Azithromycin as Single-Drug Therapy for Community-Acquired Pneumonia

Abstract & Commentary

Synopsis: Azithromycin is an effective, single-drug therapy for mild-to-moderate community-acquired pneumonia.

Source: Feldman RB, et al. Arch Intern Med. 2003;163:1718-1726.

To determine whether azithromycin alone (without an additional ß-lactam antibacterial agent) is an effective treatment for patients with mild-to-moderately severe community-acquired pneumonia, Feldman and associates performed this retrospective cohort study of patients hospitalized at the West Los Angeles Healthcare Center over a 3.5-year period ending in mid-2001.

Eligible patients included those admitted to the general medical ward with clinical symptoms and signs of pneumonia, as well as radiographic demonstration of pulmonary infiltrates that were not previously present and that appeared within the first 2 days of hospitalization. Excluded were patients immunosuppressed by virtue of HIV infection, neutropenia, malignancy, or medication, as well as patients with hospital-acquired pneumonia, mycobacterial infection, or other explanations for the pulmonary infiltrate. Patients transferred from other hospitals or hospitalized within the previous 2 weeks were also excluded. Patients who were initially admitted to the general medical ward but then transferred to the intensive care unit within 24 hours were assumed to have severe pneumonia and to have been misclassified for the purposes of this study and were not included in the analysis.

A total of 442 patients’ records remained after the exclusion criteria were applied. The majority were white, but approximately 30% were black and 10% were Hispanic. Average age was mid-to-late 60s. As expected in a VA patient series, about 95% were male, and nearly half were current smokers.

All patients were assigned a pneumonia severity index score,1 a scoring system using age, coexisting medical conditions, abnormalities of mentation and vital signs upon admission to the hospital, and laboratory and radiographic findings. Statistically, patients in classes I and II have a predicted mortality of < 1%, while patients in class III have a predicted mortality of < 4%. Patients in classes IV and V have predicted probability of death of 4-10% and > 10%, respectively.

In this VA study, patients were grouped according to the initial antibiotic regimen received (excluding the first dose of antibiotic usually given in the emergency department). During the time of this study, recommended empiric therapy for patients with less-than-severe pneumonia was azithromycin as monotherapy. As a result, the largest group of patients (221—exactly half) received azithromycin initially; 29% received an alternate American Thoracic Society (ATS)-recommended regimen,2 and 21% received an initial regimen different from the ATS-recommended guidelines. (Inexplicably, identities of antibiotics received by these non-azithromycin groups were not specified.)

The groups were comparable with respect to age and other demographic features. Pneumonia severity index scores among the 3 groups were similar. Patients receiving azithromycin as the initial therapy fulfilled early discharge criteria sooner and length of stay was shorter than in patients in the other 2 groups. Need to transfer to the intensive care unit and in-hospital mortality—markers for progressive disease—were similar in all 3 groups.

Streptococcus pneumoniae was the most frequent presumed pathogen, comprising 41% of all isolates from sputum, other respiratory tract sites, and blood. Eighty percent of S pneumoniae isolates were erythromycin-susceptible. Outcomes were similar irrespective of erythromycin susceptibility among these isolates.

In summary, Feldman et al concluded that azithromycin as single-drug therapy for hospitalized patients with mild-to-moderate community-acquired pneumonia was equivalent to other ATS-recommended regimens.

Comment by Jerry D. Smilack, MD

The literature on community-acquired pneumonia is vast. A quick Medline search found 985 articles on the subject in the English language medical literature over the last 5 years. Comparison of results is confounded by such methodologic differences as definitions of pneumonia, case entry criteria, measurement of outcome, institutional differences in quality of care, etc. Efforts to standardize medical therapy have advanced considerably in recent years by publication of guidelines by the Infectious Diseases Society of America,3 the American Thoracic Society,2 the Canadian Infectious Diseases Society and Canadian Thoracic Society,4 and the Drug-Resistant Streptococcus pneumoniae Therapeutic Working Group.5

In general, for the hospitalized patient not requiring intensive care, most guidelines call for a ß-lactam, such as a third-generation cephalosporin or a ß-lactam/ß-lactamase inhibitor combination, plus a macrolide, or monotherapy with an enhanced-activity fluoroquinolone. The ATS guidelines, on the other hand, offer several options for patients with no underlying cardiopulmonary disease or modifying risk factors: azithromycin or an antipneumococcal fluoroquinolone as monotherapy, or a combination of doxycycline and a ß-lactam.

The present VA study retrospectively analyzed the outcomes of patients admitted with a diagnosis of community-acquired pneumonia. Because hospital routine at the time called for azithromycin monotherapy, half of the patients received that regimen from the outset. However, since treating physicians were apparently not compelled to follow recommended treatment guidelines, other regimens were also used.

Feldman et al found that patients treated with azithromycin generally fared as well as patients treated with other ATS-recommended and nonrecommended regimens. They had shorter lengths of stay, met clinical stability and discharge criteria more quickly, and had equal or lesser need to return to the emergency room or require readmission following hospital discharge than the comparator patient groups.

Although Feldman et al state that the treatment groups were almost entirely comparable, it appears to this reviewer that patients treated with azithromycin alone were "less sick" overall than those in the group receiving other ATS-recommended regimens or those treated with non-ATS recommended regimens. For example, the azithromycin group had lower pneumonia severity scores (although differences just escaped statistical significance); over 55% were in classes I and II, whereas 43% and 42% of patients in the other 2 treatment groups were in these lowest severity index classes. Additional support for the notion that the azithromycin group might not have appeared to the treating physicians to be as ill was the fact that patients in this group were less likely to have blood cultures drawn and to have arterial O2 assessment performed upon admission. In addition, patients admitted from a skilled nursing facility were more likely to receive non-ATS-recommended treatment, perhaps because the treating physicians felt they may be "sicker." Questions about comparability of the groups weaken the conclusions of this study.

Other randomized studies, both retrospective6 and prospective,7 have shown similar results even though details concerning methods of selection, analysis, and outcome were somewhat different. After reviewing all of these studies, one could conclude that patients hospitalized for community-acquired pneumonia, particularly in the absence of such significant risk factors as serious underlying disease, advanced age, recent nursing home exposure, and recent hospitalization or antibiotic exposure, can safely and effectively be treated with azithromycin as monotherapy. Of course, this begs the question: Could these patients be treated as outpatients? If so, all of the expert guidelines agree that either a macrolide or doxycy cline, among other alternatives, is appropriate therapy.


1. Fine MJ, et al. N Engl J Med. 1997;336:243-250.

2. Niederman MS, et al. Am J Respir Crit Care Med. 2001;163:1730-1754.

3. Bartlett JG, et al. Clin Infect Dis. 2000;31:347-382.

4. Mandell LA, et al. Clin Infect Dis. 2000;31:383-421.

5. Heffelfinger JD, et al. Arch Intern Med. 2000;160: 1399-1408.

6. Lentino JR, et al. Int J Antimicrob Agents. 2002;19: 61-66.

7. Vergis EN, et al. Arch Intern Med. 2000;160:1294-1300.

Dr. Smilack is a Infectious Disease Consultant, Mayo Clinic Scottsdale, Scottsdale, AZ.