Mediation effect analysis of lipoprotein levels on BMI and cardiovascular outcomes in patients with heart failure

Abstract

Background: Among heart failure patients with obesity, the prognosis is better than those with normal weight, a phenomenon known as the obesity paradox. However, it is unclear whether lipoprotein levels play a mediating role in the machine of the obesity paradox.

Methods: The study included 1663 heart failure patients hospitalized from January, 2019 through August, 2022. Kaplan-Meier survival analysis and Log-rank tests were performed for three endpoints in order to determine cumulative event-free survival. We investigated the correlation between Body Max Index (BMI) and outcomes by multifactorial Cox models. Mediation analysis was applied to study the presence and magnitude of mediation effects of triglyceride, total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, apolipoprotein A1 and apolipoprotein B, with the association between BMI and endpoints.

Results: In MACCEs, the median follow-up period was 679 days. In Cox model, compared with the underweight group, a high BMI level was significantly associated with lower all-cause mortality (HR=0.47, 95%CI 0.31~0.69, p<0.001, obese vs underweight), cardiovascular mortality (HR=0.46, 95%CI 0.30~0.73, p<0.001, obese vs underweight) and the incidence of MACCEs (HR=0.68, 95%CI 0.53~0.88, p=0.003, obese vs underweight). Mediation analysis revealed that TG was the strongest mediator between BMI and endpoints, with proportions of mediated effects of 6.6% (95%CI 2.2%~18.0%, p=0.0258, in all-cause death),7.0% (95%CI 2.3%~18.9%, p=0.0301, in cardiovascular death) and 10.2% (95%CI 3.3%~27.4%, p=0.0185, in MACCEs).

Conclusions: There is an "obesity paradox" in patients with heart failure, and lipoprotein levels especially triglyceride mediate the association between BMI and cardiovascular outcomes.

Keywords: Heart failure; Lipoprotein; Mediating effect; Obesity.

Fig.1

Flow diagram of patient selection

Fig. 2

Differences of each lipid indicators in different BMI groups. BMI1: underweight group, BMI2: normal group, BMI3: overweight group, BMI4: obese group

Fig. 3

Kaplan-Meier survival curves for all-cause death (A), cardiovascular death (B), and MACCEs (C) in four groups of patients with BMI levels. BMI1: underweight group, BMI2: normal group, BMI3: overweight group, BMI4: obese group

Fig. 4

Multifactorial Cox regression modeling with MACCEs outcome as endpoint event

Fig. 5

Restricted cubic spline showing the relationship between BMI as a continuous variable and the risk ratios for all-cause mortality (A), cardiovascular death (B), and MACCEs (C)