Analogue bests human insulin in T2D patients with kidney failure on peritoneal dialysis

Analogue insulin demonstrates superiority over human insulin in terms of survival among patients with kidney failure and type 2 diabetes (T2D) who are receiving peritoneal dialysis, according to a study presented at ERA 2025.

At baseline, daily dialysate glucose load (DDGL) was higher in patients treated with analogue insulin, said lead author Dr James Fotheringham, University of Sheffield, Sheffield, UK.

Fotheringham and colleagues identified a total of 13,146 people with T2D. Of these, 3,387 (26 percent) received a purely analogue or human insulin either prior to study recruitment (n=2,789) or during follow-up (n=628). Specifically, 355 and 3,032 patients received analogue and human insulin, respectively. [ERA 2025, abstract 998]

“The proportion of type 2 diabetics receiving an insulin, and the proportion on insulin [users] who received a human insulin varied by country, but patient characteristics by insulin type were similar,” Fotheringham said.

At 2 years, crude survival was higher in patients treated with analogue insulin than human insulin users (77 percent vs 68 percent; hazard ratio [HR], 0.71, 95 percent confidence interval [CI], 0.54‒0.94). In adjusted models, the corresponding survival rates were 77 percent and 65 percent (HR, 0.66, 95 percent CI, 0.48‒0.92).

“Similar effect sizes were obtained when stratifying by incident or prevalent peritoneal dialysis use,” according to Fotheringham.

DDGL and BMI

In incident cohorts, DDGL did not vary over time or by insulin type. In prevalent cohorts, mixed effects models estimated a DDGL increase of 1.0 g/month of peritoneal dialysis in human insulin users. Such increase with time was significantly lower among analogue insulin users (difference, ‒0.9 g/month, 95 percent CI, ‒1.2 to ‒0.5), but DDGL was higher by 23 g in the analogue insulin group at baseline.

In prevalent cohorts, BMI showed no statistical change with time (0.1/month) in the human insulin group. Among analogue insulin users, BMI was not higher at baseline (1.4, 95 percent CI, ‒1.6 to 4.4) and did not increase more quickly (‒0.1/month, 95 percent CI, ‒0.4 to 0.2).

“The absence of insulin-specific changes in DDGL over time or in BMI measures generally suggest the metabolic impact of analogue insulin is limited,” Fotheringham said.

“Although residual confounding may be present, this is reassuring given recent announcements from insulin manufacturers regarding halting human insulin production,” he added.

This study involved people with kidney failure and T2D receiving peritoneal dialysis from seven countries. Fotheringham and his team used Cox proportional hazard models, with inverse probability weighting adjustment for country, study phase, and patient characteristics, to estimate the mortality risk by insulin type for over 2 years.

Using mixed effects linear regression, the researchers longitudinally assessed the association between insulin type and the variables of BMI and DDGL, estimated using dialysate prescribed glucose strength and volume.

“T2D is highly prevalent in people with kidney failure on peritoneal dialysis, with a proportion requiring insulin therapy,” Fotheringham said. “Compared to human insulin, analogue (genetically altered) insulin has been associated with superior survival in haemodialysis.”