During the last 10 years the classification of diabetes as a CHD equivalent has helped to improve standards of diabetes care by underlining the importance of cardiovascular risk factor management . However, it is also obvious that this simplified concept falls short of adequately describing the true cardiovascular risk of a highly heterogeneous population [5, 6]. The presence of the MetS, a cluster of cardiovascular risk factors, may help to refine risk factor assessment and treatment in diabetic patients. In this study, 59 patients with long standing diabetes mellitus underwent whole body MRA with quantification of atherosclerosis by determining an overall atherosclerosis score. Using this score we investigated the degree of variation in macrovascular disease and found a significant association between the score and the presence of the MetS as well as a dose-response relationship between the score and the number of MetS-components.
Despite the general high risk of the population with an average diabetes duration of 20.7 years and additional cardiovascular risk factors in many patients, we found very pronounced differences in the degree of atherosclerotic burden. The WB-MRA based score ranged from 1.00 to 2.41 with a median of 1.18. More than one third of the population (n = 22) had a score value of 1.00, indicating that these patients had no atherosclerotic changes or at most wall irregularities in all scanned vessels. In contrast, all patients with a score above 2.0 (n = 5) had at least 3 completely occluded vessels and several other stenotic lesions. These results are in line with the perspective of diabetic patients as a heterogeneous population with increased but variable prevalence of atherosclerosis and cardiovascular events [25, 26].
In our study population a high score was significantly associated with age, HDL-cholesterol, MetS and CHD status. The number of components of the MetS was associated with a linear increase in the MRI-score. Furthermore, consistent with previous reports evaluating MRI-based atherosclerosis scores [27, 28], a high score was significantly associated with estimated cardiovascular risk using an established risk algorithm.
Despite significant progress in recent years with respect to risk factor modification particularly through the use of lipid lowering agents and drugs targeting the renin-angiotensin system , several studies have found that there is still a large gap between guidelines and clinical practice [30, 31]. Furthermore, recent landmark studies aimed at investigating an even more aggressive treatment of hyperglycemia, blood pressure and dyslipidemia failed to show additional benefits in diabetic patients [32, 33]. A more individualized risk assessment could hence allow clinicians to apply flexible treatment goals to their diabetic patients and help to focus their attention and resources on the patients most at risk among the diabetic population . Including the well known concept of MetS into cardiovascular risk assessment and treatment decisions may help to individualize and further improve diabetes care.
There are several limitations to our study. First the prognostic significance of the detected vascular lesions is uncertain, due to the lack of follow up data. The association with the UKPDS risk calculations also must be interpreted with caution since the risk engine is not validated for a diabetes duration of >20 years and has an uncertain validity for patients with established cardiovascular disease, although only patients with acute cardiovascular disease were excluded in the original UPKDS study . Furthermore, there was no examination of the coronary arteries. However, cerebrovascular disease and peripheral artery disease are both well established predictors of cardiovascular mortality  and also in this study, the score showed a strong association with CHD status. Other limitations of this study are the relatively small sample size and single on treatment measurements. These factors limit the significance of the statistical associations. Another limitation relates to the scoring system. First of all, changes in all arteries were treated equally. Thus, there was no "weighting" of arteries (for example: changes of fibular arteries were considered as relevant as those of carotid arteries). From a clinical point of view, changes in some vascular beds are obviously more relevant than changes in others. However, a correct weighting of arteries is not possible. Furthermore, the grading of the changes in the individual arteries was considered to be linear, although this is not the case. For example: "multi-segmental significant stenoses with at least one exceeding 50% of the vessel diameter" (grading 4) was considered to represent double the amount of atherosclerosis as "non significant stenosis" (grading 2). However, in order to create an overall score for each patient, location and severity of changes had to be described numerically.