Utility of models for estimating cardiovascular risk in people with diabetes

Background and objective: recommendations pertaining to cardiovascular risk stratification in diabetes are conflicting. This review addresses the utility of cardiovascular risk models in people with diabetes. Methods: the discriminative power of major cardiovascular risk factors was compared in patients with and without diabetes. Opportunities for improving risk estimation in people with diabetes were examined. The performance of existing diabetes-specific and non-specific cardiovascular risk models in people with diabetes was reviewed. Results: major risk factors affect cardiovascular risk in patients with and without diabetes in a similar way, with no evidence of any significant statistical interaction between these factors and diabetes. Diabetes-specific parameters such as microvascular complications and taking into account cardio-preventative therapies can significantly improve risk estimation in diabetes. Existing cardiovascular risk models, both diabetes-specific and non-specific that have been tested, have a useful discriminative power but tend to overestimate risk in people with diabetes. Their impact on clinical care, patients' adherence to treatment and outcome remains poorly understood. Conclusions: in spite of the high-risk status conferred by diabetes, cardiovascular risk estimation using global absolute risk models is useful for prescription and intensification of preventive measures in people with diabetes. However, the choice of model should go to recent and more elaborate algorithms developed in cohorts of diabetic patients best reflecting contemporary diabetes care.