|

 

British medical journal

Primary care based randomised, double blind trial
of amoxicillin versus placebo for acute otitis
media in children aged under 2 years
Roger A M J Damoiseaux, Frank A M van Balen, Arno W Hoes, Theo J M Verheijand Ruut A de Melker 2000;320;350-354 Updated information and services can be found at: These include: This article cites 19 articles, 7 of which can be accessed free at: 37 online articles that cite this article can be accessed at: 14 rapid responses have been posted to this article, which you can access forfree at: You can respond to this article at: Receive free email alerts when new articles cite this article - sign up in the box at the top left of the article Articles on similar topics can be found in the following collections To order reprints follow the "Request Permissions" link in the navigation box Primary care based randomised, double blind trial of
amoxicillin versus placebo for acute otitis media in
children aged under 2 years
Roger A M J Damoiseaux, Frank A M van Balen, Arno W Hoes, Theo J M Verheij, Ruut A de Melker
otitis media in the Netherlands does not seem to be any General Practice, worse than that in other countries.1 University Medical Objective To determine the effect of antibiotic
Several authors have advocated restriction of treatment for acute otitis media in children between 6 antibiotic treatment for acute otitis media to children at months and 2 years of age.
increased risk of poor outcome or complications,5 8 Design Practice based, double blind, randomised,
notably children under 2 years of age,4 9–14 although, placebo controlled trial.
surprisingly, there is virtually no empirical evidence as Setting 53 general practices in the Netherlands.
to the effectiveness of such treatment in these Subjects 240 children aged 6 months to 2 years with
children.15 We therefore assessed outcome in a primary the diagnosis of acute otitis media.
care based randomised trial of amoxicillin versus Intervention Amoxicillin 40 mg/kg/day in three
professor of general Main outcome measures Persistent symptoms at day
four and duration of fever and pain or crying, or both.
Ruut A de Melkerprofessor of general Otoscopy at days four and 11, tympanometry at six weeks, and use of analgesic.
The study was conducted between February 1996 and Results Persistent symptoms at day four were less
May 1998 in the Netherlands, where all patients are common in the amoxicillin group (risk difference and Julius Centre treated initially by their own general practitioner. Chil- for Patient Oriented 13%; 95% confidence interval 1% to 25%). The dren aged between 6 and 24 months were eligible if Research, University median duration of fever was two days in the they presented with acute otitis media—defined as amoxicillin group versus three in the placebo group infection of the middle ear of acute onset and a charac- professor of clinical (P = 0.004). No significant difference was observed in teristic ear drum picture (injection along the handle of duration of pain or crying, but analgesic consumption the malleus and the annulus of the tympanic Correspondence to: was higher in the placebo group during the first 10 membrane or a diffusely red or bulging ear drum)—or days (4.1 v 2.3 doses, P = 0.004). In addition, no acute otorrhoea. In addition, one or more symptoms of otoscopic differences were observed at days four and acute infection (fever, recent earache, general malaise, 11, and tympanometric findings at six weeks were recent irritability) had to be present, in line with the similar in both groups.
Dutch guidelines.8 Conclusions Seven to eight children aged 6 to 24
The following exclusion criteria were applied: anti- months with acute otitis media needed to be treated biotic treatment in the preceding four weeks; proved with antibiotics to improve symptomatic outcome at allergy to amoxicillin; compromised immunity; cranio- day four in one child. This modest effect does not facial abnormalities; Down's syndrome; or being justify prescription of antibiotics at the first visit, entered in this study before. The 53 participating gen- provided close surveillance can be guaranteed.
eral practitioners were trained to classify ear drums byusing a standard set of slides depicting a range of com-mon ear drum appearances with the emphasis on dis- criminating between acute otitis media and otitismedia with effusion.16 The study protocol was Antibiotics are currently the treatment of choice for approved by the ethical committee of the Children's acute otitis media in nearly all countries,1 which is rather Hospital of the University Medical Centre Utrecht; surprising as their effectiveness seems limited in terms of and all parents of the children gave written informed clinical improvement.2–7 Although the worldwide crisis consent before enrolment.
