The ACCORD study showed that aggressive intensification of glycaemic control in patients with type 2 diabetes can increase mortality (hazard ratio [HR] 1·22, 95% CI 1·01—1·46), including death from cardiovascular causes (1·35, 1·04—1·76).1 The reason for this unexpected finding is unknown. Of note was the high percentage of intensively treated patients receiving insulin therapy (77%) or thiazolidinedione therapy (with or without insulin) (92%), and their greater weight gain (3·5 kg vs 0·4 kg).1 Results of post-hoc analyses did not support the hypothesis that increased hypoglycaemia in patients who were intensively treated caused the excess deaths.2 The analyses showed, however, that a higher baseline HbA1c and a failure to improve average HbA1c throughout the study were linked to the increased mortality.3

In view of the findings of the ACCORD study, we propose that development of insulin resistance in crucial tissues such as the heart is a protective response to persistently raised glucose concentrations. We postulate that overriding insulin resistance in attempts to lower glucose with exogenous insulin—particularly in overweight and obese patients with the most refractory hyperglycaemia—could undo this protection and cause harm.

For decades the dogma has been that insulin resistance is mostly pathological. However, regulation of insulin sensitivity is an integral component of normal metabolic physiology, including, at times, the induction of insulin resistance. For example, in response to even short-term overfeeding, skeletal and cardiac muscle develops insulin resistance,4 which promotes diversion of excess nutrients to adipose tissue for safe storage. We, and others, have proposed that this induction of insulin resistance protects important tissues from nutrient-induced dysfunction.5

The myocardium, with its high energy needs, adapts to the predominant nutrient source, which is free fatty acids (FFA) during fasting and glucose in the fed state. In poorly controlled type 2 diabetes, the reciprocal relationship between FFA and glucose is lost and both are simultaneously raised. This places the myocardium at increased risk of nutrient overload and myocardial glucolipotoxicity.5, 6 We propose, however, that insulin resistance is a safeguard against glucolipotoxicity because it limits myocardial glucose uptake. Treatment of patients with the most refractory hyperglycaemic and hyperlipidaemic type 2 diabetes with large amounts of exogenous insulin could override this block against glucose entry, providing all the ingredients for glucolipotoxicity,6 a process that we have termed insulin-mediated metabolic stress. In the heart, this process would cause a metabolic cardiomyopathy. Jagasia and colleagues7 provided evidence that insulin resistance is easier to override in the myocardium than in skeletal muscle in patients with type 2 diabetes. Exogenous insulin tripled myocardial glucose uptake without compensatory reduction in FFA uptake,7 which is consistent with the capacity of exogenous insulin to drive excess nutrients into the heart. If the same mechanisms of insulin-mediated metabolic stress can function in skeletal muscle, inappropriate high-dose insulin use in patients with type 2 diabetes could also induce a metabolic myopathy that impairs the patient’s ability to exercise.

If our proposition is correct—that intensive treatment of type 2 diabetes with insulin is potentially harmful to overweight and obese patients with the most refractory hyperglycaemia—there should be support for it in clinical trials. In the major trials of intensive versus conventional glucose control, and insulin versus other glucose-lowering therapies, whenever high use of insulin was associated with weight gain of greater than 1 kg per year (ACCORD,1 VADT,8 DIGAMI-29), cardiovascular or all-cause mortality, or both, were increased, reaching statistical significance only in the higher powered ACCORD study. UKPDS, ADVANCE, and ORIGIN did not show increased cardiovascular disease or mortality in the intensive control or insulin therapy groups. However, the patient characteristics and the aggressiveness of insulin use differed greatly from those in ACCORD. Careful analysis showed that UKPDS, ADVANCE, and ORIGIN were not studies of intensive insulin use in obese patients with the most refractory hyperglycaemia. While results of population-based studies have shown increased risk of mortality in patients with type 2 diabetes treated with insulin, they are observational and should be interpreted with caution.10

Patients with type 2 diabetes should be considered individually, because their relative need for insulin resistance as a protective mechanism and the potential for long-term benefit from tight blood glucose control will differ. For example, an overweight individual with type 2 diabetes unable to exercise because of comorbidities will have much more difficulty achieving lifestyle change to overcome a state of positive energy balance (figure). In these patients, insulin resistance might be necessary to protect crucial tissues such as the heart and skeletal muscle against excess nutrient entry. An aggressive approach to lower blood glucose with insulin will override this protection, increasing the risk of insulin-mediated metabolic stress in these key tissues (figure). Furthermore, these patients are least likely to be able to adequately achieve lower HbA1c concentrations—the group that was at highest risk in ACCORD.3 Careful amelioration of very high blood glucose, however, will be necessary in these individuals. At the other end of the spectrum, a patient with type 2 diabetes who can avoid positive energy balance by lifestyle change will be at much lower risk of the harmful consequences of an aggressive approach to glucose lowering with insulin.

