Glucose-lowering drugs pose no risk of severe hyponatremia

Use of glucose-lowering medications appears to have neutral effects on the risk of hospitalization due to severe hyponatraemia, with a recent study even showing an inverse association across all investigated drugs.

The study involved 14,359 hospitalized patients with hyponatraemia and 57,383 matched controls without the condition. The median age of the overall population was 76 years, and 72 percent were women.

More patients than controls had been recently dispensed a glucose-lowering medication (10.1 percent vs 7.5 percent).  The most frequently prescribed were metformin (n=3,027), insulin (n=2,510) and sulfonylureas (n=1,359).

In unadjusted logistic regression models, the use of glucose-lowering medications was associated with a higher risk of hospitalization due to hyponatraemia compared with nonuse. The odds ratios (ORs) were 1.41 (95 percent confidence interval [CI], 1.29–1.54) with insulins, 1.38 (95 percent CI, 1.27–1.50) with metformin and 1.22 (95 percent CI, 1.07–1.38) with sulfonylureas.

However, on multivariable analysis that controlled for confounding factors, the association was consistently reversed. The adjusted ORs for hospitalization due to hyponatraemia was 0.63 (95 percent CI, 0.58–0.68) with any glucose-lowering medication, 0.58 (95 percent CI, 0.52–0.65) with insulins, 0.81 (95 percent CI, 0.72–0.90) with metformin and 0.81 (95 percent CI, 0.69–0.94) with sulfonylureas.

The estimates for newly initiated medications were generally higher, whereas those for ongoing treatment were further decreased. The adjusted ORs associated with ongoing treatment with insulins, metformin and sulfonylureas were 0.54 (95 percent CI, 0.48–0.61), 0.82 (95 percent CI, 0.73–0.91) and 0.78 (95 percent CI, 0.66–0.92), respectively.

Researchers pointed out that pharmacologic mechanisms may have mediated the associations observed, but the uniform effects across drug-classes suggest that properties of the diabetic disease are of importance.