of multiple resistant strains of microbes underlines theimportance of the prevention of overuse and misuse of antibiotics, the Netherlands is still the only country Patients received either amoxicillin suspension 40 where only a minority of the episodes of acute otitis mg/kg/day in three divided doses for 10 days or media are treated with antibiotics.1 The outcome of acute placebo suspension. Most patients in the Netherlands BMJ VOLUME 320 5 FEBRUARY 2000 www.bmj.com
with acute otitis media receive decongestant nose Kaplan-Meier curves, and differences between the treat- drops, so all patients received one drop of oxymetazo- ment groups were tested by the log rank test. When line 0.025% in each nostril three times a day (Nasivin, diary data were incomplete and the last entry recorded Merck) for seven days. The use of paracetamol was fever or pain, the child was censored in the survival allowed when the child was in pain, the amount being analysis. The difference in the mean analgesic consump- recorded in the diary. For each dose children under 1 tion in the two groups was tested with the Mann- year old received a 120 mg suppository and older chil- Whitney U test. All reported P values are two sided.
dren received 240 mg.
To adjust for possible confounding due to unequal At the baseline visit the doctor recorded the history, distribution of baseline characteristics we used logistic the presence or absence of certain risk factors for acute regression analysis with the primary outcome measure otitis media, and the results of otoscopy. Parents were as the dependent variable.
instructed to keep a 10 day diary showing occurrenceof aural and gastrointestinal symptoms and adminis- Assignment and blinding
tration of study medication, paracetamol, and nose After we obtained consent the children were randomly drops. Follow up visits were scheduled on days four assigned to treatment with amoxicillin or with a and 11 at the general practitioner's clinic, inquiry was placebo suspension with the same colour and taste.
made about remaining symptoms, and the ear drum The suspensions were supplied to the participating was examined. At six weeks all children were visited at doctors in a double blind fashion with computerised home by the first author (RD), and information was two block randomisation; doctors, parents, and investi- obtained about present and past symptoms, antibiotic gators remained blinded throughout the study. During use, and any referral to a paediatrician or otolaryn- the trial the code of the allocation schedule was kept in gologist since day 11. Further otoscopy and tympan- the pharmacy of the University Medical Centre, ometry was also carried out.
Utrecht, and was broken only if severe complicationsor side effects occurred.
Outcome measures
The primary outcome measure was persistent symp-
toms at day four, assessed by the doctor and defined aspersistent earache, fever (>38oC), crying, or being irri- Participant flow and follow up—Of the 425 children with table. In addition the prescription of another antibiotic acute otitis media registered, 362 were eligible, and because of clinical deterioration before the first follow from these 240 were randomly assigned to one of the up visit was to be considered a persistent symptom.
treatment groups (figure). Children in the antibiotic Secondary outcome measures were clinical treat- group and the placebo group differed in the ment failure at day 11, defined as persistent fever, ear- prevalence of recurrent acute otitis media, regular ache, crying, being irritable, or no improvement in the attendance at a day care centre, and parental smoking appearance of the tympanic membrane, defined as habits (table 1).
persistent redness, bulging, or perforation of one or Outcome at day four—Persistent symptoms at day both tympanic membranes; the duration of fever four occurred in 69 out of 117 children (59%) in the (>38oC), pain, or crying, defined as the number of days amoxicillin group and in 89 of 123 (72%) in the until the first day on which these signs were considered placebo group (difference in risk 13%; 95% confidence absent and remained absent as recorded in the diary by interval 1% to 25%) (table 2). Adjustment for the parents; the mean number of doses of analgesics recurrence, day care, and smoking as possible given, based on the diaries; adverse effects mentioned confounders in the logistic regression analyses showed in the diaries; and the percentage of children with mid- an odds ratio of 1.79 (1.03 to 3.13). Among children dle ear effusion at six weeks. The diagnosis of effusion with persistent symptoms four (one in the amoxicillin was based on combined otoscopy and tympanometry.
