Atherosclerosis-related disease, coronary heart disease (CHD), peripheral vascular disease (PVD), and thrombotic stroke are major complications in people with type 2 diabetes mellitus [1]. A recent meta-analysis of 102 prospective studies demonstrated a hazard ratio of 2 for coronary death and non-fatal myocardial infarction (MI) and 2.5 for ischemic stroke [2]. In the United Kingdom Prospective Diabetes Study (UKPDS), for each 1% increase in HbA1c there was a 28% inc rease in PVD [3].

The main focus for CVD risk management relates to patients with type 2 diabetes, but the increased lifetime risk for those with type 1 diabetes should be remembered when considering lipid lowering, particularly those with albuminuria, hypertension, and chronic kidney disease [4].

The pathogenesis of atherosclerosis in diabetes is multifactorial and the task for the physician is to manage all modifiable risk factors to prevent CVD events. However, it is clear from prospective studies that plasma cholesterol and low-density lipoprotein (LDL)-cholesterol in particular are major independent risk factors. In the United Kingdom Prospective Diabetes Study (UKPDS) of newly presenting patients with type 2 diabetes, LDL-cholesterol was the strongest predictor of MI. The second strongest predictor of MI was low levels of high-density lipoprotein (HDL)-cholesterol ahead of glycated hemoglobin, systolic blood pressure, and cigarette smoking [5].

The dyslipidemia of metabolic syndrome, insulin resistance, and type 2 diabetes consists of both quantitative and qualitative lipid and lipoprotein abnormalities [6]. Moderate hypertriglyceridemia is accompanied by low levels of HDL-cholesterol and an increase in cholesterol-rich remnant particles of chylomicrons and very low-density lipoprotein (VLDL) metabolism. LDL-cholesterol concentrations reflect those of the background population. However, important qualitative changes are present in the LDL particle distribution, with the accumulation of smaller, denser particles that are thought to be more atherogenic [7].

This complex phenotype is present at the time of diabetes diagnosis as it is part of the metabolic syndrome and prediabetes. In an individual patient it will be influenced by gender and lifestyle factors, particularly central obesity, the degree of physical activity, poor glycemic control, cigarette smoking, and alcohol intake. In addition, other secondary causes including renal and hepatic dysfunction, hypothyroidism, and concurrent medication may have a significant effect. Concurrent primary dyslipidemias such as familial hypercholesterolemia, familial combined hyperlipidemia, and type III dyslipidemia should be identified and managed appropriately.

Although understanding of the impact of insulin resistance on lipid and lipoprotein metabolism has increased enormously, much remains to be learned. A basic abnormality is the overproduction of large VLDL from the liver, partly as a result of an increased flux of fatty acids from adipose tissue combined with lack of inhibition of VLDL assembly [8]. In the postprandial state, hepatic VLDL production is not suppressed and this, together with exogenous fat absorbed in the form of chylomicrons, saturates activity of the enzyme lipoprotein lipase (LPL). LPL activity itself can also be reduced by increased levels of apoprotein C-III, apoprotein A-V, excess levels of fatty acids, low adiponectin levels, and insulin resistance.

Prolongation of the postprandial phase of lipid metabolism is associated with increased cholesterol and triglyceride exchange through the activity of cholesterol ester transport protein (CETP). CETP facilitates a mole-for-mole transfer of cholesterol esters from HDL to VLDL, IDL and chylomicron remnants, and LDL in exchange for triglycerides. As a result, LDL and HDL are triglyceride enriched and become substrates for the enzyme hepatic lipase, the activity of which is increased in diabetes. As a result of the triglyceride hydrolysis by this enzyme, LDL and HDL become smaller and denser. Smaller, denser HDL particles are cleared more rapidly, contributing to the low plasma levels observed [7, 9].

It is those patients with diabetes and concomitant metabolic syndrome including dyslipidemia that are at highest risk. In the National Health and Nutrition Examination (NHANES III) performed in the USA, the prevalence of metabolic syndrome in diabetes was 86%. The prevalence of CHD in this group was 19.2%. In those with diabetes and no evidence of metabolic syndrome, CHD prevalence was 7.5%, which is comparable to those without diabetes or metabolic syndrome [10].

Many studies in different populations have confirmed that dyslipidemia is a common finding in type 2 diabetes. The prevalence of low HDL-cholesterol (<0.9 mmol/l in men; <1.0 mmol/l in women) and/or raised triglycerides (>1.7 mmo/l) was increased about threefold compared to the background population in the Botnia study from Finland [11]. In a Canadian study, the prevalence of dyslipidemia ranged from 55% to 66% depending on the duration of disease: the longer the diabetes duration, the higher the prevalence of dyslipidemia [12].

LDL-cholesterol concentrations are generally similar to those of the background population. However, LDL-cholesterol remains a major risk factor and was indeed the best predictor of risk of MI in the UKPDS [5]. Qualitative changes in LDL particles increase their atherogenicity. The particles are smaller and denser with less lipid core. Parts of the apoprotein B molecule are exposed which have increased affinity to glycosaminoglycans. As a result, the particles are more likely to be retained in the subintimal space of the artery. Small, dense LDL are also more susceptible to oxidation, and it is oxidized LDL that is central to the development of atherosclerosis. Glycation of apoprotein B may also contribute to the increased atherogenicity [6].

HDL-cholesterol concentrations are inversely related to the risk of CVD events. In UKPDS, low HDL was the second best predictor of MI risk [5]. Baseline HDL concentrations remain a significant risk predictor in the major CVD outcome trials with statins, even in those subjects who achieved LDL-cholesterol concentrations <1.8 mmol/l [13]. The mechanism(s) by which HDL protects remains to be fully understood, although its role in reverse cholesterol transport has received considerable attention. Other potential mechanisms include antioxidant, anti-inflammatory, and antithrombotic effects [14].

