Cardiovascular complications are the leading cause of morbidity and mortality among patients with type 2 diabetes, and cardiovascular disease (CVD) risk is 2- to 8-fold higher in the diabetic population than it is in non-diabetic individuals of a similar age, sex and ethnicity [1, 2]. Furthermore, macrovascular complications are the largest contributor to the direct and indirect costs of diabetes [3]. However, when describing cardiovascular disease in diabetic patients, the patient population is lumped together as a heterogeneous group with CVD, without separation into subgroups according to the macrovascular disease type. Furthermore multiple vascular sites are often simultaneously involved in type 2 diabetic patients, which contributes to the highest cardiovascular risk in this population. To the best of our knowledge, no extensive clinical studies of type 2 diabetic subjects have stratified patients according to their macrovascular complications. The aim of our retrospective study was to evaluate macrovascular disease in a large cohort of type 2 diabetic adults by dividing the subjects into several subgroups according to the vascular areas that were involved. We also studied differences between these subgroups in their anthropometric, clinical and laboratory characteristics, as well as their MetS prevalence/severity.

The first goal of our study was to assess the prevalence of different macroangiopathic conditions in our patients. Our study demonstrates that 46.4% of our patient population has macrovascular disease. It is well known that CVD prevalence, incidence, and mortality are strikingly greater in the diabetic than in the non-diabetic population. According to the World Health Organization, CVD prevalence in diabetic patients ranges from 26 to 36% [6]. The significantly higher prevalence of macrovascular complications in our patients is likely linked to patient characteristics (average age, diabetes duration, poor glycaemic control over time) and due to our inclusion of NSCS patients. As expected, almost half of our patients had polydistrectual involvement.

Advanced atherosclerotic vascular changes are often preceded by impairment of endothelium-dependent vasodilation, vascular smooth muscle dysfunction and increased arterial stiffness. Today all of these factors are recognized as predictors of vascular dysfunction in T2DM patients [7]. Atherosclerotic disease usually causes systemic involvement, which is more frequent in the diabetic population [8]. Moreover, atherosclerosis in diabetic patients is different from that in non-diabetic subjects because both pathologic studies and angiographic reports in individuals with coronary heart disease and PAD have shown that diabetic patients have more blood vessels involved, with a more diffuse atherosclerotic lesion distribution [9, 10].

Genetic studies have also shown that different macrovascular phenotypes (as CAD and CBVD) and T2DM share a major linkage at the chromosome 12q24 locus. In this regard, the gene of proteasome modulator 9 (PSMD9) is linked to macrovascular pathology of T2DM [11].

Estimated twenty years incidence of different cardiovascular complication phenotypes (ischaemic heart, myocardial infarction, heart failure, cerebrovascular disease, amputation of lower limbs) was evaluated in a Mexican diabetic population by a simulation model indicating that a large portion of this diabetic population is at risk of myocardial infarction and cerebrovascolar disease in subsequent years [12]. Also in our study, among patients with macrovascular-isolated events, the majority of patients were placed in the CAD (16%) and CBVD (14.6%) groups.

PAD, CAD and CBVD risk factors are similar and are also typical atherosclerosis risk factors. These risk factors include smoking, dyslipidaemia, diabetes and hypertension. However, specific risk factors, such as genetic background, could be more important for the development of macro-vascular disease at certain sites. To the best of our knowledge, no study has compared the anthropometric, clinical and laboratory features in type 2 diabetic patients by stratifying patients according to the type of macrovascular involvement. For this reason, as a second study target, we tested the hypothesis that different macrovascular involvement types might correspond to different phenotypes. By analysing the anthropometric and clinical characteristics of subgroups of patients, we found that PAD and PVD patients were older, had had diabetes for a longer period of time, were more likely to have been smokers and had a lower weight, BMI and WC than did the patients in the other groups. A multivariate analysis demonstrated that age, male sex, diabetes duration and the use of statins were the most important independent PAD and PVD risk factors. It is noteworthy that the same characteristics were not found to be independent risk factors in the CAD group and that the average patient age in this group was comparable to that of the population without macrovascular complications. The question arises of whether a genetically determined predisposition alone may explain the early onset of cardiac involvement in the CAD group. In recent years, many genetic risk factors for both diabetes and coronary artery disease have been discovered through genome-wide association studies. Genetic aspects of diabetes, diabetic macrovascular complications and CAD may share mechanisms, leading to a common effector hypothesis. However, only a few genetic risk factors could be identified that modulate the risk for both conditions. Polymorphisms in the TCF7L2 and near the CDKNZA/B genes may be of great importance for CAD development because these genes modulate both conditions and are not necessarily related to hyperinsulinaemia or hyperglycaemia [13].

