Correlation Between Waist-to-Height Ratio and Skeletal Muscle Mass in Women with Obesity: A Cross-Sectional Study

Introduction

Obesity, a chronic disease characterized by pathological fat accumulation, is a growing global health crisis [1,2]. While its metabolic complications (eg, diabetes, cardiovascular diseases) are well-documented, the impact of obesity on skeletal muscle mass (SMM) – a key determinant of physical function and metabolic health – remains underrecognized in clinical practice [3,4]. The absolute SMM of individuals with obesity is similar to or greater than that of nonobese individuals of the same age and sex [5]. Fat infiltration into the visceral compartment and skeletal muscle is commonly observed with increased body fat, which is associated with muscle-catabolic derangements, altered muscle composition, and reduced peak force production [6,7]. This sarcopenic obesity (SO) phenotype accelerates obesity-related chronic diseases and aging, resulting in frailty, dependence, and increased mortality risk [8].

Notably, basic and clinical research shows that lean mass loss is closely linked to visceral adiposity [9,10]. Unlike body mass index (BMI), waist circumference (WC) and the waist-to-hip ratio (WHR), the waist-to-height ratio (WHtR) is less affected by sex or ethnicity, thus eliminating the need for sex-specific cut-offs and providing a more accurate measure of abdominal adipose tissue [11]. Emerging evidence suggests an inverse correlation between WHtR and grip strength [12,13], yet its predictive value for weight-standardized skeletal muscle mass (SMM/W) remains debated [14,15]. Few studies in Asian populations have systematically examined the association between WHtR (as an abdominal adiposity marker) and diagnostic parameters of SO (particularly SMM/W) [16].

As obesity management medications inevitably reduce muscle mass, individuals with obesity and inadequate muscle mass may present with a heterogeneous phenotype requiring clinical attention [17]. Therefore, this study aimed to evaluate the predictive value of the WHtR for SMM/W in women with obesity. Data for this retrospective study were extracted from the YiduCloud platform, a big-data intelligence system that integrates multi-source, heterogeneous electronic health record (EHR) data from our hospital [18].

Material and Methods

ETHICS APPROVAL:

Approval for the study was granted by the Ethics Committee of Tianjin Medical University Chu Hsien-I Memorial Hospital (Approval No.: ZXYJNYYkMEC2024-38; Date: November 29, 2024) in compliance with the Declaration of Helsinki. Written informed consent was obtained from all subjects prior to data collection, and all data were anonymized to protect patient confidentiality.

SAMPLE SIZE ESTIMATION:

For this cross-sectional study, the sample size was calculated based on the expected prevalence of SO using the formula for prevalence studies:

where Z=1.96 (95% confidence level), P=0.06 [16], and e=0.05 (margin of error). Accounting for a 20% potential data loss, a minimum sample size of 104 participants was calculated to ensure adequate power (1-β=0.8) for detecting the target prevalence with 5% margin of error (α=0.05).

INCLUSION CRITERIA:

We included women with obesity (aged 18 to 59 years and BMI ≥30 kg/m²) who first attended the obesity clinic at Tianjin Medical University Chu Hsien-I Memorial Hospital between 2023 and 2024.

EXCLUSION CRITERIA:

We excluded women with: (1) diabetes mellitus; (2) hepatorenal insufficiency; (3) hypothalamic pituitary endocrine axis dysfunction; (4) acute or chronic infection; (5) cancer; (6) current steroid/thyroid therapy; or (7) incomplete laboratory information, or missing body composition analysis data.

DATA COLLECTION:

The initial cohort was identified through the outpatient Hospital Information System (HIS) and comprised all individuals who attended the obesity clinic between 2023 and 2024. Medical record numbers and dates were recorded. Authorized staff accessed the YiduCloud platform (YiduCloud Technology Ltd., Beijing, China) using secure accounts to retrieve full medical records (since 2001) by patient ID, screening for individuals for whom this was their first-ever recorded encounter in the obesity clinic in the system. Longitudinal clinical, demographic, medical history, medication, and biochemical data were extracted through the platform’s export tool. The entire extraction procedure was independently duplicated by 2 researchers, and their outputs were cross-verified to ensure complete agreement. Body composition analysis data were exported separately from its dedicated system and integrated into the main dataset via double-data entry. Data meeting inclusion/exclusion criteria were statistically analyzed.