group and three in the placebo group) received other Type B and C2 tympanograms (modified Jerger's antibiotics. Three of these children were admitted to classification) were regarded as indicative of the hospital (one in the amoxicillin group, two in the presence of fluid in the middle ear.17 18 placebo group); one (placebo group) was admitted onthe third day with meningitis but because of deteriora- Sample size and data analysis
tion this child had already been started on another Calculation of the sample size was based on the antibiotic on day two. The culture of cerebrospinal fluid assumption of a minimum difference of 20% in yielded negative results, but the Gram stain suggested primary outcome between the groups, with an á of 5%, streptococcal meningitis. The two other children were a discriminating power of 80%, and an estimated 60% admitted because of dyspnoea (amoxicillin group) and persistent symptoms in the placebo group.4 The total dehydration (placebo group). All four recovered number of children required in each treatment arm without residual symptoms. Inclusion of the one child lost to follow up (amoxicillin group) in either outcome All analyses were carried out with SPSS on an inten- group did not materially change the findings.
tion to treat basis. We performed best and worst case Outcome at day 11—Clinical treatment failure at day analyses when necessary because of loss to follow up.
11 occurred in 72 out of 112 children (64%) in the The prevalence of persistent symptoms at day four, amoxicillin group and in 84 of 120 (70%) in the clinical treatment failure at day 11, occurrence of middle placebo group (6%; − 6% to 18%) (table 2). Eleven ear effusion at six weeks, and side effects in the two children received other antibiotics (three in the amoxi- groups were compared by calculating differences in risk cillin group, eight in the placebo group) and were with 95% confidence intervals. Durations of fever and of recorded as treatment failures. One of these children pain or crying, or both, were plotted by means of (placebo group) needed admission to the hospital BMJ VOLUME 320 5 FEBRUARY 2000 www.bmj.com
Patients with acute otitis media (n=425) Not eligible (n=63) Not randomised (n=122) Patients randomised Not fulfilling inclusion criteria Exclusion criteria valid Infant needed antibiotic, in blocks of 2 (n=240) according to doctor (n=27) Refused consent (n=46) Diagnosis in doubt (n=1) Entered study before (n=18) Logistical reasons (for example Complaints present for Received antibiotics in concerning follow up) (n=49) > 3 weeks (n=6) preceding 4 weeks (n=29)Allergic to amoxicillin (n=7)Belong to risk group (n=2) Received other antibiotics (n=1) Patients receiving amoxicillin Patients receiving placebo Received other antibiotics (n=3) Lost to follow up (n=1) as allocated (n=117) as allocated (n=123) Received other antibiotics (n=3) Patients evaluated Patients evaluated Received other antibiotics (n=8) Lost to follow up (n=3) Lost to follow up (n=3) Lost to follow up (n=2) Patients evaluated Patients evaluated Had grommets inserted (n=1) Lost to follow up (n=3) Patients evaluated Patients evaluated Trial profile and participant flow because of deterioration of symptoms of acute otitis Duration of fever and pain or crying—The median media. Six children (three in each group) were lost to time to cessation of fever was two days with amoxicillin follow up between day four and day 11, and in one case and three days with placebo (P = 0.004; log rank test).
(amoxicillin group) the evaluation of the ear drum was Median time to cessation of pain or crying was eight missing, although the symptoms were gone. A best case days with amoxicillin and nine days with placebo scenario (amoxicillin group analysed as "cured" and (P = 0.432; log rank test).
placebo group as "not cured") did not show any signifi- Analgesic consumption—During the first three days, cant difference in clinical treatment failure at day 11 mean analgesic consumption in the amoxicillin group (9%; − 3% to 21%).
was 1.7 doses and in the placebo group 2.5 doses(P = 0.018). Over the whole 10 days these figures were2.3 and 4.1, respectively (P = 0.004).
Table 1 Baseline characteristics of 240 children randomised in
Outcome at six weeks—At six weeks 212 children trial of antibiotic use for treatment of acute otitis media. Figures were examined. Middle ear effusion was present in are numbers of children except for mean age 69/107 (64%) in the amoxicillin group and in 70/105 (67%) in the placebo group (3%; − 10% to 16%). The proportion of children with bilateral effusion was 48% Mean age (months) in both groups. In addition, no clear differences were observed between the two groups as regards recurrent Breastfed for >6 months acute otitis media, use of antibiotics in this period, >2 children in family referrals to the otolaryngologist or paediatrician, or Season of inclusion (Oct-March) Smoking in household Attendance at day care centre Adverse effects—De novo diarrhoea was reported on day four in 17% (20/117) of the amoxicillin group and in 10% (12/123) of the placebo group (difference Recurrent AOM in family − 7%; − 16% to 2%). On day 10 these figures were 12% (14/117) and 8% (10/123), respectively (difference − 4%; − 12% to 4%). Of the children lost to follow up, Clinical presentation: five were withdrawn (all between day four and day 11) >3 days complaints because of possible side effects, two because of diarrhoea (both in the amoxicillin group) and three because of skin rashes (all in the placebo group).
