Screening strategies for dysglycaemia in relation to cardiovascular risk

Abstract

Background: Dysglycaemia, defined as type 2 diabetes (T2DM) or impaired glucose tolerance (IGT), is a major risk factor for cardiovascular disease (CVD). Dysglycaemia and CVD together account for more than half of worldwide mortality. Despite abundant evidence stressing the need for early detection and improved cardiovascular prevention in patients with dysglycaemia, approximately 50% of people with T2DM are undiagnosed. The early identification of dysglycaemia is key to initiating lifestyle changes and pharmacological interventions successfully to prevent cardiovascular complications.

Aims: The general aim of this thesis was to evaluate different screening strategies for dysglycaemia with respect to CVD. There were four specific goals. 1) Compare the prognostic value of fasting plasma glucose (FPG), 2h postload glucose (2hPG) and HbA1c in patients with coronary artery disease (CAD) (Study I). 2) Study the prevalence of dysglycaemia in people treated for hypertension and/or dyslipidaemia but without CVD and examine whether using the Finnish Diabetes Risk Score (FINDRISC) decreases the need for blood tests in this patient population (Study II). 3) Examine the effectiveness of different outreach strategies to detect people at high risk of T2DM and/or CVD using FINDRISC (Study III). 4) Study the accuracy of a point-of-care technique used for glucose measurement compared with laboratory-based methods (Study IV).

Methods: Studies I and II were based on data from the EUROASPIRE IV survey, a multicentre study of two large cohorts of patients, providing both questionnaire data and blood samples: one patient cohort with established CVD and one with hypertension and/or dyslipidaemia but without established CVD. In Study III, FINDRISC questionnaire data were collected through several outreach channels in the municipality of Södertälje. Response rate, risk level and cost effectiveness were compared across channels. In Study IV, glucose measurement data from the same patients using different laboratory methods were obtained from the EUROASPIRE V survey.

Results: In Study I, 4,004 patients with CAD but no history of dysglycaemia at baseline were followed for a median of 2.03 years. The 2hPG was a significant predictor of future cardiovascular events (p=0.01) in contrast to FPG (p=0.45) or HbA1c (p=0.36). In Study II, of 2,395 patients with hypertension and/or dyslipidaemia but no history of dysglycaemia, 19% had undetected T2DM and 20% IGT. Contrary to the hypothesis, a relatively large proportion of patients with low FINDRISC scores had dysglycaemia. In Study III, the highest response rate to the FINDRISC was obtained through workplaces. The largest proportion of individuals at high risk (≥15/26 points) were outreached through the Syrian Orthodox church. The cost of identifying a person at high risk varied greatly depending on the choice of outreach strategy. Finally, in Study IV, glucose values obtained from the HemoCue® point-of-care system correlated well with gold standard hospital laboratory measurements (FPG r=0.94 and 2hPG r=0.96; p<0.05) in CAD patients in the EUROASPIRE V survey.

Conclusions: Future cardiovascular events were only predicted by 2hPG in the present cohort of CAD patients. A large proportion of patients with CVD risk factors but without established CVD had undetected dysglycaemia. The use of the FINDRISC did not reduce the need for blood testing in patients with hypertension and/or dyslipidaemia. The outreach strategy affected participation rate, risk level and cost in a large-scale dysglycaemia screening. The HemoCue® point-of-care system is accurate for dysglycaemia screening, which should save time and costs when performing OGTT.