Fluoroquinolone-Associated Peripheral and Central Nervous System-Related Disorders: A Large German Claims-Based Cohort Study

Abstract

Purpose: To examine the association between newly prescribed fluoroquinolones (FQ) and the occurrence of polyneuropathy and certain neuropsychiatric events.

Methods: German statutory health insurance-based cohort study (2013-2019) of patients exposed to FQ compared to different reference antibiotics. Patients with an incident antibiotic prescription were followed up for 365 days after initial dispensing. Outcomes were defined by incident diagnoses of polyneuropathy/other peripheral nervous system-related diseases (PNS), depression/other affective disorders (DEP), mood-related symptoms (MOOD), somnolence/stupor/coma (SSC), and consciousness-related symptoms (CONSCIOUS). Piece-wise exponential additive mixed models adjusted for person-time, age, comorbidities, year, and quarter of cohort entry were applied.

Results: The outcome-specific cohorts included 10.7-14.3 million antibiotic episodes of which 1.7% had incident PNS, 4.8% DEP, 1.5% MOOD, 0.6% SSC, 0.8% CONSCIOUS. Relative risks for these outcomes ranged from 1.04 to 1.10 for FQ versus reference antibiotics. Stratified by gender and age groups, relative risk for PNS was high in ≤ 39-year-old males (aHR = 1.31 [95% 1.19; 1.45]), for MOOD and CONSCIOUS in 40-69-year-old females (aHR = 1.12 [1.08; 1.15], aHR = 1.14 [1.09; 1.19]) for DEP in ≥ 70-year-old males (aHR = 1.16 [1.14; 1.19]), whereas relative risk for SSC was high in ≤ 39-year-old females (aHR = 1.24 [1.10; 1.40]). FQ-associated PNS was mainly comprised of drug-induced polyneuropathy (aHR = 1.68 [1.58; 1.79]). Relative risks for each endpoint varied by choice of active comparator.

Conclusions: FQ episodes were associated with an increased risk for neurological and neuropsychiatric events, especially within the first 92 days after initial dispensing. Risk estimates vary by age and gender subgroups.

Keywords: active comparator new user design; administrative cohort; adverse drug reactions; fluoroquinolones; real‐world data.