The trusted source for
healthcare information and
Pneumonia is an extremely common and still devastating illness. It accounts for several million hospitalizations, although 75% of patients are treated in the outpatient setting. Mortality and prolonged disability are high especially for elderly patients, in whom more than 90% of deaths occur.1 Because of the enormous proliferation of antimicrobial agents and the wide proliferation of antimicrobial prescribing practices, the American Thoracic Society (ATS) developed a clinical practice guideline for the initial diagnosis of patients suspected of having community-acquired pneumonia.2 Treatment algorithms based on age, comorbidity, and site of treatment relied heavily on expert opinion for their development since, in most cases, high level of evidence (i.e., randomized clinical trials) were simply unavailable.3 In outpatients, the approach to healthy younger individuals (no significant medical illness) advised the use of erythromycin, clarithromycin, or azithromycin in smokers or the use of tetracycline in patients allergic to macrolide antibiotics. In individuals over age 60 or with comorbidity (COPD, CHF, diabetes mellitus, renal disease, liver disease, etc.), a second generation cephalosporin, trimethoprim-sulfamethoxazole (TMP/SMX), or a beta lactam/beta lactamase inhibitor with or without a macrolide was recommended.
The current study reported by Gleason et al evaluated the prescribing practices of physicians for pneumonia treated in the outpatient setting and compared medical outcomes and antimicrobial costs based on whether patient treatment was consistent with the ATS guidelines. A total of 864 patients, 560 individuals 60 years or younger and 318 older than 60 or with at least one significant medical illness, were interviewed in a variety of clinical settings (emergency departments, medical clinics, and practitioners’ offices). The database was completed prior to the publication of the ATS guidelines in 1993. In this outpatient study, few had specific microbiology etiology assigns (5%), with Pneumococcus being most commonly identified. Overall, 62% of the young no comorbid illness group had ATS-consistent treatment. This group was generally younger and received predominately erythromycin. Inconsistent therapy was predominantly amoxicillin in this group (89%). Older patients or with comorbidity had only 18% of their treatment consistent with ATS guidelines, usually cefuroxime or TMP/SMX. Young patients without comorbidity did exceedingly well, without any mortality and earlier return to work compared to those given ATS-inconsistent treatment. Those whose physicians prescribed ATS-consistent treatment had no difference in medical outcome, quality of life, or antibiotic-related complications compared to those given inconsistent treatment but most significantly had antibiotic costs that were three-fold lower ($5.43 vs $18.51).
In the older group (> age 60) and those with coexisting medical illness, mortality was 5%, with a trend toward higher likelihood of death if the ATS guidelines were employed (or 6.1), while other pneumonia and drug-related complications were similar. Median antimicrobial costs were 10-fold increased, with the magnitude of these costs maintained when inpatient antimicrobial costs were excluded for the 31 patients subsequently hospitalized ($73.50 vs $7.50).
This study provides both good and bad news and guidance on guidelines. First the good news. The ATS recommendations using an expert panel provided a template for the treatment of uncomplicated pneumonia in the outpatient setting. As validated by this prospective database, in young patients (~ age 35) without comorbidity, the use of erythromycin for approximately 12 days resulted in excellent outcomes and limited morbidly at three times less the cost. Unfortunately, the PORT pneumonia database called into question the recommendations of the ATS expert panel. Relatively few of those patients over age 60 or with medical illness actually received recommended therapy (17%), receiving for the most part macrolide therapy. In this group, those whose treatment conformed to ATS guidelines actually had no better (with a trend toward worse) outcome but with 10-fold higher costs for antibiotics. The excellent outcomes seen with the use of macrolides even in a higher risk older group suggests that perhaps the microbial spectrum between these two groups is not as different as the expert panel presumed and that atypical organisms, especially C. pneumoniae and M. pneumoniae, are important pathogens in relatively mild pneumonia even in those over age 60 and with at least one significant medical illness.
Additionally, we must continue to refine guidelines after they are formulated. Although evidence-based medicine is increasingly a part of medical practice in general and clinical practice guideline formulation in particular, much of medical practice is not founded on randomized clinical trials. Thus, it is unlikely that even if the ATS guidelines were being done today for the first time conclusions would have been different. Fortunately, the ATS and other medical societies (infectious disease) are actively working to revise and refine their initial advice. The work of Gleason et al gives them the evidence to make those guidelines clinically accurate.
1. Bartlett JG, Mundy LM. Community acquired pneumonia. N Engl J Med 1995;33:1618-1624.
2. Niederman MS, et al. American Thoracic Society guidelines for the initial management of adults with community acquired pneumonia: Diagnosis, assessment of severity, and antimicrobial therapy. Am Rev Resp Dis 1993;148:1418-1426.
3. Fein AM, Niederman MS. Guidelines for the initial management of community-acquired pneumonia: Savory recipe or cookbook for disaster? Am Rev Resp Crit Care Med 1995;152:1149-1153.