Will Insulin Be The Death Of Our Patients?

Does insulin worsen cardiovascular outcomes and cause cancer? Should insulin be reserved as the last resort? In 2008, ACCORD raised questions about the cardiovascular safety of insulin, at least in patients with type 2 diabetes.¹  And an analysis in Diabeteologica suggested insulin glargine increased the risk of cancer.² So while intensive blood glucose control slows the progression of microvascular complications and most clinicians agree that there is a link between poorly controlled diabetes and cardiovascular disease (CVD), the data is not clear regarding the risks and benefits of insulin.

UKPDS showed a non-significant 16% reduction in cardiovascular events in the intensive glycemic control arm (sulfonylurea or insulin) which became significant after 10 years of follow-up (15% reduction in MIs and 13% reduction in all-cause mortality).^3,4 However ACCORD, ADVANCE and VADT trials, which enrolled patients with more advanced diabetes, showed no significant reduction in CVD outcomes with intensive glycemic control (A1c goal of <6-6.5%).^1,5,6

The two co-primary outcomes were: 1) death from CVD, non-fatal MI, or non-fatal stroke; and 2) a composite of these CVD events, plus a revascularization procedure, or hospitalization for heart failure. Secondary outcomes included composite microvascular outcomes, incidence of new onset diabetes, all-cause mortality, and new or recurrent cancers. Patients were followed for a total of 6 years.As expected, hypoglycemia and weight gain were more common in patients who received insulin glargine (p<0.001). Median weight increased by 1.6 kg in patients in the treatment group vs. 0.5 kg in the control group.

The strengths of ORIGIN include its large sample size, double-blinding and randomized study design. There were no differences in the two primary outcomes between the two treatment arms (see Table 1).  There were no differences in CVD events, mortality or cancer. Not surprisingly, fewer patients with IFG or IGT at baseline in the insulin glargine arm developed diabetes (ARR=8.6%, NNT=11.6). These results confirm previous studies that suggested early insulin use delays the progression from pre-diabetes to diabetes.^8,9

As expected, hypoglycemia and weight gain were more common in patients who received insulin glargine (p<0.001). Median weight increased by 1.6 kg in patients in the treatment group vs. 0.5 kg in the control group.

In contrast to UKPDS and ACCORD, ORIGIN suggests that insulin has a neutral effect on cardiovascular outcomes.^1,3,4,7 There are a few possible reasons for this. First, the patients followed in ORIGIN had IFG, IGT or diabetes with A1c < 9%. UKPDS included newly diagnosed patients with two consecutive fasting plasma glucose levels that were greater than ~110 mg/dL.³ Therefore, patients in ORIGIN were more likely to have uncontrolled diabetes for longer and therefore more likely to have cardiovascular complications already. Also, more patients in ORIGIN were taking beta-blockers, ACE inhibitors, ARBs, and statins, all drugs with cardioprotective properties. It is important to note that ORIGIN only followed patients for approximately 6 years. Despite no change in macrovascular events at the end of UKPDS, at the 10 year follow-up, significant reduction was seen.⁴ This suggests that the cardioprotective properties of insulin may take longer than 6 years to emerge. Though some might disagree, it is possible that a trend towards benefit would emerge if the ORIGIN study was extended to 10 years. Of note, only 47% of the patients in the insulin glargine arm and 60% of the patients in the standard care arm received metformin. This confounder is a significant limitation given that metformin is known to be cardioprotective and likely narrowed the difference between the two groups.^1,2As expected, hypoglycemia and weight gain were more common in patients who received insulin glargine (p<0.001). Median weight increased by 1.6 kg in patients in the treatment group vs. 0.5 kg in the control group.

ORIGIN confirms that insulin does not worsen cardiovascular outcomes or cause cancer.  Even though the study demonstrated that insulin decreases progression to diabetes in high-risk people, it failed to show that insulin has any cardioprotective benefit. Moreover, the study reminds us that weight gain and hypoglycemia are common problems, even with basal insulin. This should alleviate the persistent fear that insulin is harmful. Indeed, ORIGIN affirms insulin’s place in therapy.  In patients with type 2 diabetes and an A1c < 9%, ADA guidelines suggest using metformin first. If the A1c target is not met or maintained, then a 2^nd oral agent, GLP-1 agonist or basal insulin can be added.¹⁰However, in patients with type 2 diabetes and an A1c > 9%, ORIGIN reassures clinicians that insulin should be used without fear.