The relationship of plasma triglycerides to CVD risk remains unresolved. Present in univariate analyses, the relationship is not maintained after other factors are adjusted for, particularly non-HDL-cholesterol [15]. Remnants of triglyceride-rich lipoproteins, enriched in cholesterol through lipid exchange mediated by CETP in prolonged postprandial lipemia, are atherogenic, as they are rapidly taken up by arterial wall macrophages to form foam cells. In several studies including the more recent FIELD and ACCORD studies, subjects with raised triglyceride (>2.3 mmol/l) and low HDL-cholesterol (<0.9 mmo/l) have been shown to be at higher CVD risk. Clearly, these parameters are intimately linked through postprandial lipemia [16, 17]. In the Copenhagen General Population Study, which included over 2,000 subjects with diabetes, nonfasting triglyceride concentrations were highly predictive of CVD events independent of other factors [18]. This relationship probably reflects the link between nonfasting triglycerides and remnant lipoprotein cholesterol.

Management of dyslipidemia should be part of overall CVD risk prevention, with attention to all modifiable risk factors. A lipid profile including total cholesterol and triglycerides, HDL-cholesterol with calculation of LDL-cholesterol by the Friedwald formula generally provides sufficient information for clinical management. Non-HDL-cholesterol is an important measure readily calculated by subtracting HDL-cholesterol from total cholesterol; this value is closely correlated with measurements of apoprotein B and therefore the number of atherogenic particles. It is often inconvenient for patients to fast for these measurements and this is not crucial, as apart from triglycerides, nonfasting concentrations do not differ significantly. Furthermore, as has been discussed, nonfasting triglycerides appear to be a strong CVD predictor.

As already discussed, the lipid phenotype may be influenced by other primary and secondary dyslipidemias [19].These other conditions should be diagnosed and treated appropriately. In the individual patient poor glycemic control, central obesity, excess alcohol intake, suboptimal diet and lack of physical activity are common and open to lifestyle intervention. It cannot be overemphasized that lifestyle measures should be the cornerstone of therapy in the management of vascular risk. The reader is referred to a comprehensive review of the topic [20].

Are all patients with type 2 diabetes at sufficient CVD risk (20% 10-year CVD risk) to receive pharmacotherapy for dyslipidemia? In the author’s opinion, risk calculation is not necessary, as most patients above the age of 40 years will fulfill this risk criterion. However, risk engines such as the one based on the UKPDS epidemiology data are available [21]. In the recent European Society of Cardiology/European Atherosclerosis Society guidelines for the management of dyslipidaemias [19], in patients with type 2 diabetes and CVD or chronic kidney disease (CKD), and those without CVD who are over the age of 40 years with one or more other CVD risk factors or markers of target organ damage, the recommended goal for LDL-cholesterol is <1.8 mmo/l and the secondary goal for non-HDL-cholesterol is <2.6 mmo/l. These guidelines also give a target for apoprotein B of less than <0.8 g/L. This, in the author’s opinion, is forward thinking and particularly helpful (if available) in diabetic dyslipidemia, as potentially atherogenic cholesterol is carried on lipoproteins other than LDL. There is one molecule of apoprotein B per particle of the VLDL, IDL, LDL cascade and its concentration therefore gives important information on particle numbers. For all other people with type 2 diabetes, an LDL-cholesterol <2.5 mmol/l is the primary target. The non-HDL-cholesterol target is below 3.3 mmol/l and apoprotein B <1.0 g/L. In this and other guidelines, different targets are set depending on the risk. The author fails to see the rationale for this and in his practice, once the decision to introduce pharmacotherapy has been taken, the more intensive target is applied to all.

Statins are first-line pharmacotherapy for diabetic dyslipidemia. Their use is based on a wealth of data from robust, randomized trials for both primary and secondary prevention of CVD events. First discovered in the 1970s by the Japanese scientist Dr. Akiro Endo, the introduction of these drugs into clinical practice in the 1980s enabled the first definitive CVD endpoint trials of cholesterol lowering to be performed. They act by decreasing hepatic cholesterol synthesis (by about 40%) by specific competitive inhibition of the rate-determining enzyme, HMG-CoA reductase, which catalyzes the first committed step in cholesterol synthesis. As a result, the expression of hepatic LDL receptors is increased, which bind and take up more plasma LDL, thereby decreasing plasma LDL. The Scandinavian Simvastatin Survival Study (4S) was the first landmark statin trial [22] performed in patients with established CHD (n = 4,444, 827 females). The primary endpoint was overall mortality. Simvastatin reduced LDL-cholesterol concentration by 35% and, after a mean follow-up of 5.4 years, there were 182 deaths in the treated group compared to 256 in the placebo group (HR 0.7; 95% CI 0.59–0.85; p < 0.0003). In addition, there were highly significant reductions in all coronary events.

In 4S 202 known diabetic patients (age 60 years, 78% male) were included and approximately half of those on placebo suffered a major coronary event during the study period [23]. In the simvastatin group, CVD events were reduced by 55% (p = 0.002). Numbers were too small to assess the effect on overall mortality, although there was a 47%, nonsignificant reduction. In a further analysis, additional diabetic patients (n = 483) were identified on the basis of a baseline fasting glucose >7.0 mmol/l [24]. In addition, 678 patients were identified with impaired fasting glucose (IFT) with glucose levels between 6.1 and 6.9 mmol/l. Major CHD events were reduced by simvastatin (HR 0.58; 95% CI 0.42–0.81, p < 0.001). The 28% reduction in overall mortality did not reach significance. In the IFT group there was a significant reduction in overall mortality (HR 0.57; 95% CI 0.31–0.91, p < 0.02) [24].