Compliance—According to the diaries the mean number of doses of study medication taken was 24.6(82% of possible total) in the amoxicillin group and URTI=upper respiratory tract infection.
AOM=acute otitis media.
23.2 (76%) in the placebo group (P = 0.9). According to BMJ VOLUME 320 5 FEBRUARY 2000 www.bmj.com
the suspension remaining in the returned bottles, 80% Table 2 Main outcome measures in infants with acute otitis media randomised to
of the children in both groups had received the full receive amoxicillin or placebo amount, and 95% received at least 80% of the amountprescribed.
Difference in %
Persistent symptoms* at day 4 No improvement in eardrum at day 4 In this study resolution of symptoms on day four was Clinical treatment failure‡ at day 11 more common in those treated with amoxicillin than Median duration of fever (days) in those taking placebo. At day 11, no significant differ- Median duration of pain/crying (days) ences in symptoms and otoscopy results were Mean consumption of analgesia in first 10 observed. Amoxicillin shortened the duration of fever by one day, and analgesics were used more often in the *Defined as still having earache or having fever, crying, being irritable, or having received other antibiotics.
†÷2 test.
placebo group.
‡Defined as still having symptoms or no improvement, or both, in tympanic membrane.
The significant reduction of the duration of fever §Log rank test for Kaplan-Meier plot.
observed in the amoxicillin group is in accord with the ¶Mann-Witney U test.
results of Burke et al in children aged 3 to 10 years.2 Weobserved no difference between the two groups in pain The treatment regimen we used (amoxicillin 40 or crying. This was also reported by Burke et al, and the mg/kg/daily) is still the treatment of first choice.22 The amounts of analgesics taken in their study were dosage was deemed sufficient because incidences of comparable with those in ours.2 The fact that more resistant Streptococcus pneumonia and Haemophilus influ- analgesics were used in the placebo group could enzae in the Netherlands remain low at < 1 %23 and 6% explain the lack of difference in duration of pain or (data on file 1998, Dutch National Institute of Public Health and Environmental Protection), respectively, The number of children with persistent symptoms and compliance in this study was good.
in our study was high compared with other studies.2 19 As primary outcome measure we combined earache, Burke et al, however, included only older children and crying, and irritability because in these little children it is in young children symptoms are often prolonged.4 9 difficult to establish earache as such. We have shown that Contrary to our results complete resolution of seven to eight children aged 6 to 24 months with acute symptoms was not mentioned by Kaleida et al.19 otitis media needed to be treated to improve Our diagnoses were based on acute signs of symptomatic outcome at day four in one child. This is infection and abnormality of the ear drum; this has not sufficiently important clinically to prescribe antibiot- shown to be adequate in other studies3 4 and is in ics for every affected child within this age group. Routine accord with day to day practice in the Netherlands. An prescription of antibiotics would not prevent all cases of abnormal ear drum had to be seen because diagnosis meningitis.24 Our conclusion is that watchful waiting at based only on symptoms is not specific.20 According to the first visit is justified for these children. Instead of the baseline characteristics the results in our sample antibiotics analgesics could be given for proper are generalisable to the population seen in primary resolution of symptoms but more research is needed as care in the Netherlands.1 21 to whether this is a good alternative.
We thank all the general practitioners who included patients for What is already known about this
Funding: Netherlands Organisation for Scientific Research (grant no 904-58-074).
Several meta-analyses have shown that the Competing interests: Nasivin nose drops for this study were effectiveness of antibiotics for acute otitis media is donated by E Merck Nederland BV.
Contributors: RAMJD was responsible for the planning of limited in terms of clinical improvement the study, data collection and analysis. FAMvB, and RAdMdesigned the protocol and were the supervisors of RAMJD.