BODY COMPOSITION ANALYSIS:

Body composition was estimated using DBA-550 bioelectrical impedance analysis (BIA) (Donghuayuan, Beijing, China). Before measurement, participants were required to maintain a fasting state and thoroughly dry their skin. Subjects stood barefoot in an upright position on the instrument’s electrode platform, holding the hand electrodes with both hands and placing their thumbs on the electrode pads. The upper arms were relaxed in a natural position, slightly abducted (~30°) from the torso. The procedure for the measurements was carried out in compliance with the manufacturer’s instructions. Body components parameters (eg, body fat weight, skeletal muscle weight, percentage of fat) were obtained at the end of the analysis. SMM/W was calculated as SMM (kg) divided by body weight (kg).

ANTHROPOMETRIC MEASUREMENTS:

An ultrasonic height and weight measuring instrument was used to assess height and weight. DBA-550 BIA (Donghuayuan, Beijing, China) was used to measure WC and hip circumference. BMI was calculated as weight (kg) divided by height² (m²).

BIOCHEMICAL ANALYSIS:

After a minimum 8-hour overnight fast, fasting blood samples were collected to measure fasting plasma glucose (FPG), glycated hemoglobin A1c (HbA1c), fasting serum insulin (FINS), lipid profile, serum 25-hydroxyvitamin D3 (25-(OH)D3), hepatic/renal function parameters, and hormone concentrations. A 2-h plasma glucose blood sample was obtained after a 75-g oral glucose tolerance test. Fasting blood biochemical tests were performed at the biochemical testing center in the hospital. The plasma concentrations of metabolic markers were determined with an AU5800 automatic biochemical analyzer (Beckman Coulter, Inc., USA) and supporting reagents. 25-(OH)D3 concentrations were assessed via an enzyme-linked immunosorbent assay, whereas FINS was measured by means of a Cobas 602 automatic electrochemiluminescence immunoanalyzer (Roche, Switzerland) and supporting reagents. Based on FPG and FINS levels, homeostasis model assessment of insulin resistance (HOMA-IR) values were computed as FINS (μU/mL)×FPG (mmol/L)/22.5.

DISEASE THRESHOLD MEASUREMENT:

According to the European Society for Clinical Nutrition and Metabolism (ESPEN) and the European Association for the Study of Obesity (EASO) consensus, low SMM is defined as SMM/W ≤0.28 [4]. Groups with normal or lower SMM/W values were identified according to the above cut-off values. Obesity was defined as BMI ≥30 kg/m2.

STATISTICAL ANALYSIS:

The normality of continuous variables was assessed using the Kolmogorov-Smirnov test. Normally distributed data are expressed as the means±standard deviations (SDs), and non-normally distributed data are expressed as medians (interquartile ranges, [IQRs]). Continuous variables with normal distribution were analyzed using the t test, while non-normally distributed variables were analyzed with the Mann-Whitney U test. Non-normally distributed variables (eg, HOMA-IR) were natural log-transformed to meet the normality assumption before statistical analysis. Pearson correlation analysis was used to assess the relationships between variables. Multiple linear regression using both enter and stepwise methods was employed to examine associations between SMM/W and covariates (age, WHtR, WHR, triglyceride (TG), cholesterol (TC), ln HOMA-IR, FPG, HbA1c, uric acid (UA), and 25-(OH)D3 [19–21]), all of which had variance inflation factors (VIF) <5. Statistical analyses were performed with SPSS 27.0 (IBM Corp.), using a two-tailed significance level of P<0.05.

