Abstract
BACKGROUND: In outpatient populations, hypoglycemia has been associated with tramadol. We sought to determine the magnitude of risk for hypoglycemia associated with tramadol use in hospitalized patients. METHODS: During a 2-year period of observation, adult inpatients who received ≥1 dose of tramadol were identified and their medical records were reviewed. Patients were included if they had blood or plasma glucose (BG) concentrations measured on at least two occasions within five days after the initial administration of tramadol. A contemporary comparator group of hospitalized oxycodone recipients was similarly reviewed. RESULTS: Tramadol was administered to 2927 patients who met inclusion criteria. Among these, hypoglycemia (BG ≤70 mg/dL) was documented in 22 (46.8%) of 47 patients with type 1 diabetes, 113 (16.8%) of 673 patients with type 2 diabetes, and 103 (4.7%) of 2207 patients who did not have a diabetes mellitus diagnosis. In those without a diabetes diagnosis, the causality association between hypoglycemia and tramadol use was probable in 77 patients (3.5%). By comparison, hypoglycemia was documented in 8 (1.1%) of 716 matched oxycodone recipients without diabetes (p = 0.002). As compared with tramadol recipients who did not develop low BG concentrations, those who experienced tramadol-related hypoglycemia were relatively young (mean age 52.0 versus 59.8 years; p = 0.027) and predominantly female (74.0% versus 59.8%; p = 0.012). CONCLUSIONS: Tramadol use was causally associated with hypoglycemia in hospitalized patients. The proportion of patients without diabetes who developed hypoglycemia was higher among those who received tramadol than among those who received oxycodone. TRIAL REGISTRATION: Colorado Multiple Institutional Review Board Protocol № 15-2215. Registered/approved 8 December 2015.