Fluoroquinolones and aortic aneurysms: Which fluoroquinolones are at risk?

A recent cohort study described a 66% increased rate of aortic aneurysms in patients receiving fluoroquinolones (Pasternak et al. BMJ 2018; 360:k678). The authors did not study the different fluoroquinolones. The present study was performed to investigate the fluoroquinolones more at risk of aortic aneurysms. We used Vigibase ® , the World Health Organization Global Individual Case Safety Reports (ICSRs) database, performing a case non-case study where cases were ICSRs of “aortic aneurysms and dissections” (HLGT) with fluoroquinolones registered until the 31st December 2017 as “suspected” or “concomitant” in patients ≥ 50 years. In order to reduce possible confounding, comparison was made with amoxicillin. Results are shown as Reporting Odds Ratios (ROR) of the exposure odds among reported cases of aortic aneurysms to the exposure odds among reported non-cases. Among the 6,462,912 ICSRs in Vigibase ® , 172,588 involved fluoroquinolones with 113 aortic aneurysms and 40,658 amoxicillin with 8 aneurysms. Most of the patients were females (54.9%) with a mean age of 67.6 ± 11.1 years. The first ICSR was reported in 2000. The ROR value for fluoroquinolones versus amoxicillin was 3.33 [1.63–6.82] ( P  = 0.01) and 3.13 [1.53–6.43] ( P  = 0.002) after adjustment on age and sex. Analyses of individual fluoroquinolones found an association with levofloxacin (3.16 [2.54–3.94]) ( P P  = 0.0014) but not with ciprofloxacin, gatifloxacin, ofloxacin or tosufloxacin. The present study, first, confirms the previous signal of aortic aneurysms with fluoroquinolones using another validated method. This ADR could be explained by upregulation of metalloproteinases by fluoroquinolones. Second, we described, for the first time, a differential risk among the different drugs, with a signal for levofloxacin and moxifloxacin but not the other fluoroquinolones. The pharmacological mechanisms explaining these differences remain unknown.