(P11-011-23) The Association Between Diabetes Medication Use and Hypomagnesemia: The Boston Puerto Rican Health Study

Objectives: Hypomagnesemia is commonly observed in individuals with diabetes, but whether diabetes medications alter magnesium (Mg) status remains unclear. Therefore, we aimed to examine the cross-sectional association between diabetes medication use and hypomagnesemia.

Methods: Adults from the Boston Puerto Rican Health Study (BPRHS) who participated in health and sociodemographic interviews and provided a blood sample between 2006 and 2011 were included in analyses (n=1062). Diabetes medication use within the past year was obtained during home interviews and was classified in three ways: 1) use of any or specific types of diabetes medications (users vs. nonusers), 2) count of medications (nonusers vs. 1 medication vs. 2 medications vs. use of ≥3 medications), and 3) duration of use (nonusers vs. use ≤5 y vs. use >5 y). Serum Mg was measured using the Xylidyl blue-I chromogenic method. Hypomagnesemia was defined as serum Mg < 0.75 mmol/L. Logistic regression models were used to estimate odds ratios (OR) and 95% confidence intervals (CI) after multivariable adjustments for hemoglobin A1c and potential confounders, including dietary and supplemental Mg intakes. A stratified analysis by Mg intake (≥ vs. < Recommended Dietary Allowance) was performed.

Results: Mean age was 59.0 ± 7.5 y, with 72.5% female and 45.7% with hypomagnesemia. The odds of hypomagnesemia in users of any diabetes medications were three times that in nonusers [OR (95% CI) = 3.23 (2.26-4.62)]. Higher count of diabetes medications was associated with higher odds of hypomagnesemia [1 vs. nonusers: OR (95% CI) = 2.32 (1.54-3.49), 2 vs. nonusers: OR (95% CI) = 3.98 (2.43-6.53), and ≥3 vs. nonusers: OR (95% CI) = 10.5 (4.41-24.9), P trend < 0.001]. Longer duration of medication use was associated with higher odds of hypomagnesemia [≤5 y vs. nonusers: OR (95% CI) = 2.72 (1.80-4.12), >5 y vs. nonusers: OR (95% CI) = 4.02 (2.57-6.27)]. Use of metformin [OR (95% CI) = 3.67 (2.52-5.35)] or glitazones [OR (95% CI) = 1.96 (1.05-3.66)], but not insulin or sulfonylureas, retained significance after mutual adjustment. Results were consistent in both Mg intake groups.

Conclusions: Diabetes medication use was cross-sectionally associated with elevated odds of hypomagnesemia. Mg status merits more attention among people who take metformin, glitazones, or multiple diabetes medications.

Funding Sources: NIH