For children under 2 years of age—a risk group AWH and TJMV assisted with the analysis of the results. The with regard to poor outcome—the evidence of the manuscript was prepared by RAMJD and commented on by all effectiveness of antibiotics for this common authors. RAMJD is the study guarantor.
condition is not conclusive Froom J, Culpepper L, Grob P, Barteld A, Bowers P, Bridges-Webb C, et What this paper adds
al. Diagnosis and antibiotic treatment of acute otitis media: report fromInternational Primary Care Network. BMJ 1990;300:582-6.
Burke P, Bain J, Robinson D, Dunleavy J. Acute red ear in children: con- This randomised study shows that seven to eight trolled trial of non-antibiotic treatment in general practice. BMJ children, aged 6 to 24 months, with acute otitis Van Buchem FL, Dunk JHM, van't Hof MA. Therapy of acute otitis media: media need to be treated with amoxicillin to myringotomy, antibiotics, or neither? A double blind study in children.
improve symptomatic outcome at day four in one Appelman CLM, Claessen JQPJ, Touw-Otten FWMM, Hordijk GJ, de Melker RA. Co-amoxiclav in recurrent acute otitis media: placebocontrolled study. BMJ 1991;303:1450-2.
This is not sufficiently important clinically to Froom J, Culpepper L, Jacobs M, de Melker RA, Green LA, van BuchemFL, et al. Antimicrobials for acute otitis media? A review from the Inter- prescribe antibiotics for every child with acute national Primary Care Network. BMJ 1997;315:98-102.
otitis media in this age group Rosenfeld RM, Vertrees JE, Carr J, Cipolle RJ, Uden DL, Giebink GS, et al.
Clinical efficacy of antimicrobial drugs for acute otitis media:meta-analysis of 5400 children from thirty-three randomized trials. J Watchful waiting at the first visit is therefore justified for these children Del Mar C, Glasziou P, Hayem M. Are antibiotics indicated as initial treat-ment for children with acute otitis media? A meta-analysis. BMJ1997;314:1526-9.
BMJ VOLUME 320 5 FEBRUARY 2000 www.bmj.com
Appelman CLM, Bossen PC, Dunk JHM, Lisdonk EH, de Melker RA, van 17 Jerger J. Clinical experience with impedance audiometry. Arch Otolaryngol Weert HCPM. NHG standard otitis media acuta. (Guideline on acute oti- tis media of the Dutch College of General Practitioners.) Huisarts Wet 18 Zielhuis GA, Heuvelmans-Heinen EW, Rach GH, van den Broek P. Envi- ronmental risk factors for otitis media with effusion in preschool Hoberman A, Paradise JL, Burch DJ, Valinski WA, Hedrick JA, Aronovitz children. Scand J Prim Health Care 1989;7:33-8.
GH, et al. Equivalent efficacy and reduced occurrence of diarrhea from a 19 Kaleida PH, Casselbrant ML, Rockette HE, Paradise JL, Bluestone CD, new formulation of amoxicillin/clavulanate potassium for treatment of Blatter MM, et al. Amoxicillin or myringotomy or both for acute otitis acute otitis media in children. Pediatr Infect Dis J 1997;16:463-70.
media: results of a randomized clinical trial. Pediatrics 1991;87:466-74.
10 Laxdal OE, Merida J, Jones RHT. Treatment of acute otitis media: a con- 20 Kontiokari T, Koivunen P, Niemela M, Pokka T, Uhari M. Symptoms of trolled study of 142 children. Can Med Assoc J 1970;102:263-8.
11 Mandel EM, Casselbrant ML, Rockette HE, Bluestone CD, Kurs-Lasky M.
acute otitis media. Pediatr Infect Dis J 1998;17:676-9.
Efficacy of 20- versus 10-day antimicrobial treatment for acute otitis 21 Bruijnzeels MA, van Suijlekom-Smit LWA, van der Velden J, van der media. Pediatrics 1995;96:5-13.
Wouden JC. The child in general practice. Dutch national study of morbidity 12 Alho OP, Laara E, Oja H. What is the natural course of recurrent acute and interventions in general practice. Utrecht: NIVEL, 1993.
otitis media in infancy? J Fam Pract 1996;43:258-64.
