Diabetes mellitus (DM) is an established risk factor for atherosclerotic cardiovascular disease
(CVD), and patients with DM are at a two to four-fold higher cardiovascular risk, including myocardial infraction,
unstable angina, stroke, and heart failure. All of the above have arisen interest in CVD preventive strategies by
the use of non-invasive methods, such as risk scores. The most common approach is to consider DM as a CVD
equivalent and, therefore, to treat patients with DM in a similar way to those who required secondary CVD prevention.
However, this approach has been disputed as all patients with DM do not have the same risk for CVD,
and since other potentially important factors within the context of DM, such as DM duration, presence of albuminuria,
and comorbidities, should be taken into consideration. Thus, the second and third approach is the application
of risk models that were either developed initially for the general population or designed specifically
for patients with DM, respectively. This review summarizes the evidence and implications for clinical practice
regarding these scores. Up to date, several models that can be applied to the diabetic population have been proposed.
However, only a few meet the minimum requirement of adequate external validation. In addition, moderate
discrimination and poor calibration, which might lead to inaccurate risk estimations in populations with
different characteristics, have been reported. Therefore, future research is needed before recommending a specific
risk model for universal clinical practice in the management of diabetes.