Adults with nonsevere community-acquired pneumonia (CAP) responded nearly equally to three first-line and alternative antibiotic regimens, based on data from more than 23,000 individuals.
Current recommendations for the treatment of CAP vary across guidelines, wrote Anthony D. Bai, MD, of Queen’s University, Kingston, Ont., and colleagues. However, most guidelines were based on studies that were not powered to examine the effect of treatments on mortality, they said.
“Large observational studies could fill this gap by comparing multiple treatment arms, including patients not well represented in trials, and having a large sample size powered to detect a difference in mortality,” they noted.
In a study published in Chest, the researchers reviewed data from 23,512 consecutive patients admitted to 19 hospitals in Canada for CAP between 2015 and 2021. Patients were treated with one of four initial antibiotic regimens: beta-lactam plus macrolide (BL+M), beta-lactam alone (BL), respiratory fluoroquinolone (FQ), or beta-lactam plus doxycycline (BL+D). Of these, BL+M is generally considered the first-line regimen, the researchers noted.
Patients were divided into four groups according to their initial antibiotic treatment within 48 hours of admission; 9,340 patients received BL+M, 9,146 received BL, 4,510 received FQ, and 516 received BL+D. The duration of any antibiotic that was active against CAP was at least 4 days, or until hospital discharge or death.
The primary outcome was all-cause in-hospital mortality, which was 7.5%, 9.7%, 6.7%, and 6.0% for patients in each of the four treatment groups, respectively. Relative to the first-line therapy of BL+M, the adjusted risk differences for BL, FQ, and BL+D were 1.5%, –0.9%, and –1.9%, respectively.
The adjusted in-hospital mortality was not significantly different between BL+M and either FQ or BL+D, but the difference of 1.5% seen with BL alone suggested a “small but clinically important difference,” the researchers noted.
Key secondary outcomes were the length of hospital stay and being discharged alive. The median length of stay was 4.6 days for BL+M, 5.2 days for BL, 4.6 days for FQ, and 6.0 days for BL+D. Patients treated with BL also had a longer time to hospital discharge, which suggests that BL may not be as effective as the other regimens, the researchers said. In addition, patients in the BL group had a subdistribution hazard ratio of 0.90 for being discharged alive, compared with the BL+M group after adjustment with propensity scores and overlap weighting.
Overall, the results support dropping BL as a first-line regimen in the current ATS/IDSA guidelines, and support the recommendation of BL+M, FQ, and BL+D as similarly effective options as listed in other guidelines, applied according to other patient characteristics. For example, “Doxycycline may be preferred over a macrolide in many cases such as macrolide allergy, prolonged QT, or high [Clostridioides] difficile risk,” the researchers said.
The findings were limited by several factors including the lack of follow-up data after hospital discharge.
However, the results were strengthened by the large sample size and use of a comprehensive database that allowed adjustment for many variables, as well as the availability of complete follow-up data for the time spent in the hospital. Based on this study, clinicians may choose a respiratory fluoroquinolone, a beta-lactam plus macrolide, or a beta-lactam plus doxycycline for equally effective antibiotic treatment of CAP, based on the best fit for each individual patient, the researchers concluded.
The study received no outside funding. The researchers had no financial conflicts to disclose.