Combined associations of obesity and metabolic health with subclinical left ventricular dysfunctions: Danyang study

Abstract

Aims: The association of strictly defined metabolic healthy obese (MHO) with subclinical cardiac function was unclear. Our study aims to examine the role of MHO in subclinical cardiac dysfunction in a Chinese population.

Methods and results: The study subjects were recruited from Danyang from 2017 to 2019. Obesity was defined by body mass index (BMI) categories (normal weight, overweight and obesity). Metabolic health was strictly defined as having neither any of the guidelines recommended metabolic syndrome components nor insulin resistance. Thus, subjects were grouped by BMI categories and metabolic health status as six groups. Preclinical systolic (global longitudinal strain [GLS]) and diastolic function were assessed by 2D speckle tracking, and transmitral and tissue Doppler imaging, respectively. The 2757 participants (mean age ± standard deviation, 52.7 ± 11.7 years) included 1613 (58.5%) women, 999 (36.2%) obese, 2080 (75.4%) metabolically unhealthy and 93 (3.4%) MHO participants. After adjustment for covariates, the trend was similar for left ventricular (LV) ejection fraction (P_trend ≥ 0.07) but significantly worse for GLS, e' and E/e' (P_trend ≤ 0.02) across the six groups or passing from normal weight to obese individuals irrespective of metabolic status. MHO participants had lower GLS (20.4 vs. 21.4%) and e' (9.6 vs. 10.6 cm/s) compared with controls (P < 0.0001) but had similar GLS (P = 0.47) compared with metabolically unhealthy obese (MUO). Regardless of obesity status, metabolically unhealthy participants had worse diastolic function compared with their metabolically healthy counterparts (P ≤ 0.0004). Compared with controls, MHO individuals were at higher risk of subclinical LV systolic dysfunction (OR = 3.44, 95% CI = 1.25-9.49, P = 0.02). These results were robust to sensitivity analysis.

Conclusions: MHO was substantially associated with worse subclinical systolic function although early diastolic dysfunction seemed to be more accentuated in MUO.

Keywords: Insulin resistance; Metabolic status; Obesity; Subclinical myocardial abnormalities.

Figure 1

Comparisons of crude mean values of systolic and diastolic function measures among study groups. (A) Left ventricular ejection fraction. (B) Global longitudinal strain. (C) e′. (D) E/e′ ratio. White and black bars indicate mean values of echocardiography parameters in metabolically healthy and unhealthy individuals, respectively. The P‐value for trend across six groups (in the order of MHNW, MHOW, MHO, MUNW, MUOW and MUO) and across obesity groups (in the order of normal weight, overweight and obese) stratified by metabolically healthy status are given, respectively. *P < 0.05; **P < 0.01; ***P < 0.001; metabolically unhealthy vs. healthy counterparts. ^& P < 0.05; ^&& P < 0.01; ^&&& P < 0.001; obese vs. normal weight individuals. MHNW, metabolically healthy normal weight; MHOW, metabolically overweight; MHO, metabolically healthy obese; MUNW, metabolically unhealthy normal weight; MUOW, metabolically overweight; MUO, metabolically unhealthy obese. Metabolic health was defined as individuals having none of five metabolic syndrome components and without insulin resistance.

Figure 2

Comparisons of adjusted mean values of systolic and diastolic function measures among study groups after multivariable adjustment. (A) Left ventricular ejection fraction. (B) Global longitudinal strain. (C) e′. (D) E/e′ ratio. White and black bars indicate adjusted mean values of echocardiography parameters in metabolically healthy and unhealthy individuals, respectively. The analysis was adjusted for age, sex, heart rate, current smoking and alcohol drinking, education, physical activity, γ‐glutamyltransferase, estimated glomerular filtration rate, LVMI, LVEF (except for measure of LVEF) and e′ as a marker of diastolic dysfunction (except for measures of diastolic function). The P‐value for trend across six groups (in the order of MHNW, MHOW, MHO, MUNW, MUOW and MUO) and across obesity groups (in the order of normal weight, overweight and obese) stratified by metabolically healthy status are given, respectively. *P < 0.05; **P < 0.01; ***P < 0.001; metabolically unhealthy vs. healthy counterparts. ^& P < 0.05; ^&& P < 0.01; ^&&& P < 0.001; obese vs. normal weight individuals. MHNW, metabolically healthy normal weight; MHOW, metabolically overweight; MHO, metabolically healthy obese; MUNW, metabolically unhealthy normal weight; MUOW, metabolically overweight; MUO, metabolically unhealthy obese. Definition of metabolic health is the same as Figure 1 .

Figure 3

Multivariate odds ratios of subclinical left ventricular systolic dysfunction in study groups. The metabolically healthy normal weight participants were considered as a reference group. The analysis was adjusted for age, sex, heart rate, current smoking and alcohol drinking, education, physical activity, γ‐glutamyltransferase, estimated glomerular filtration rate, LVMI, LVEF and e′ velocity. Definition of metabolic health is the same as Figure 1 .