ABSTRACT
INTRODUCTION: Patients with type 2 diabetes mellitus (T2DM) are at higher risk of developing cardiovascular diseases, and assessment of their cardiac risk is important for preventive strategies. PURPOSE: The Ministry of Health of Sri Lanka has recommended World Health Organization/International Society of Hypertension (WHO/ISH) charts for cardiac risk assessment in individuals with T2DM. However, the most suitable cardiac risk assessment tool for Sri Lankans with T2DM has not been studied. This study was designed to evaluate the performance of two cardiac risk assessments tools; WHO/ISH charts and UK Prospective Diabetes Study (UKPDS) risk engine. METHODS: Cardiac risk assessments were done in 2,432 patients with T2DM attending a diabetes clinic in Southern Sri Lanka using the two risk assessment tools. Validity of two assessment tools was further assessed by their ability to recognize individuals with raised low-density lipoprotein (LDL) and raised diastolic blood pressure in a cohort of newly diagnosed T2DM patients (n=332). RESULTS: WHO/ISH charts identified 78.4% of subjects as low cardiac risk whereas the UKPDS risk engine categorized 52.3% as low cardiac risk (P<0.001). In the risk categories of 10%-<20%, the UKPDS risk engine identified higher proportions of patients (28%) compared to WHO/ISH charts (7%). Approximately 6% of subjects were classified as low cardiac risk (<10%) by WHO/ISH when UKPDS recognized them as cardiac risk of >20%. Agreement between the two tools was poor (κ value =0.144, P<0.01). Approximately 82% of individuals categorized as low cardiac risk by WHO/ISH had higher LDL cholesterol than the therapeutic target of 100 mg/dL. CONCLUSION: There is a significant discrepancy between the two assessment tools with WHO/ISH risk chart recognizing higher proportions of patients having low cardiac risk than the UKPDS risk engine. Risk assessment by both assessment tools demonstrated poor sensitivity in identifying those with treatable levels of LDL cholesterol and diastolic blood pressure.
