Uncovering Heart Failure in Type 2 Diabetes

Abstract

In existing disease-management-programs for type 2 diabetes there is no routine attention for uncovering latent or early stages of cardiovascular diseases, including heart failure. Between February 2009 and March 2010, 605 patients aged 60 years or over with type 2 diabetes in the south-west of the Netherlands participated in our study. The prevalence of previously unknown heart failure in these patients is high (27.7%), steeply rises with age, and is overall higher in women (31.0%) than in men (24.8%). The majority (83%) of the patients with newly detected heart failure had preserved ejection fraction (HFpEF); 17% had heart failure with reduced ejection fraction (HFrEF). Based on the findings of our study, we developed a diagnostic rule to detect or exclude heart failure in these patients. Variables from electronic medical files combined with items from history taking and physical examination provides a good to excellent accuracy with a C-statistic of 0.82; 95% CI 0.79-0.86. Both electrocardiography and natriuretic peptides had independent added value beyond the clinical model and increased the C-statistic to 0.86; 95% CI 0.83 to 0.89. Annual screening appears to be cost-effective. Using information from the medical record (age and comorbidities) and suggestive symptoms of heart failure performed best for a low willingness-to-pay threshold of €20,000 per QALY, a commonly used threshold in Europe. This strategy had an incremental cost-effectiveness ratio (ICER) of €6,115 for men and €6,318 for women compared to no screening. A screening strategy applying echocardiography in all patients performed better, but the ICER was much higher with €29.100 for men and €39.226 for women. Cost-effectiveness would be better if convincing mortality-reducing treatment for HFpEF becomes available. Both screen-detected HFrEF and HFpEF were associated with a clinically relevant lower health status in patients with type 2 diabetes. This was already observed at the time of screening and persisted during the 1-year follow-up period. Patients with screen-detected heart failure had a lower Euroqol (EQ5D)-score (0.73) than those without heart failure (0.85). Patients with HFrEF had a lower EQ5D-score (0.66) at baseline than those with HFpEF (0.74). Especially physical functioning was decreased in patients with screen-detected heart failure as measured with the short form 36. Also the prognosis of these patients was worse than of those without heart failure. After adjustment for age and gender, the hazard ratio for all-cause mortality was 1.5 (95% CI 0.8-2.7), for cardiac hospitalizations 2.2 (95% CI 1.5-3.3), and for the composite end-point combining these two 1.8 (95% CI 1.3-2.6). The prognostic impact of screen-detected HFrEF was much more evident than of screen-detected HFpEF. Undetected heart failure appears to be an important health problem. Screening seems feasible, and in newly detected cases of HFrEF prognostically beneficial drugs can be initiated. We concluded more attention should be paid to detect heart failure in older patients with type 2 diabetes. Future research should further focus on the cardiovascular risk stratification, the different pathophysiological pathways resulting in HFpEF, the exact diagnostic criteria, and new treatments that improve prognosis in patients with this ‘phenotype’ of heart failure.