Obesity is a serious health problem, which alters the state of metabolism, leading to dyslipidemia, insulin resistance, and inflammation, and is therefore an important risk factor for cardiovascular diseases (CVD) and type 2 diabetes (T2D) (1). However, individuals in the same body mass index (BMI) category can have substantial heterogeneity of metabolic features. Individuals of “metabolically healthy obesity” (MHO) phenotype may not be at an increased risk for cardiovascular complications, and have a favorable lipid and glucose profiles (2, 3). In contrast, not all non-obese individuals present with a healthy metabolic profile, and were described to have “metabolically unhealthy non-obesity (MUNO)” (3, 4). If people with different healthy status could be identified, this could mean great benefits both for the individual and the health care system.

Body composition is closely related to metabolic health status. Previous studies indicated that MHO individuals were characterized by lower visceral fat mass, less fat accumulation in liver and skeletal muscle as compared to metabolically unhealthy obese (MUO) individuals (5–7), emphasizing the role of adipose tissue function in metabolic health. The effects of subcutaneous adipose tissue on metabolic health status are complex. Some studies supported beneficial effects of subcutaneous adipose tissue on metabolic health status while others not (8–13). Moreover, abdominal subcutaneous adipose tissue mass and gluteofemoral fat mass might have different effects on metabolic health status. German Tübingen diabetes family study showed that more metabolically unhealthy patients had a high percentage of subcutaneous abdominal fat mass, and a low percentage of subcutaneous gluteofemoral and leg fat mass, than did metabolically healthy patients (12, 13). However, the Dallas Heart Study did not show that subcutaneous abdominal fat mass was significantly associated with an increased risk (10). Meanwhile, previous studies showed conflicting results on the contribution of skeletal muscle mass to the pathogenesis of insulin resistance and metabolic status (14–17). Abdominal fat distribution and skeletal muscle mass vary significantly between racial and ethnic groups (18), however, few studies have investigated the abdominal fat distribution and skeletal muscle mass in different metabolic phenotypes in a population-based study with a large sample size in China.

The aim of this study was to investigate the characteristics of abdominal fat distribution and skeletal muscle measured by using quantitative computed tomography (QCT) in different metabolic status in Chinese participants. We explored whether higher abdominal visceral adipose tissue (VAT) and lower abdominal subcutaneous adipose tissue (SAT) and skeletal muscle index (SMI) were associated with an increased risk of metabolically unhealthy phenotype. We also assessed the ability of abdominal fat distribution and skeletal muscle mass to identify subjects with different metabolically phenotypes.

In this population-based study using CT techniques to ascertain body composition and its relation with metabolically healthy status in a large sample of Chinese population, we found 63.4%, 39.5%, and 23.3% participants were classified as MHNW, MHOW, and MHO, respectively. Higher levels of abdominal visceral adipose tissue were closely associated with metabolically unhealthy phenotype in Chinese adults with different BMI categories. Abdominal subcutaneous adipose tissue might be a protective factor for metabolic health status only in obese individuals. However, there was no significant association between the levels of SMI and metabolic status in different BMI categories.

The mechanisms underlying metabolic health status are complex. The different impact of specific fat compartments on insulin sensitivity and lipid levels was well described, with excess abdominal VAT imparting a greater risk of metabolic syndrome than excess abdominal SAT. Visceral adipose tissue acted not only as a fat-deposit site, but also as a highly secretory organ with a differential production of adipokines capable of regulating lipid metabolism, insulin sensitivity, and inflammation (23). Previous studies found MHO individuals had significantly lower levels of visceral fat than postmenopausal women with MUO phenotype (7, 24). In our study, we used QCT, an accurate method, to assess abdominal fat area, and we found abdominal VAT was closely associated with metabolically unhealthy phenotype in different BMI categories. These findings supported the opinion that a smaller amount of abdominal VAT was a protective factor in the maintenance of the favorable metabolic profile.

Some evidence indicated ectopic fat accumulation was a plausible mechanism for metabolic syndrome (25). The liver was an important location for ectopic fat deposition. Norbert Stefan et al. found ectopic fat in the liver may be more important than visceral fat in the determination of a beneficial phenotype in obesity (13). Though we did not assess ectopic fat accumulation in the liver, lower concentrations of hepatic enzymes in metabolically healthy individuals may reflect lower hepatic insulin resistance and lower liver fat content (23). However, we also need to emphasize that liver fat content is high but insulin resistance and the risk of diabetes are not in patients with fatty liver disease with a strong hepatic genetic component (ie, the 148Met allele in PNPLA3 and the 167Lys allele in TM6SF2) (26, 27). Future studies will be required to unravel underlying mechanisms of ectopic fat accumulation and its contribution to metabolic unhealthy phenotype in Chinese population.

Massive expansion and remodeling of adipose tissue during obesity differentially affects specific adipose tissue depots and significantly contributes to metabolic unhealth status (28). Subcutaneous adipocytes have distinctly different gene expression patterns (higher adiponectin expression and lower expression of proinflammatory adipokines), are better differentiated, and have increased adipogenesis and browning potential compared with visceral adipocytes (29, 30). In most obese individuals, the SAT might fail to expand to store energy surplus, and lipids might accumulate ectopically in VAT, liver, and skeletal muscle; whereas in healthy obese individuals, SAT had the intrinsic ability to expand, leading to preserved insulin sensitivity (31, 32). Observations of Framingham Heart Study showed that the inability to store fat in subcutaneous adipose tissue depot increases the propensity for visceral fat storage (33). It was found that regulation of lipid storage-related genes was defective in the SAT of subjects exhibiting the largest fat accumulation in the VAT of non-obese subjects when overfed for 56 days (32). In our study, abdominal SAT might be a protective factor for MHO, which might be explained by a greater ability of MHO individuals to store free-fatty acids in the SAT instead of in ectopic fat depot. However, the levels of abdominal SAT were similar in metabolically healthy and metabolically unhealthy individuals with normal weight and overweight. This finding was not consistent with previous studies in Caucasians. A low percentage subcutaneous leg fat mass can be found with high prevalence in both unhealthy normal weight subjects and overweight and obese subjects (12). In addition, low percentage leg fat mass, followed by fatty liver, is the strongest independent predictor of metabolic risk in normal weight subjects, however, it is not a significant determinant of metabolic risk in obese subjects (12). The associations between subcutaneous fat at different sites and metabolic status warrant further investigation.

