Is there an ideal antibiotic for mild to moderate community-acquired pneumonia?

Community-acquired pneumonia (CAP) is a common and potentially serious infectious condition, and one of the first decisions to be made by the emergency physician is about the most appropriate treatment scenario: outpatient vs. hospital.

The decision regarding the need for hospitalization is supported by prognostic scores, such as the PSI (Pneumonia Severity Index) and the CURB-65, which aim to complement the individualized perception of the case by the health professional, including socioeconomic parameters, health literacy and conditions of access to emergency medical services.

The Brazilian Society of Pulmonology and Phthisiology (SBPT), in 2018, reiterated the recommendation of monotherapy with beta-lactam or macrolide as the first line of treatment for mild to moderate CAP, in accordance with most global guidelines. Therefore, it is up to the prescriber to select the antimicrobial class and the drug to be used, leading us to the following question: is there any hierarchy regarding the effectiveness of the different options available?

Severe community-acquired pneumonia

Study

The study published by Acharya and collaborators in February 2024 in Journal of General Internal Medicine (JGIM) consisted of a network meta-analysis, covering 24 randomized clinical studies (RCTs) selected from 1,060 cataloged in PubMed and Cochrane, with a number of 9,361 patients and published between 1999 and 2022. The objective was to compare the clinical response rate and overall mortality between different empiric oral antibiotic therapy regimens among adults with mild to moderate CAP. The average age ranged between 38 and 60.8 years, with a predominance of males (62%). The majority of studies aimed at outpatient treatment (17 studies), although 6 were hybrid and only 1 was exclusively inpatient.

The populations assessed were diverse, including South Africa, Argentina, Brazil, Canada, Chile, China, Colombia, Spain, USA, Estonia, France, India, Israel, Italy, Lithuania, Mexico, Peru, Poland, Russia, Taiwan and Turkey. Cases of mild (PSI class I or II) and moderate (PSI class III and IV) CAP were included. Studies on ventilator-associated pneumonia (VAP), severe CAP (PSI class V) and in which there was initial exposure to parenteral antibiotic therapy were excluded. The risk of bias was classified as low in 6 RCTs, uncertain in 6 and high in 12, especially at the expense of lack of clarity about allocation criteria, failure of blinding and funding by the pharmaceutical industry in most of them.

The antimicrobial classes evaluated were quinolones (levofloxacin, moxifloxacin, nemonoxacin and sparfloxacin), beta-lactams (amoxicillin-clavulanate), cephalosporins (cefuroxime, cefditoren and cefpodoxime), macrolides (azithromycin, clarithromycin and roxithromycin) and ketolides (telithromycin).

Results

The primary outcome assessed was composed of all-cause mortality and clinical response rate.

The ranking results were based on the p-scorewhose result varies between 0 (less probability of being the most effective antibiotic) and 1 (highest probability of being the most effective antibiotic).

The recorded clinical response rate was high: 86.2% (8,073 among 9,361 patients), while the overall mortality was 0.97% (57 deaths among 5,883 patients).

The antibiotics most likely to achieve clinical response were levofloxacin, nemonoxacin and telithromycin, with p-scores of 0.71, 0.79 and 0.69, respectively (risk ratio: 1.01 to 1.03).

In contrast, the antibiotics with the lowest response rates were penicillin and amoxicillin, with p-scores of 0.09 and 0.19, respectively (RR 0.85 and 0.93).

The drugs most likely associated with reduced mortality were levofloxacin, nemonoxacin, azithromycin and amoxicillin-clavulanate, with p-scores of 0.85, 0.75, 0.74 and 0.68, respectively (RR of 0.04 to 0.12).

Regarding antibiotic classes, quinolones and macrolides were the most effective in terms of clinical response, with a p-score of 0.71 and 0.7, respectively (RR of 1.0); while the combination of amoxicillin-clavulanate and macrolides and beta-lactams in monotherapy were the least effective, with a p-score of 0.11 and 0.22, respectively (RR of 0.93 and 0.94).

Quinolones were most associated with reduced mortality (p-score of 0.63).

It is noteworthy that all confidence intervals overlapped with 1, in addition to being wide and partially overlapping.

Conclusion and practical messages

• The definition of the first line of oral antibiotic therapy in cases of mild to moderate CAP must be individualized based on clinical and epidemiological data, recognizing regional and temporal variations in antimicrobial resistance patterns.
• The meta-analysis summarized here demonstrated a tendency towards superiority of quinolones in relation to beta-lactams, macrolides and ketolides, without, however, reaching statistical significance.
• The data obtained, therefore, do not allow identifying a clear superiority of one drug or antimicrobial class over another, partly due to the high clinical response rate (> 80%) and low mortality (< 1%) with the regimens used.

Read too: Hydrocortisone in severe community-acquired pneumonia

Additional comments

• To the network meta-analysesor network meta-analysisare statistical tools that enable the comparison of effectiveness between 3 or more treatments for the same condition, allowing direct and indirect comparison between them and their ranking.
• A nemonoxacin is a non-fluorinated quinolone whose first studies in vitro date back to 2009. Its advantages over other quinolones include coverage of resistant gram positive bacteria (MRSA and pneumococcus). It has currently been used for the treatment of CAP and pyelonephritis predominantly in Asia, Turkey and Russia. The need for coverage of MRSA and resistant pneumococcus in cases of mild to moderate CAP in Brazil is lower, due to the low prevalence of these germs for now.
• A telithromycin, in turn, is a ketolide derived from erythromycin, classifying as a semi-synthetic macrolide. It has bactericidal activity directed against resistant bacteria (pneumococcus and hemophiles), being indicated for the treatment of CAP and rhinosinusitis. Approved by the FDA in 2004, it was withdrawn from the American market in 2016 due to serious adverse events among patients with myasthenia gravis.