This article is more than 5 years old. Information may no longer be current.
Higher insulin doses not associated with increased mortality, CV risk in type 2 diabetes
Click Here to Manage Email Alerts
For adults with type 2 diabetes initiating insulin therapy, there is no dose-response relationship between the amount of daily insulin and all-cause death or cardiovascular risk, according to a retrospective study.
“For all-cause mortality, it was not until after we used a marginal structural model to account for potential time-dependent confounders within the putative causal pathways that an insulin dose-response relationship was no longer observed,” John-Michael Gamble, PhD, professor of medicine at the School of Pharmacy at Memorial University of Newfoundland in St. John’s, Canada, and colleagues wrote. “These findings could partly explain, and help to reconcile, the opposing conclusions reached by previous observational studies and randomized trials with respect to the safety of insulin.”
Gamble and colleagues analyzed health care records from 6,072 adults with type 2 diabetes aged at least 30 years who started metformin monotherapy from 2001 to 2012 and were new users of insulin (mean age, 60 years; 54% men; mean HbA1c, 8.5%; 6% had a previous CV event). Within the cohort, 3,599 were included in the subcohort linked to hospital records and death certificate information.
Patients attended primary care clinics participating in the U.K.-based Clinical Practice Research Database (CRPD). Researchers stratified patients into five categories by mean insulin dose per day (units) for all-cause mortality within 180-day time segments throughout follow-up (mean, 3.1 years): less than 25 units per day (reference group); 25 to less than 50 units per day; 50 to less than 75 units per day; 75 to less than 100 units per day and at least 100 units per day. Patients were stratified by three categories of insulin use for CV events: less than 50 units per day (reference group); 50 to less than 100 units per day and at least 100 units per day. Mean insulin dose during follow-up was 57.8 units per day.
Researchers used Cox proportional hazard models to assess relative differences in mortality and major adverse CV events; marginal structural models were then applied to reduce bias introduced by the time-dependent confounders affected by previous treatment.
After adjustment for baseline covariates, researchers found that higher insulin doses were associated with increased mortality. Patients receiving 25 to less than 50 units of insulin per day were at increased risk for all-cause death vs. those receiving less than 25 units (HR = 1.41; 95% CI, 1.12-1.78), as were patients receiving 50 to less than 75 units (HR = 1.37; 95% CI, 1.04-1.8), 75 to less than 100 units (HR = 1.85; 95% CI, 1.35-2.53) and at least 100 units per day (HR = 2.16; 95% CI, 1.58-2.93). However, adjustments for time-dependent changes in glycemic control, body weight, frequency of hypoglycemic events and occurrence of CV events decreased the association, with only the two highest doses of insulin linked to increased mortality. Any association between insulin dose and all-cause mortality was attenuated once researchers applied marginal structural models (P > .1 for all).
Higher insulin doses were not associated with composite major adverse CV events, according to researchers; the only significant association was between insulin dose and CV-related death for an insulin dose of 50 to less than 100 units per day (adjusted HR = 1.68; 95% Ci, 1.17-2.42) for insulin use of more than 100 units per day (adjusted HR = 2.65; 95% CI, 1.65-4.25).
“Overall, the result from this study reassures both patients and their physicians of the overall safety and absence of major [CV] harms of insulin use in the treatment of type 2 diabetes,” the researchers wrote. “The final result also contributes to understanding some of the potential reasons for the contrasting conclusions from nonrandomized and randomized studies regarding dose-dependent effects of insulin on mortality and [CV] outcomes.” – by Regina Schaffer
Disclosure: One of the researchers reports receiving an unrestricted research grant from Sanofi-Aventis for work unrelated to diabetes.
Click Here to Manage Email Alerts