Antibiotic-Induced Pulmonary Fibrosis: National Database Analysis

Background: Pulmonary fibrosis (PF) is a major global health issue associated with substantial morbidity across all age groups. One of the important etiological factors contributing to PF is drug-induced lung injury, which can result from both direct and indirect damage to the pulmonary parenchyma caused by various pharmacological agents, including chemotherapeutics, antirheumatic drugs, cardiovascular medications, and certain antimicrobial agents. The aim of our study was to assess the structure of antibacterials involved in drug-induced PF (DIPF) and analyze signals of DIPF, calculating the reporting odds ratio (ROR) and proportional reporting ratio (PRR) using spontaneous reports (SRs) extracted from the Russian National Pharmacovigilance database. Methods: A retrospective, descriptive pharmacoepidemiological analysis of SRs from the AIS database for the period 1 April 2019-31 March 2025 was conducted. Results: A total of 130 SRs with data on DIPF associated with antibacterial agents were identified, with patients' mean age of 59.1 ± 14.46 years. Death was reported in 65 SRs (50%) with a mean age of 53.0 ± 13.66 years. Next, antibacterials were identified as leaders: sulfamethoxazole (used alone or in combination with trimethoprim, 20.7% (n = 50)), azithromycin (18.2%, n = 44), levofloxacin (12.4%, n = 30), doxycycline (11.6%, n = 28), and cefuroxime (10.7%, n = 26). Disproportionality analysis performed with PRR and ROR calculation revealed the strongest association with DIPF for cefuroxime (PRR = 15.11, 95% confidence interval, CI: 10.25-22.27; ROR = 15.31, 95% confidence interval, CI: 10.33-22.68). Conclusions: Cefuroxime was revealed as a drug with an unexpected but robust safety signal for DIPF, warranting heightened clinical awareness and further investigation. The observed associations between antibacterial agents and DIPF should be interpreted with caution, as they may reflect protopathic bias (antibiotics prescribed for early symptoms of unrecognized pulmonary fibrosis) or context-dependent biological effects rather than true pro-fibrotic drug properties. Our findings do not establish causality but rather generate safety signals that warrant validation through prospective studies with detailed clinical phenotyping and mechanistic investigations using human cell lines.