The safety of insulin sensitisers is probably associated with the mechanism by which they work. Those that enhance nutrient detoxification should be beneficial. Such agents are very different from insulin in that they reduce and do not override insulin resistance. Metformin and thiazolidinedione drugs both have mechanisms of action that include nutrient detoxification.5 The development of new insulin sensitisers that do not have a nutrient detoxification mechanism might not be advisable.

Further studies of the effect of insulin therapy on myocardial and skeletal muscle nutrient uptake, storage, and function in obese patients with type 2 diabetes are necessary to further explore the concept of insulin-mediated metabolic stress. Ultimately, carefully designed randomised clinical trials with long-term outcome data will be necessary to assess the safety of insulin therapy, and how best to use it, in obese patients with type 2 diabetes who do not achieve acceptable glycaemic control by other therapies. The results of the ACCORD study suggest that the combination of insulin with thiazolidinedione agents should be used with considerable caution.1 The use of insulin in combination with newer agents that avert positive energy balance, such as the glucagon-like peptide 1 mimetics, warrants particular attention in this challenging group of patients.

CJN has received speaking fees from AstraZeneca, Eli Lilly, GlaxoSmithKline, Merck Sharp and Dhome, Novartis, and Novo Nordisk; and has been a member of an advisory board for Sanofi Aventis. NBR and MP declare that they have no conflicts of interest.

1 Gerstein HC, Miller ME, Byington RP, et al. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med 2008;358: 2545-2559. CrossRef | PubMed

2 Bonds DE, Miller ME, Bergenstal RM, et al. The association between symptomatic, severe hypoglycaemia and mortality in type 2 diabetes: retrospective epidemiological analysis of the ACCORD study. BMJ 2010; 340: b4909. CrossRef | PubMed

3 Riddle MC, Ambrosius WT, Brillon DJ, et al. Epidemiologic relationships between A1C and all-cause mortality during a median 3·4-year follow-up of glycemic treatment in the ACCORD trial. Diabetes Care 2010; 33: 983-990. CrossRef | PubMed

4 Kraegen EW, Saha AK, Preston E, et al. Increased malonyl-CoA and diacylglycerol content and reduced AMPK activity accompany insulin resistance induced by glucose infusion in muscle and liver of rats. Am J Physiol Endocrinol Metab 2006;290: E471-E479. CrossRef | PubMed

5 Nolan CJ, Damm P, Prentki M. Type 2 diabetes across generations: from pathophysiology to prevention and management.Lancet 2011; 378: 169-181. Summary | Full Text | PDF(856KB) | CrossRef | PubMed

6 Taegtmeyer H, Stanley WC. Too much or not enough of a good thing? Cardiac glucolipotoxicity versus lipoprotection. J Mol Cell Cardiol 2010; 50: 2-5. CrossRef | PubMed

7 Jagasia D, Whiting JM, Concato J, Pfau S, McNulty PH. Effect of non-insulin-dependent diabetes mellitus on myocardial insulin responsiveness in patients with ischemic heart disease. Circulation 2001; 103: 1734-1739. CrossRef | PubMed

8 Duckworth W, Abraira C, Moritz T, et al. Glucose control and vascular complications in veterans with type 2 diabetes. N Engl J Med 2009; 360: 129-139. CrossRef | PubMed

9 Mellbin LG, Malmberg K, Norhammar A, Wedel H, Ryden L. Prognostic implications of glucose-lowering treatment in patients with acute myocardial infarction and diabetes: experiences from an extended follow-up of the Diabetes Mellitus Insulin-Glucose Infusion in Acute Myocardial Infarction (DIGAMI) 2 Study. Diabetologia 2011; 54: 1308-1317. CrossRef |PubMed

10 Currie CJ, Poole CD, Evans M, Peters JR, Morgan CL. Mortality and other important diabetes-related outcomes with insulin vs other antihyperglycemic therapies in Type 2 Diabetes. J Clin Endocrinol Metab 2013; 98: 668-677. CrossRef | PubMed

Source: Lancet