PerspectiveIn the Journals

Higher insulin dosage for type 2 diabetes may increase mortality risk

Show Citation

December 10, 2014

Patients with type 2 diabetes taking exogenous insulin could be at increased risk for death from all causes, major adverse cardiovascular events and cancer with increasing dosage, according to research published in Diabetes, Obesity and Metabolism.

“We found that there was an association between the dose of prescribed insulin and increased risk of all-cause mortality,” the researchers wrote. “A statistically significant association with major adverse cardiovascular events and cancer was also observed.”

Sarah E. Holden, a PhD student at Cardiff University School of Medicine, United Kingdom, along with Craig Currie, PhD, and colleagues wrote that the findings add to ongoing concerns regarding insulin use to treat the chronic disease but warrant additional investigation.

“Limitations associated with both the estimation of insulin dose and also the retrospective observational nature of the study mean that this hypothesis should be tested using an interventional study design,” they wrote.

Using the UK Clinical Practice Research Datalink, the researchers identified 6,484 patients (mean age at baseline, 64 years) with type 2 diabetes who progressed to insulin monotherapy from 2000 onward and followed them for an average of 3.3 years.

Risks of progression to serious adverse outcomes — all-cause mortality, incident major adverse cardiovascular events (MACE) and incident cancer — were compared using Cox proportional hazards models.

Insulin exposure as a cumulative, continuous, annually updated, time-dependent covariable was introduced into the model as prescribed international units per kilogram per day; results were reported in four patient groups, defined by the doses ranging from <0.5 units per kilogram of bodyweight per day, up to doses >1.5 units.

The analysis of medical history showed 1,110 deaths, 342 MACE and 382 incident cancers. Unadjusted event rates, per 1,000 person-years, were 61.3 deaths, 26.4 incident MACE and 24.6 incident cancers. Adjusted HRs, relative to 1-unit increases in insulin dose, were 1.54 (95% CI, 1.32-1.78) for all-cause mortality, 1.37 (95% CI, 1.05-1.81) for MACE and 1.35 (95% CI, 1.04-1.75) for cancer.

“When compared to patients who received doses of less than 0.5 units, our findings indicate that patients receiving doses of between 1 to 1.5 units and more than 1.5 units were associated with a much greater death rate over time,” Currie said in a press release.

The researchers noted that the potential risks associated with insulin therapy should be considered in the context of its benefits for glucose control, inherently reducing the risk for microvascular events, and that the findings do not relate to type 1 diabetes.

Retrospective observational studies can only show associations with events, they said.

“What this study does show is a need for prospective randomized controlled trials to investigate the cause behind these patterns to provide conclusive evidence as to the real effect insulin is having on the thousands of patients who rely on glucose-controlling medication,” Currie said in the release.

Disclosure: Please see study for full list of disclosures.

itj+ Infographic

itj+ Perspective

Zachary T. Bloomgarden

Zachary T. Bloomgarden

The research by Holden et al raises the question of whether high insulin dosages mediate the possible adverse effects of cardiovascular disease, cancer and other complications. It should be noted that the study focuses on a rather unusual group, of persons with type 2 diabetes but yet receiving insulin as monotherapy. Since the vast majority of type 2 diabetic patients receiving insulin also receive other glucose-lowering medications, there may be further peculiarities of the group analyzed that limit our ability to extrapolate from the observations in the fashion described. This all suggests that the need for high insulin dosages is a marker of an underlying insulin-resistant state, which is the explanatory mechanism for the observed outcomes, and which is certainly well-recognized to be associated with weight gain, dyslipidemia, hypertension, and many other conditions. If high insulin dose requirement is a marker of underlying diabetes characteristics, and not in itself the mediator of adverse outcomes, then it would be adding insult to injury to deny patients the amount of insulin they require because of a putative association with adverse outcome. We certainly would not want to tell patients to refrain from taking sufficient insulin to control their blood glucose under those circumstances. Furthermore, even if insulin itself is in some fashion causally related to adverse outcomes, it still would not necessarily be the case that one should limit insulin dosages and allow blood glucose levels to remain high, which we know without doubt to have a myriad of adverse consequences.

Zachary T. Bloomgarden, MD, MACE
Endocrine Today Editorial Board Member

Disclosure: Bloomgarden reports consultancy/advisory fees from BMS, AstraZeneca, Johnson & Johnson, Merck, Novartis, Dainippon Sumitomo Pharma America, Forest Laboratories, Boehringer Ingelheim and Medtronics.