4.

13 Incesu L, Coskun A, Selcuk MB, Akan H, Sozubir S, Bernay F. Acuteappendicitis: MR imaging and sonographic correlation. Am J Roentgenol1997;168:669-74.

14 Rao PM,Boland GWL. Imaging of acute right lower abdominal quadrantpain. Clin Radiol1998;53:639-49.

15 Ford R, Passinault W, Morse M. Diagnostic ultrasound for suspectedappendicitis: Does the added cost produce a better outcome? Am Surg1994;60:895-8.

16 Sivit C. Imaging children with acute right lower quadrant pain. PediatrClin North Am1997;44:575-89.

17 Ohmann C, Yang Q, Franke C. Diagnostic scores for acute appendicitis.Eur J Surg1995;161:273-81.

18 Macklin CP, Radcliffe GS, Merei JM, Stringer MD. A prospectiveevaluation of the modified Alvarado score for acute appendicitis in chil-dren.Ann R Coll Surg Engl1997;79:203-5.

19 Tobias M, Samuel E. Ultrasound and the Alvarado score. J R Soc Med

1992;85:585-6.20 Davidson PM, Douglas CD, Hosking CS, Graded compressionultrasonography in the assessment of the tough decision acuteabdomen in childhood.Pediatr Surg Int1999;15:32-5.

21 Newcombe R. Two-sided confidence intervals for the single proportion:comparison of seven methods.Stat Med1998;17:857-72.

(Accepted 18 May 2000)

Papers

6 BMJ VOLUME 321 14 OCTOBER 2000 bmj.com

Documents

Randomised Controlled Trial of Ultrasonography in Diagnosis of Acute Appendicitis_incorporating the Alvarado Score-1