Insulin therapy in people on dialysis

Abstract

Insulin therapy in people with type 2 diabetes mellitus (T2DM) and kidney disease undergoing hemodialysis requires integrating glucose loss during the session, insulin adsorption by the membrane, and the subsequent counterregulatory response. The result is a dynamic profile, with some characteristics that are repeated in different studies using continuous glucose monitoring (CGM), showing that most patients arrive with pre-dialysis values, with a typical decrease in blood glucose during dialysis and an increase after dialysis¹

In 59 adults treated with insulin, CGM detected hyperglycemia >180 mg/dL in 98% during the pre-dialysis period, 61% during the session, and 100% after dialysis; for values ​​>250 mg/dL, the proportions were 80%, 20%, and 91%. Hypoglycemia (<70 mg/dL) was observed in 32%, 7.4%, and 25% of patients during these periods. Compared to capillary blood glucose monitoring, the CGM detected hypoglycemic episodes in a higher proportion of patients (<70 mg/dL - 47% versus 25%) and identified nocturnal and prolonged episodes in 29% and 12%, respectively, which were not recorded by capillary blood glucose monitoring².

These findings demonstrate that insulin therapy should be individualized by comparing days with and without dialysis and integrating information from the CGM (or available capillary blood glucose monitoring), dialysis session time, food intake, and the action profile of the prescribed insulins. A reduction of approximately 25% of the mid-day or prandial dose, whose peak coincides with hemodialysis, can be considered, but an automatic 25% reduction of the total daily insulin on hemodialysis days is not recommended. Prolonged basal insulin regimens should maintain a stable schedule, with adjustments guided by the glycemic pattern, and in many cases, the post-dialysis schedule will be preferred for once-daily analogs. The bolus dose should be linked to the food actually ingested. Systematic aggressive corrections before the session should be avoided.

In a European observational cohort of 1,446 patients, treatment with analog insulins was associated, compared to human insulin, with lower three-year all-cause mortality (22.0% vs. 31.4%; adjusted HR 0.808), fewer major cardiovascular events (26.8% vs. 35.9%; HR 0.817), and fewer hospitalizations (58.2% vs. 75.0%; HR 0.757), with no difference in intradialytic hypoglycemia^3.

The goal is to reduce overall glycemic fluctuations through simple and predictable regimens, conservative titration, and monitoring capable of recognizing sustained hyperglycemia and undetected hypoglycemia.