The clinical diagnosis of acute bacterial rhinosinusitis in general practice and its therapeutic consequences

https://doi.org/10.1016/S0895-4356(02)00590-5Get rights and content

Abstract

In a randomized double-blind trial 251 adults with sinusitis-like symptoms were given amoxcillin/clavulanate or placebo for 6 days. Seven diagnostic indicators for acute bacterial rhinosinusitis are compared by their accuracy assuming a latent class model and by the treatment effect that they would have had if used to select a subset of patients for antibiotic treatment. Under a latent class model, radiography is a more efficient indicator then C reactive protein (CRP), which is, in turn, more efficient than other clinical signs and symptoms. However, a history of purulent nasal discharge, and signs of pus in the nasal cavity and throat, are better criteria than radiography or CRP for selecting those patients who will benefit from antibiotic treatment. These contradictory results are a salutary reminder that diagnostic indicators need to be evaluated in terms of therapeutic consequences for the patient.

Introduction

Acute bacterial rhinosinusitis is a common reason for prescribing antibiotics [1]. Its signs and symptoms are “extraordinarily similar” to those of an acute viral upper respiratory tract infection, and so viral cases are frequently misdiagnosed as bacterial [2]. Using antibiotics to treat viral infections is an unnecessary cost to the health care system and contributes to increasing bacterial resistance to antibiotics [3]. On the other hand, if the patient has acute bacterial rhinosinusitis, antibiotic treatment may shorten time to recovery and avoid rare complications [4], [5]. Guidelines advocate prescribing antibiotics only where acute maxillary sinusitis has been diagnosed by radiography or by sinus puncture [6] or where symptoms are severe [2].

To confirm a diagnosis of acute maxillary sinusitis requires maxillary sinus puncture, aspiration, and bacterial culture [2], [4]. Typically the final step of bacterial culture is omitted, and diagnosis is confirmed on finding purulent secretions [7]. There is at least indirect evidence that this procedure is suitable as a gold standard for maxillary sinusitis [8]. But the procedure is too painful, too inconvenient, and too costly for routine use, and it provides little information about sinuses other than the maxillary [9], [10].

Two recent meta-analyses compare diagnostic tests for acute bacterial rhinosinusitis [7], [11]. Both conclude that sinus puncture is the most accurate test, and that radiography is more accurate than ultrasound. They reach different conclusions on the value of a clinical examination. One concludes that a clinical examination is not reliable [11]; the other that a clinical examination and particularly risk scores are useful [7]. However, physicians are unlikely to use imaging tests regardless of their accuracy, if these tests are costly or inconvenient, and so effort is needed to perfect simple diagnostic tools [7]. To date, diagnostic tests for acute bacterial rhinosinusitis have not been evaluated in terms of their therapeutic consequences [11].

We report on the accuracy of clinical signs and symptoms and simple tests suitable for use in general practice. We consider the therapeutic consequences of these diagnostic indicators using data from a randomized clinical trial where adults with symptoms of acute bacterial rhinosinusitis were recruited in a general practice setting and given either an antibiotic therapy or placebo.

Section snippets

The clinical trial

Over four winter seasons, 251 patients aged 18 or over were recruited at 26 general practices in the northwest of Switzerland and at two outpatient clinics at the Basel University Hospital for a randomized placebo-controlled double-blind trial [12]. Those recruited had to have a history of both purulent nasal discharge and unilateral or bilateral frontal or maxillary pain for at least 48 h but less than 4 weeks. Initially patients had to show signs of purulent nasal discharge under rhinoscopy,

Indicator accuracy

Data for all seven indicators were available for 239 of the 251 patients in the trial. The two class latent model suggests that of the seven indicators, two should be dropped from further latent variable modeling. Both cacosmia and a history of purulent nasal discharge appear to have no predictive power (Table 1). Both have positive predictive values (0.15 and 0.14, respectively) below the estimated latent prevalence of the disease (0.18), and negative predictive values (0.75 and 0.79) below