22 Berman S. Otitis media in children. N Engl J Med 1995;332:1560-5.
13 Iino Y, Nakamura Y, Koizumi T, Toriyama M. Prognostic factors for per- 23 Hermans PWM, Sluijter M, Elzenaar K, van Veen A, Schonkeren JJM, sistent middle ear effusion after acute otitis media in children. Acta Nooren FM, et al. Penicillin-resistant Streptococcus pneumoniae in the Otolaryngol (Stockh) 1993;113:761-5.
Netherlands: results of a 1-year molecular epidemiologic survey. J Infect 14 Hathaway TJ, Katz HP, Dershewitz R, Marx TJ. Acute otitis media: who needs posttreatment follow-up? Pediatrics 1994;94:143-7.
24 Rothrock SG, Harper MB, Green SM, Clark MC, Bachur R, McIlmail DP, 15 Damoiseaux RAMJ, van Balen FAM, Hoes AW, de Melker RA. Antibiotic et al. Do oral antibiotics prevent meningitis and serious bacterial treatment of acute otitis media in children under two years of age: infections in children with streptococcus pneumoniae occult bacteremia? evidence based? Br J Gen Pract 1998;48:1861-4.
A meta-analysis. Pediatrics 1997;99:438-44.
16 Wormald PJ, Browning GG, Robinson K. Is otoscopy reliable? A structured teaching method to improve otoscopic accuracy in trainees.
Clin Otolaryngol 1995;20:63-7.
(Accepted 11 November 1999) General practice—time for a new definition
Frede Olesen, Jim Dickinson, Per Hjortdahl
Editorial by Heath After some 30 years of medical development focusing largely on hospitals, organ specialisation, and high Research Unit for technology, the 1960s saw the revitalisation of general General Practice, practice (in some countries known as family practice— It is time to create a new definition of general Aarhus, Vennelyst see note at the end of the article), allowing this field of practice based on the ideal content of the specialty medicine to develop into a cornerstone of the DK-8000 Aarhus C,Denmark healthcare system. At the end of the millennium Any new definition should describe the core academic general practice is now established in all content and function of general practice and developed countries. General practice is recognised as should supplement the description of the medical a special discipline and in many countries as a medical specialty equally important as and complementary to Family Medicine,Chinese University other specialties. Participation in specific training pro- It should also be universal, not country specific grammes has therefore become mandatory for anyone who wishes to become a specialist in this field.
It should provide a framework for teaching and The time has come to reflect on the education of training and describe where to find evidence to general practitioners, which depends on the scientific support science based work Department ofCommunity and content of general practice. A proper starting point is to consider the ideal content of the discipline as it A new proposed definition fulfils these criteria, University of Oslo, would be described in a definition.
emphasising the need for general practitioners to be able to take a biomedical, psychological, and social approach to patients and their problems Old definitions of general practice
Correspondence to:F Olesen Many definitions of primary care and general practice have been proposed.1–5 One of the most frequently general practice gave prominence to systems, settings, quoted is the Leeuwenhorst definition from 1974: and methods creating opportunities for good general "The general practitioner is a licensed medical graduate who gives personal, primary and continuing We question whether these dimensions in them- care to individuals, families and a practice population selves distinguish between those doctors who are gen- irrespective of age, sex and illness. It is the synthesis of eral practitioners and those who are not. Not only that, these functions which is unique."5 Other definitions but they may hamper change and promote failure.
also focus on the patient as an individual in a family They are rooted in a model of long term, full time, year and cultural context, continuity of care, and the round service in a stable community. Although sustained relationship with patients. Hence most rewarding for the doctor, this is a difficult role, with textbooks describe the particulars of the general prac- high personal and social costs for doctors in a modern titioner in terms of working methods such as continu- ity, comprehensiveness, work in a society (or even in We contend that many definitions confuse the patients' homes), a family approach, and good setting with the role and the person. They do not help communication.6 Early definitions and descriptions of us in defining the academic agenda for universities or BMJ VOLUME 320 5 FEBRUARY 2000 www.bmj.com

Source: http://www.ibe.med.uni-muenchen.de/lehre/lehrveranst/lehr_a/modul1-l5/material/archiv/archiv_ss09/damoiseaux_bmj_2000.pdf