Significance of non-traditional risk factors for CHD in diabetes: the ARIC study

Original article:
Nontraditional risk factors for coronary heart disease incidence among persons with diabetes: the Atherosclerosis Risk in Communities (ARIC) study. Saito I, Folsom AR, Brancati FL et al. Ann Intern Med 2000; 133: 81–91.

Summary
The investigators of the prospective Atherosclerosis Risk in Communities (ARIC) cohort study assess in this paper the association of several traditional and non-traditional risk factors with the incidence of coronary heart disease in subjects with diabetes mellitus.
A total of 1676 subjects with diabetes, but without known coronary heart disease, participated in the study. Several risk factors were recorded at baseline, and patients were followed up for a maximum of 8 years.
Of the entire cohort, 186 participants developed a new coronary heart disease event during follow-up (11%). The incidence of coronary heart disease in these diabetic subjects was positively associated with known classical risk factors such as hypertension, smoking, total serum cholesterol and low HDL cholesterol. An association was also found between incident coronary heart disease and waist-hip ratio, levels of HDL3 cholesterol, apolipoproteins (apo) A-1 and B, serum albumin and fibrinogen levels as well as von Willebrand factor, factor VIII activity and white blood cell count.
After applying statistical adjustment for traditional risk factors for coronary heart disease, only the levels of serum albumin, fibrinogen, von Willebrand factor, factor VIII activity and white blood cell count remained independently associated with coronary heart disease (Fig. 1). 

Fig. 1: Relative risks and 95% confidence intervals for incident coronary heart disease by tertiles or quartiles of selected baseline risk factors among diabetic participants in the ARIC study.

The authors postulate that these associations may either reflect the underlying inflammatory reaction or microvascular injury related to ath-erosclerosis and the tendency towards thrombosis, or they may reflect common antecedents for both diabetes and coronary heart disease.

