Anti-tumor Necrosis Factor-Alpha Therapy and Hypoglycemia: A Real-World Pharmacovigilance Analysis.

Abstract

An association between tumor necrosis factor (TNF)-α inhibitors and hypoglycemia has been detected in a few case reports and small case series; however, no relevant pharmacovigilance data have been published yet. The objective of this study was to detect and characterize relevant safety signals between hypoglycemia and TNF-α inhibitor use. Indication-focused disproportionality analysis was conducted to detect increased reporting of TNF-α-associated hypoglycemia compared with all other reports with the same indication during the same time period. Reporting odds ratios (RORs) with 95% confidence intervals (CIs) were calculated to determine disproportionality. To reduce potential confounding factors, adjusted RORs were further calculated by logistic regression to control for age, sex, diabetes status, and concomitant drugs that potentially affect blood glucose levels. In all, 1086 adverse drug reactions related to TNF-α inhibitors were reported as 'hypoglycemia'. There were no disproportionality signals of hypoglycemia in TNF-α inhibitor users with indication of inflammatory bowel disease. When TNF-α inhibitors were considered as a class, disproportion for hypoglycemia only emerged in indication of psoriasis (n=267, ROR 1.20, 95% CI 1.02-1.41). In further analyses of specific TNF-α inhibitor type, significant RORs for hypoglycemia were found in indication of rheumatic disease, including adalimumab in ankylosing spondylitis (n=37, ROR 1.97, 95% CI 1.28-3.04), psoriasis (n=160, ROR 1.64, 95% CI 1.37-1.97), and rheumatoid arthritis (n=230, ROR 1.35, 95% CI 1.16-1.56) and infliximab in psoriasis (n=18, ROR 2.14, 95% CI 1.33-3.42). After adjusting for confounding factors, only the signals of adalimumab were stable. Our study identified some potential pharmacovigilance signals between hypoglycemia and TNF-α inhibitors, which warrants further validation.