Improved heart failure screening and prevention in type 2 diabetes mellitus by cardiologist in rural areas

Abstract

Abstract Purpose Guidelines recommend screening in type 2 diabetes mellitus (T2DM) for unrecognised asymptomatic left ventricular systolic dysfunction (LVSD), reduced (HFrEF), mid-range (HFmrEF) or preserved (HFpEF) heart failure (HF) with periodic evaluation of natriuretic peptides (NTproBNP) or echocardiography (echo). Tissue doppler and speckle-tracking echo may be needed to facilitate HF diagnosis at earlier, more treatable stage. Methods 2185 consecutive asymptomatic T2DM (age 62 ±4 years, 62% women) patients (pts) were referred from January 2015 by rural diabetologists (dia) to echolab with elevated NTproBNP (>125 pg/mL) or 5-year HF risk score WATCH-DM ≥14 points. After exclusion of abnormal electrocardiogram or chest x-ray and significant valvular heart disease was compared continued usual care by diabetologist with randomly assigned prospective follow-up by regional cardiologist. Results At referral in both groups were similar baseline characteristics, 20% pts were treated with ACE-inhibitor (ACEi), 2% beta-blockers (BB). After cardiologist consultation ACEi and BB use increased significantly to 96% and 84% in both groups (all p <0.001). Global LVSD (EF≤50%) was detected in 8% at baseline and 29% pts after 5 years in diabetologist group, respectively 9% at baseline and 13% after 5 years in cardiologist group (p <0.001). Systolic dysfunction (S’≤6.0 cm/s) occured in 20% at baseline and 41% pts after 5 years in dia group, respectively 19% at baseline and 24% after 5 years in cardiologist group (p <0.001). Reduced global longitudinal (GLS<16%) observed in 26% at baseline and 59% pts after 5 years in dia group, respectively 26% at baseline and 36% after 5 years in cardiologist group (p <0.001). At baseline we found 1% pts with new-onset HF in both groups (only HFmrEF). After mean follow-up of 54 ±7 months in cardiologist group developed new HF in 4% pts (2% HFmrEF, 2% HFpEF) compared with 25% pts (3% HFrEF, 7% HFmrEF, 15% HFpEF) in diabetologist group (p <0.001). Significant (p <0.001) predictors of incident HFmrEF were systolic dysfunction (S’≤6.0 cm/s), hazard ratio (HR) 3.6, 95% CI 2.96-4.2) and reduced GLS (HR 4.6, 95% CI 3.8-5.2), all p <0.01. The strongest (p <0.001) predictors of incident HFpEF were new-onset atrial fibrillation (HR 2.5, 95% CI 1.2-3.7) and echo-detected left ventricular hypertrophy (septal and posterior wall thickness ≥13 mm, HR 1.7, 95% CI 1.2-2.1), all p <0.01. Conclusions For preventing HF progression is required additional research focused on cooperative care between diabetologists and specialised HF outpatient clinic.