Muscle mass and strength were protective factors against cardiometabolic risk and metabolic syndrome, although the conclusions were inconsistent (34, 35). Protective effects of muscle mass were likely due to mechanisms involving glucose utilization, myokines secretion and ectopic fat accumulation (36, 37). However, in this study, we did not find significant association between the levels of SMI and metabolic status. In addition, muscle and fat ratio (SMA/TAT ratio) was not a more useful indicator than VAT or VAT/TAT ratio for screening metabolic syndrome. Further studies on potential associations and mechanisms underlying skeletal muscle and metabolic status were needed.

In our study, we found MHO individuals represent 23.3% of the adult obese population, with higher prevalence in women, which was similar to previous studies (38–40). Though, MHO individuals present a favorable blood lipid profile, favorable hepatic enzyme profile, lower blood pressure and lower insulin resistance compared with MUO individuals, MHO individuals have an intermediate-stage of cardiovascular risk profile that is between MUO and non-obese phenotypes, and might shift to an MUO phenotype with time (41–43). Hwang et al. demonstrated that a higher conversion to MUO was associated with greater visceral abdominal fat, female gender, higher fasting insulin levels, and lower baseline HDL-C levels (44). Some researchers even suggested that using the label ‘metabolically healthy’ to describe this group in clinical medicine might be misleading (43). Therefore, a more profound understanding of the underlying metabolic regulation in MHO and MUO needed further research. Importantly, it was necessary to encourage exercise and weight control for MHO population to prevent metabolic related disorders due to its transitory state.

Our findings had important clinical implications and public health importance. We found that patients within the same BMI range had heterogeneous phenotypes. We also found that obese subjects or normal weight subjects ran different risks of complications. Therefore, stratification of individuals, based on metabolic health status, would help to identify high-risk subjects and to optimize prevention and treatment strategies in order to bring benefits to the individual, and lessen the burden on the health care system. According to this study, the analysis of ROC curves demonstrated that abdominal VAT and VAT/TAT ratio might help to predict metabolic healthy status. It meant that patients with normal weight and overweight might have metabolic syndrome if the VAT and VAT/TAT ratio exceeded cut-off levels, and screening interim might be shortened for risk prevention of complications in these patients. More importantly, there might be metabolism healthy subjects if the VAT and VAT/TAT ratio was below cut-off levels for obese individuals. In these subjects, the intensity of treatment and the frequency of screening might be weakened.

There were several strengths of our study. Firstly, this was the first large population-based study focusing on the associations of adipose distributions and muscle mass measured by QCT with the relation of metabolic status in different BMI category among Chinese adults. Secondly, most previous studies used waist to height ratio, waist to hip ratio or indices measured by bioelectrical impedance analyzer or dual-energy x-ray absorptiometry as measures of body composition (23, 45). In our study, we used QCT, which was considered as the gold standard for quantifying and comparing regional fat distribution and skeletal muscle. Thirdly, most previous studies evaluated characteristic of MHO and MUO (41). In our study, we evaluated not only obese individuals, but also normal weight and overweight participants according to metabolic status.

However, our study has several limitations. Firstly, given the inconsistency of metabolically healthy definitions, there was high degree of variability surrounding the estimated prevalence of different phenotypes. Secondly, we did not measure hepatic fat content, and only used serum activities of hepatic enzymes to represent hepatic fat content. Gluteofemoral fat mass had effects on metabolic health status, however, we did not measure lower-body fat mass in our study. In addition, we measured skeletal muscle mass without considering the muscle quality. Good-quality muscle and poor-quality muscle might have different contributions to metabolic status. Thirdly, chronic or acute diseases might influence BMI and metabolic health status. However, comorbidity and associated pharmacological treatment were not described in detail in this study. Fourthly, human obesity is associated with an imbalance in adipokines, which could act as classic hormones affecting the metabolism of tissues and organs. What is more, adipokines may decrease the insulin sensitivity of tissues and induce inflammation and development of chronic complications. However, we did not measure adipokines in our study. Fifthly, the cross-sectional design did not allow us to elucidate the causal relationship between abdominal fat distribution and skeletal muscle and the metabolically phenotypes.

In conclusion, metabolically healthy individuals were relatively protected against metabolic disease as compared to metabolically unhealthy individuals, which was at least partly due to a better adipose tissue function and less ectopic fat storage. Measurement of abdominal fat distribution and skeletal muscle may provide a more complete understanding of metabolic risk.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding authors.

The studies involving human participants were reviewed and approved by Peking University Health Science Center. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

LJ and XC conceptualized this study and designed the systematic review protocol. FL, YL, XYZ, XH, XHZ, and ZF performed the study selection and data extraction. FL and XC performed the statistical analyses. FL and XC prepared the outlines and wrote the manuscript. All authors contributed to the critical revision of manuscript drafts.

This work was supported by National Natural Science Foundation of China (No.81970698, No.81970708, and No.81900805), and Beijing Natural Science Foundation (No.7202216). The funding agencies had no roles in the study design, data collection or analysis, decision to publish or preparation of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.