Prognostic significance of growth differentiation factor-15 across the BMI spectrum in patients with chronic heart failure: Insights from the SUPPORT trial

Abstract

Abstract Background Obesity is a risk factor for cardiovascular diseases including heart failure (HF), yet patients with obesity have better outcomes than their leaner counterparts in HF (i.e. the obesity paradox). We sought to investigate whether growth differentiation factor-15 (GDF15), an emerging biomarker in HF reflecting inflammation and oxidative stress, is a prognostic indicator across the body mass index (BMI) spectrum in chronic heart failure (CHF). Methods Serum levels of GDF15 and its prognostic significance across BMI were examined in a total of 1,040 Japanese patients with CHF enrolled in a randomized clinical trial of the SUPPORT trial (supplemental benefit of an angiotensin receptor blocker in hypertensive patients with stable heart failure using olmesartan) (mean age 66.0±9.9 years, 26% female). Patients were categorized into four groups: underweight (BMI <18.5 kg/m2; n = 56), normal (BMI 18.5-22.9; n = 304), overweight (BMI 23-24.9; n = 257), and obese (BMI ≥25; n = 423) defined based on the WHO recommendation for Asian-Pacific population. Results The underweight group consisted of older patients with a higher prevalence of NYHA class III and reduced left ventricular ejection fraction (LVEF) compared to obese patients (mean age 69.0 vs 64.7 years, NYHA class III 17.9 vs 5.9%, and mean LVEF 49.3 vs 55.5%; all p<0.05). Median GDF15 level in the underweight group was higher (1,864 pg/mL [IQR 1,070-2,626]) but was similar in overweight and obese groups (1,285 [953, 1772] and 1,220 [888-1,789], respectively) compared to normal BMI group (1,276 [879-2,018]). During follow-up (median 10.3 years), 540 composite events of HF hospitalization or all-cause death occurred. The risk was higher in the underweight group (adjusted HR 1.8, 95%CI [1.3-2.5]) and lower in the obesity group (adjusted HR 0.8 [95% CI 0.6-0.9]) compared to the normal BMI group. An increased GDF15 incurred a consistently higher risk from underweight to obese groups; per unit increase in logGDF15 (pg/mL) was associated with 4.1, 2.4, 1.7 and 2.4 times higher risk, respectively. The optimal cut-off point for discriminating those with and without the composite events was 1,343 pg/mL in overall patients. This cut-off value performed well as a prognostic indicator across the continuous BMI spectrum including those with higher BMI (Figure). Conclusions The prognostic significance of GDF15 persisted across the BMI spectrum in patients with CHF. GDF15 may be useful for providing prognostic information across patients with a diverse range of BMI even in those with modest outcomes.HF events in high GDF15 across BMI