Discussion

The prevalence of bacterial rhinosinusitis for adult patients recruited in a general practice setting is estimated from sinus puncture to be around 50% [7]. Our estimate of latent prevalence is 28% based on the five indicator data set. In a similar clinical trial, the adjusted hazards ratio for the effect of antibiotics on time to a “resumption of daily activities” was 1.31 (95% confidence interval 0.96–1.78) [30]. In this trial, the adjusted hazards ratio for the effect of antibiotics on time

Conclusions

This study sides with those who have faith in the clinical examination. A history of purulent nasal discharge and pus in the nose and throat are better criteria for selecting adult patients for antibiotic treatment than radiography or a C reactive protein test. The patient is not interested so much in diagnosis but in the treatment of their condition. Some patients will benefit from antibiotics even though they do not appear to have the disease, and some will not benefit even though they do

Acknowledgments

We thank the patients, general practice and outpatient physicians, and radiologists who took part in the original trial; Christian Schindler, who helped design the trial; Hans-jörg Züst, who helped recruit patients; and the Journal's reviewers. We thank John Uebersax for help with the two-latent trait model. His Web site on latent class models http://ourworld.compuserve.com/homepages/jsuebersax) was most influential. This study was supported by grants from Santésuisse and the Gottfried und

References (40)

  • Williams JW Jr, Aguilar C, Makela M, Cornell J, Holleman DR, Chiquette E, Simel DL. Antibiotics for acute maxillary...
  • M. Lindbaek et al.

    Use of symptoms, signs, and blood tests to diagnose acute sinus infections in primary care: comparison with computed tomography

    Fam Med

    (1996)
  • D. Gatland

    Predicting acute maxillary sinusitis. Trial was wrong to ignore other sinuses

    BMJ

    (1995)
  • Bucher HC, Tschudi P, Young J, Périat P, Welge-Lüssen A, Züst H, Schindler C. Effect of amoxicillin/clavulanate in...
  • N.P. van Duijn et al.

    Use of symptoms and signs to diagnose maxillary sinusitis in general practice: comparison with ultrasonography

    BMJ

    (1992)
  • J.W. Williams et al.

    Clinical evaluation for sinusitis. Making the diagnosis by history and physical examination

    Ann Intern Med

    (1992)
  • J.G. Hansen et al.

    Predicting acute maxillary sinusitis in a general practice population

    BMJ

    (1995)
  • J.W. Williams et al.

    Does this patient have sinusitis? Diagnosing acute sinusitis by history and physical examination

    JAMA

    (1993)
  • D.E. Low et al.

    A practical guide for the diagnosis and treatment of acute sinusitis

    CMAJ

    (1997)
  • B.S. Dahler-Eriksen et al.

    Evaluation of a near-patient test for C-reactive protein used in daily routine in primary healthcare by use of difference plots

    Clin Chem

    (1997)
  • Cited by (32)

    • Microbiology of acute sinusitis in Mexican patients

      2006, Archives of Medical Research
      Citation Excerpt :

      Resolution of symptoms by day 7 occurred in 73 and 47% of patients, respectively (11). Current antibiotic regimens consist of amoxicillin/clavulanic acid, a quinolone or a second-generation cephalosporin (12). Their judicious use should be observed to decrease excess antibiotic use in order to prevent the emergence of antibiotic-resistant S. pneumoniae and other resistant organisms (13).

    • Maxillary sinusitis

      2005, EMC - Oto-Rhino-Laryngologie
    • Antibiotics for adults with clinically diagnosed acute rhinosinusitis: a meta-analysis of individual patient data

      2008, The Lancet
      Citation Excerpt :

      In our analysis, we assessed the prognostic value of any sign or symptom recorded in at least four trials. Although point-of-care tests might have prognostic value, C-reactive protein concentrations were recorded in only one trial.16 We used random-effects models for the odds ratio (OR) for cure because of the close relation between such models and a proportional hazards model for time to cure (webappendix).

    View all citing articles on Scopus
    View full text