Background: Fluoroquinolones (FQ) are commonly used antibiotics known to cause hypoglycemia in rare occasions. The risk of hypoglycemia with FQ may become clinically relevant in populations with a high baseline risk of this adverse effect such as patients with type 2 diabetes treated with sulfonylureas (SU), antidiabetic drugs that strongly reduce glucose levels via stimulating insulin secretion.
Objectives: To assess whether use of FQ is associated with an increased risk of severe hypoglycemia compared to use of amoxicillin in patients treated with SU.
Methods: We conducted a population-based cohort study using the UK Clinical Practice Research Datalink Aurum linked to hospitalization and vital statistics data. Out of a base cohort of patients initiating SU between 1998 and 2020, we assembled a study cohort of patients co-exposed to SU and either FQ or amoxicillin. Study cohort entry date was the date of antibiotic initiation while on SU. Using an intent-to-treat (ITT) exposure definition, we assessed the 30-day risk of severe hypoglycemia (hospitalization with or death due to hypoglycemia) associated with use of FQ compared to use of amoxicillin. The two groups were matched on number of prior SU prescriptions, antibiotic use (other than FQ or amoxicillin) in the 30 days prior to study cohort entry, and propensity score (1:5; greedy matching; caliper 0.2). Covariates with imbalances after matching (standardized difference >10%) were included in the outcome models. Cox proportional hazards models estimated hazard ratios (HRs) with 95% confidence intervals (CIs) of severe hypoglycemia in the matched cohort. In secondary analyses, we stratified by age and sex after repeating the matching process. Sensitivity analyses addressed the impact of misclassification of exposure (15- and 60-day ITT) and outcome (hospitalization codes in primary position only), and pharmacologic heterogeneity (restriction to the FQ ciprofloxacin).
Results: Out of 325,000 SU initiators, 143,417 were subsequently co-exposed to SU and either FQ or amoxicillin. Among patients treated with SU, FQ use was not associated with an increased risk of severe hypoglycemia (incidence rates, 38.3 vs 31.2 per 1000 person-years; HR, 1.17; 95% CI, 0.91 to 1.50) compared to amoxicillin use. There was no effect modification by sex. However, FQ use was associated with an increased risk of severe hypoglycemia in patients aged < 65 years (HR, 2.90; 95% CI, 1.41 to 5.97) but not in those aged ≥65 years (HR, 1.03; 95% CI, 0.79 to 1.35). Sensitivity analyses were consistent with the primary analysis (HR range from 0.94 to 1.15).
Conclusions: In a population with a high baseline risk of hypoglycemia, FQ were not associated with an excess risk of this adverse effect compared to amoxicillin. An excess risk among younger adults is possible.