Results

PARTICIPANT CHARACTERISTICS:

A total of 300 women with obesity were enrolled (mean age 34±9 years). As shown in Table 1, participants with lower SMM/W values (≤0.28) demonstrated significantly higher BMI, fat mass (FM), percentage body fat (PBF), WC, WHR, and WHtR compared to those with normal SMM/W values (all P<0.05), while maintaining comparable metabolic profiles (all P>0.05).

CORRELATION ANALYSES:

The Pearson correlation coefficients are presented in Table 2. WHtR exhibits a stronger inverse association with SMM/W in obese women than traditional central obesity (WC, WHR) (r=−0.42, P<0.001) (Figure 1). Notably, ln HOMA-IR exhibited a negative correlation with SMM/W (r=−0.32, P<0.001) but positive correlation with WHtR (r=0.28, P=0.002) (Figure 2).

MULTIVARIATE REGRESSION ANALYSIS:

Stepwise regression identified WHtR as a significant independent predictor of SMM/W (β=−0.102, P=0.006), along with age (β=0.001, P<0.001), WHR (β=−0.375, P<0.001) and TG (β=0.005, P<0.001) (Table 3). The model explained 36.5% of the variance (R2=0.365, P<0.001). Results from enter-method regression are provided in Table 4.

Discussion

Our findings revealed a negative correlation between the WHtR and SMM/W in women with obesity, with a stronger inverse relationship than traditional central obesity markers such as WC and the WHR. After multivariable adjustment for clinically relevant confounders, WHtR showed an independent inverse association with SMM/W. The novel anthropometric index WHtR may be superior to traditional central adiposity markers (eg, WC) for screening SO, owing to its ability to simultaneously assess both abdominal fat accumulation and body proportionality. This aligns with National Health and Nutrition Examination Survey (NHANES) research showing WHtR had the strongest association with SO among multiple body shape indices (body roundness index, the z-score of the log-transformed, a body shape index) after full adjustment [16], while data from the Korean Frailty and Aging Cohort Study revealed that central obesity (measured by WC) was associated with a lower prevalence of sarcopenia in menopausal women but not in men [15]. Variations in participant characteristics and evaluation criteria may explain the outcome differences seen across studies.

Obesity independently contributes to muscle loss and dysfunction through metabolic disturbances induced by adipose tissue, including insulin resistance (IR), inflammation, and oxidative stress [22]. In our study, we found that ln HOMA-IR was significantly associated with both WHtR and SMM/W (Figure 2). Jablonowska-Lietz et al also reported substantial associations between WHtR and glucose levels, and between WHtR and HOMA-IR [11]. Utilizing large-scale national data from Korea and the United States, Moon et al demonstrated an inverse association between low appendicular skeletal muscle mass index (ASMI) and IR [23]. IR-induced compensatory hyperinsulinemia reduces protein synthesis, accelerates protein degradation, and increases myostatin expression, contributing to skeletal muscle atrophy [24]. Importantly, this sarcopenic transformation establishes a vicious cycle by reducing glucose disposal capacity, which further exacerbates systemic IR and precipitates functional decline in muscle strength [25].

This study has several limitations. First, as a single-center retrospective study, the generalizability of our findings may be limited; in particular, the results cannot be generalized to males. Second, while BIA was used instead of the criterion-standard dual-energy X-ray absorptiometry (DXA) for muscle mass assessment, this approach is clinically justified by its radiation-free nature, and demonstrated high concordance with DXA (r=0.921) for SMM quantification in previous validation studies [26]. Third, potential confounders affecting body composition (eg, physical activity, dietary intake, and medication use) were not systematically controlled for in our analysis.

Conclusions

The WHtR demonstrates a significant negative correlation with SMM/W and may serve as a practical screening indicator for SO in women with obesity. Incorporating WHtR into routine clinical assessments could enhance early risk stratification and guide personalized weight-management interventions. Future prospective studies utilizing DXA are warranted to validate its predictive value and establish optimal cut-off thresholds.