Microvascular Residual Risk Studies

COMMENT

The results of this analysis imply that statin use is not associated with an increased risk of diabetes-related microvascular complications, specifically the risk for diabetic retinopathy, diabetic neuropathy and gangrene of the foot. Statin use had no benefit on diabetic nephropathy. 

Undoubtedly, the size, quality and coverage of the national registry are strengths of this study. Moreover, the finding that statin use was associated with an increased risk of diabetes (1·17, 1·14–1·21; p<0·0001), was consistent with other analyses which have shown a link between diabetes and risk for incident diabetes,1,2 and thus provides an external validity to the study. Despite this, there are a number of issues, notably the lack of data on important predictors of microvascular disease, such as HbA1c, urine albumin, blood pressure, as well as lipids. There are also issues with the definition of microvascular endpoints. Notably, gangrene of the foot is defined as one diabetes-related microvascular endpoint. However, impaired tissue perfusion in the extremities in patients with diabetes, resulting in development of gangrene, receives contributions from both microvascular and macrovascular disease.3

It is important to note that statins had no effect on the risk for diabetic nephropathy. This is not inconsistent with some studies, although some have suggested a slowing in the decline of estimated glomerular filtration rate in diabetes patients.4,5 Other lipids beyond LDL cholesterol are also implicated. Indeed, the recent REALIST Micro study showed a robust association between elevated triglycerides and low high-density lipoprotein cholesterol (HDL-C), components of atherogenic dyslipidaemia, and risk for diabetic kidney disease.6 These findings build on a growing body of evidence linking triglycerides and HDL-C, with risk for diabetic microvascular complications. Statins are undoubtedly effective in lowering low density lipoprotein cholesterol and have modest effects on triglycerides and HDL-C, translating to macrovascular benefit in patients with atherogenic dyslipidaemia, as shown by the 4S study.7 However, other lipid-modifying treatments, such as peroxisome proliferator activated-receptor alpha agonists, have greater efficacy in lowering triglycerides, and this may be implicated in preventive effects in risk for diabetic kidney disease, as suggested by REALIST Micro.6

Finally, a key point which is overlooked by this paper is that statin therapy does not prevent the development of microvascular complications in diabetes patients. Over the median follow-up of 2.7 years, 3% of statin users developed diabetic retinopathy, and 2% each developed diabetic neuropathy or diabetic nephropathy. Even with optimal multifactorial treatment, as reported by the STENO-2 study, up to 50% of patients with type 2 diabetes show development or progression of microvascular disease.8 Thus, as indicated by the authors, there is a clear need for randomised controlled studies in this area, so as to define the optimal therapy for reducing the high residual risk of diabetes-related microvascular complications that persists despite best evidence-based therapy.

Another confounder is the fact that statin use prior to diabetes diagnosis was analysed with respect to incident microangiopathy in the years following diabetes diagnosis. Since microangiopathy cannot, by definition, arise prior to diabetes diagnosis in new-onset diabetes, the design of the study cannot address a potential protective effect of previous statins use on future microangiopathies, since the conditions to generate microvascular damage were not present over the time period during which statins were administered. An appropriate trial should test whether prolonged statin use in the presence of hyperglycaemia (i.e. subsequent to diabetes diagnosis) decreases incident microangiopathy.