Fluoroquinolones and the risk for incidental seizures: a comparative retrospective study

Abstract Background Over the years, reports have associated fluoroquinolones (FQ) with seizures. The incidence and whether FQ compared to non-epileptogenic antibiotic are associated with increased risk of seizures has yet to be examined. Methods A retrospective observational study of hospitalized patients treated with FQ (ofloxacin, ciprofloxacin, levofloxacin, moxifloxacin) or macrolides (MA: azithromycin or roxithromycin) between January 2009 and January 2021 in a large tertiary academic medical centre. The outcome was the occurrence of a seizure during treatment. The Naranjo scale was used to assess causality between FQ treatment and seizures. Comparative analysis was conducted using propensity score matching to correct for possible bias due to non-random selection, followed by inverse probability weighting (IPW) to estimate the difference in seizure risk between FQ and MA. Results Overall, 52 722 patients were treated with FQ during a total of 178 982 days. Mean age was 65 (±19) years and 47% were females. Thirty-three patients (0.06%) experienced a seizure, yielding an incidence of 1:5422 treatment days. Causality was deemed probable and possible among 9/33 and 24/33, respectively. The MA group composed of 8522 patients treated during 17 954 treatment days. Mean age was 65 (±21) years, 49% were females. Six (0.07%) patients experienced each a single seizure. IPW estimated OR for seizures among the FQ versus MA group was 1.44 (95%CI 0.59–3.5, P = 0.42). Discussion The incidence of FQ associated seizures among hospitalized patients is low and the risk did not significantly exceed that under macrolides. Our results provide evidence for clinicians and decision-makers when balancing fluoroquinolones risks and benefits.