The complex interactions between genetic and environmental factors on cardiometabolic risk were analyzed by twin studies indicating that some cardiometabolic risk factors had strong heritability (as weight, waist circumference, SBP, DBP) while others were substantially influenced by environmental factors [14].

By analysing glico-metabolic parameters (HbA1c, fasting glucose, postprandial glucose and Δ post-preprandial blood glucose) we found no differences among the six subgroups. Despite a clear association between diabetes and atherosclerotic vascular disease, the underlying mechanism responsible for the two diseases is not fully understood. The relative importance of “non-glycaemic” risk factors and hyperglycemia “per sè” has always been debated [15]. Results on the causal relationship between hyperglycemia and macroangiopathy have been contradictory. Plenty of evidence suggest a significative relationship between HbA1c levels, post-prandial hyperglycemia, and risk of CV events and adverse outcomes especially in overweight and obese patients [16, 17] . However, recently three major studies, ACCORD [18], ADVANCE [19] and VADT [20] evaluated the impact of attaining euglycemia (ACCORD) or near-euglycemia (ADVANCE and VADT) in patients with long-lasting diabetes and high cardio-vascular risk. None of these studies, either individually or on pooled analysis, demonstrated any reduction in all cause or cardiovascular mortality, although a meta-analysis revealed a 15-17% reduction in the incidence of non-fatal myocardial infarction in those patients exposed to tight glucose control [21]. A higher mortality was observed in the intensive glucose control arm of ACCORD, leading to the premature termination of the glucose-lowering component of this study. The weak association between glycemic control and macro-vascular disease observed in UKPDS [22] has been confirmed and amplified by these recent intervention studies. Accordingly, our data show that glycated hemoglobin, post-prandial and fasting glucose were similar in each subgroup considered. However, the relationship between hyperglycemia and macrovascular complications is made even more complex by the potential role of epigenetic mechanisms, as the metabolic memory, by which a prior exposure to hyperglycemia predisposes diabetic patients to the continuing development of vascular diseases despite a subsequent good glycemic control [23]. Furthermore, in our study, referral bias (diabetic patients with poor metabolic control referred to us for the first time) did not allow to clarify the causal or temporal relationship among macrovascular events and glycemic control over time.

By analysing the lipid profile, we found a clear difference in HDL-C among the groups, which was lower in CAD, PAD, and PVD patients. These data are in accordance with other scientific evidence that support the importance of this macroangiopathy-associated risk factor. In a multivariate analysis, we also confirmed that reduced HDL-C values are independent CAD and PVD risk factors (the association was borderline significant for the PAD group). The importance of HDL-C as “target therapy” is now emerging from recent cardiovascular trials of CETP (cholesteryl ester transfer protein) inhibitors, such as Anacetrapib [24] and Evacetrapib [25]. This new pharmacological group is very promising for cardiovascular risk reduction either when administered alone or in conjunction with conventional therapy.

Many reports have suggested that metabolic syndrome may precede/predict vascular disease. It has been reported that insulin resistance and metabolic syndrome increase the risk of new cardiovascular events also in patients without known diabetes but with manifest arterial disease [26]. In type 2 diabetes, metabolic syndrome is highly prevalent and often precedes hyperglycaemia onset [27, 28]. Furthermore, insulin resistance and metabolic syndrome predict atherosclerosis in type 2 diabetic patients [29]. Our findings support the clinical relevance of MetS component detection, which may be a simple, quick tool to stratify diabetic patients according to the expected macrovascular complication severity (as a polydistrectual disease).

Microvascular complications were not equally present in the various groups. We found more diabetic nephropathy, as assessed by micro/macroalbuminuria or by calculating the eGFR in the CAD, PAD and PVD groups. Diabetic neuropathy was more frequent in the PAD and PVD groups, and diabetic retinopathy was strikingly present in the PAD group. As expected, patients in the CAD, PAD and PVD groups were the largest statin and antihypertensive drug users, and those in the PAD and PVD groups were more frequently treated with multiple insulin injections.

Our study shows that nearly half of diabetic patients, especially those who are elderly and male with a long disease duration, have polydistrectual atherosclerotic involvement.

The characterization of different phenotypes may have a clinical significance for the everyday clinician.

Firstly, different factors were associated with isolated coronary artery disease, like a mean age comparable to that of patients without macro-vascular complications and a high frequency of severe metabolic syndrome. These data raise the interesting hypothesis that coronary artery disease, metabolic syndrome and diabetes may be genetically linked in some individuals. Further studies may clarify whether this is because of genetic background and/or because coronary heart disease might be the first event in polydistrectual atherosclerotic involvement.

Secondly, low HDL-C values are a marker of PAD, CAD and PVD, confirming the importance of HDL-C as a potential therapeutic target. Alternatively, lifestyle modification strategies to prevent a low HDL-cholesterol should be implemented.