Comment
Secondary macrovascular complications are frequently observed in patients with diabetes mellitus and can have severe consequences. The role of diabetes as an independent risk factor for cardiovascular disease is well established [1]. Type 2 diabetic patients have two- to fourfold increased morbidity and mortality attributable to coronary heart disease [2, 3]. In addition, they have reduced survival following myocardial infarction, coronary artery bypass grafting and, in particular, percutaneous transluminal angioplasty, and an increased occurrence of peripheral vascular disease.
Epidemiological and prospective studies in Type 2 diabetes have suggested a relationship between hyperglycaemia or decrease in metabolic control and the risk of coronary heart disease [4, 5]. Hypertension, hyperlipidaemia and smoking are classical risk factors which negatively influence the prognosis for patients with Type 2 diabetes. Yet, they do not completely explain the severely increased cardiovascular risks in diabetic patients.
Lipid disturbances that can be frequently observed in diabetic patients are very diverse [6]. Increases in the level of VLDL cholesterol have been observed as well as changes in the composition of this particle which render it more atherogenic. Also, postprandial hypertriglyceridaemia, characterized by increased postprandial levels of triglycerides and chylomicron random particles, lowering of serum HDL cholesterol, slight elevations of LDL cholesterol and an increase in the number of small dense LDL cholesterol particles have been observed in Type 2 diabetic subjects.
Several new lipid risk factors have been proposed for patients with diabetes. The apolipoproteins are the protein constituents of cholesterol particles. Apo A-1 is the most important protein component of HDL, as is apo B of LDL and VLDL. It has been shown that the serum apolipoprotein level is more stable than the level of cholesterol and triglycerides. Several epidemiological studies have indicated that apo A-1 and apo B are better predictors of the development of cardiovascular disease than cholesterol or the cholesterol particles LDL and HDL. The ARIC study assessed several of these non-traditional risk factors for coronary heart disease among diabetic patients. The investigators showed that participants with incident coronary heart disease had higher apo B levels and lower levels of HDL3 cholesterol and apo A-1 than participants without coronary heart disease. These findings confirm the results of previous studies, especially the 15-year follow-up of the Finnish cohort described in 1998 [7].
In the ARIC study several factors were assessed in relation to coagulation and inflammation. The results showed that participants with incident coronary heart disease also had higher levels of fibrinogen, von Willebrand factor, factor VII, factor VIII activity and white blood cell count and lower levels of serum albumin. Antithrombin III and protein C levels were not different. Also it was shown that after adjustment for sex, age and ethnicity, coronary heart disease incidence was positively associated with these factors. After adjustment for these factors, as well as the classic risk factors (total cholesterol, HDL cholesterol, hypertension, smoking), parameters of inflammation such as albumin level, fibrinogen and white blood cell count, as well as the correlation factors von Willebrand factor and factor VIII activity were still significantly related to incident coronary heart disease. The relative risks for the highest compared with the lowest quartile of these values were 1.71–1.90.
Acute-phase reactions as well as inflammatory markers have attracted much attention in recent years. It has been shown that the concentration of lipoprotein(a) [Lp(a)] is an independent indicator of the risk of cardiovascular disease. Although the function of Lp(a) is not yet known, it may be regarded as an acute-phase protein which significantly increases when there are considerable vascular abnormalities as a consequence of atherosclerosis. In this case the serum level of Lp(a) is a good marker of the existence of (as yet undiscovered) cardiovascular disease. A similar case may be made for the inflammation markers and coagulation markers described in the ARIC study. These markers of inflammatory reactions may reflect vascular injury and also play a role in the worsening of atherosclerosis. Higher levels of fibrinogen may reflect thrombin activation in the blood of those with diabetes, and several studies have indicated that von Willebrand factor is a marker of endothelial cell dysfunction, which promotes platelet adhesion and aggregation.
From the paper it is not evident whether these inflammatory and coagulation markers can be used in the clinic to further improve diagnostic and prognostic accuracy for coronary heart disease. Although there is a clear association between these markers and new coronary heart disease events, the differences between patients who develop and those who do not develop new coronary heart disease are only moderate. The average fibrinogen level was 3.44 g/l in the incident cases and 3.25 g/l in patients without coronary heart disease. Although the leukocyte count was highly statistically significantly greater in patients developing coronary heart disease, the average value in this group was only 7.2 x 109 cells/l against 6.6 in patients not developing coronary heart disease. In patients with a leukocyte count >7.7 x 109 cells/l, there was an almost twofold higher incidence of coronary heart disease than in patients with leukocyte counts <5.2 x 109 cells/l. However, since white blood cell count may vary in patients considerably from day to day and can also be influenced by several intercurrent diseases such as the common cold, it is doubtful whether this is a marker which can be used in daily practice to identify high-risk patients who need more intensive treatment of their traditional risk factors or who may be eligible for early treatment with, for instance, aspirin.
Another example can be found in the use of fibrates in dyslipidaemic states. Although some of these compounds have been shown not only to favourably influence dyslipidaemia but also to lower fibrinogen levels, there is no compelling evidence that fibrates which lower fibrinogen levels are more beneficial in preventing cardiovascular disease than fibrates which have no effect on fibrinogen levels.

References
1. Nathan DM, Singer DE, Godine JE, Perlmuter LC. Non-insulin-dependent diabetes in older patients — complications and risk factors. Am J Med 1986; 81: 837–42.
2. Fein FS, Scheuer J. Heart disease in diabetes. In: Porte Jr D, Sherwin RS, eds. Ellenberg and Rifkin’s diabetes mellitus. 5th ed. New Hyde Park, NY: Elsevier Science Publishing Co, 1997; 1105–26.
3. Jarrett RJ. Epidemiology and public health aspects of non-insulin-dependent diabetes mellitus. Epidemiol Rev 1989; 11: 151–71.
4. Kuusisto J, Mykkänen L, Pyörälä K, Laakso M. 
NIDDM and its metabolic control predict coronary heart disease in elderly subjects. Diabetes 1994; 43: 960–7.
5. Turner RC, Millns H, Neil HAW et al. Risk factors for coronary artery disease in non-insulin dependent diabetes mellitus: United Kingdom Prospective Diabetes Study (UKPDS 23). Br Med J 1998; 316: 823–8.
6. Betteridge DJ. Diabetic dyslipidemia. Am J Med 1994; 96: 25–31.
7. Niskanen L, Turpeinen A, Penttila I, Uusitupa MI. Hyperglycemia and compositional lipoprotein abnormalities as predictors of cardiovascular mortality in type 2 diabetes: a 15-year follow-up from the time of diagnosis. Diabetes Care 1998; 21: 1861–9.

Summary and Comment:
B.H.R. Wolffenbuttel, Maastricht, 
The Netherlands