Abstract
Abstract
Background
Multiple randomized controlled studies have compared numerous antibiotic regimens, including new, recently commercialized antibiotics in the treatment of nosocomial pneumonia (NP). The objective of this Bayesian network meta-analysis (NMA) was to compare the efficacy and the safety of different antibiotic treatments for NP.
Methods
We conducted a systematic search of PubMed, Medline, Web of Science, EMBASE and the Cochrane Library databases from 2000 through 2021. The study selection included studies comparing antibiotics targeting Gram-negative bacilli in the setting of NP. The primary endpoint was 28 day mortality. Secondary outcomes were clinical cure, microbiological cure and adverse events.
Results
Sixteen studies encompassing 4993 patients were included in this analysis comparing 13 antibiotic regimens. The level of evidence for mortality comparisons ranged from very low to moderate. No significant difference in 28 day mortality was found among all beta-lactam regimens. Only the combination of meropenem plus aerosolized colistin was associated with a significant decrease of mortality compared to using intravenous colistin alone (OR = 0.43; 95% credible interval [0.17–0.94]), based on the results of the smallest trial included. The clinical failure rate of ceftazidime was higher than meropenem with (OR = 1.97; 95% CrI [1.19–3.45]) or without aerosolized colistin (OR = 1.40; 95% CrI [1.00–2.01]), imipemen/cilastatin/relebactam (OR = 1.74; 95% CrI [1.03–2.90]) and ceftazidime/avibactam (OR = 1.48; 95% CrI [1.02–2.20]). For microbiological cure, no substantial difference between regimens was found, but ceftolozane/tazobactam had the highest probability of being superior to comparators. In safety analyses, there was no significant difference between treatments for the occurrence of adverse events, but acute kidney failure was more common in patients receiving intravenous colistin.
Conclusions
This network meta-analysis suggests that most antibiotic regimens, including new combinations and cefiderocol, have similar efficacy and safety in treating susceptible Gram-negative bacilli in NP. Further studies are necessary for NP caused by multidrug-resistant bacteria.
Registration PROSPERO CRD42021226603
Publisher
Springer Science and Business Media LLC