This cross-sectional study involved in individuals having routine health examinations from May 2016 to August 2017 at the First Affiliated Hospital of Wenzhou Medical University. Subjects with age≥18, having serum lipid profile and bioelectrical impedance analysis (BIA) measures were involved in our study. Exclusion criteria included individuals with aged<18 years; taking lipid-lowering medicine; with a history of stroke, malignant tumor, chronic kidney or liver disease, or thyroid disease.
The research was agreed by the Institutional Review Board (IRB) of the hospital. Given its cross-sectional nature, the consent was waived by the IRB and the medical privacy was protected. All data were reviewed complying with the Declaration of Helsinki.
Data regarding the health examinations were collected, including questionnaires concerning to lifestyles and pre-existing conditions, and results of BIA, blood, biochemical and anthropometric measurements.
Living habits contained smoking and drinking. Alcohol drinking higher than 70 g per week for women and higher than 140 g per week for men were regarded as heavy drinking. The 3 kinds of smoking status were defined as follows: current smoker (currently smokes and has smoked for at least the past 6 months or who abandoned smoking <2 years ago), past smoker (smoked in the past and had quitted smoking for at least 2 years) and never (never smoked).
Pre-existing conditions included diabetes mellitus (DM), hypertension (HTN), and hyperuricemia. DM was referred to random plasma glucose(PG), fasting plasma glucose (FPG) or 2-h PG equal or greater than 200 mg/dL, 126 mg/dL, or 200 mg/dL, respectively (18). Every individual took a blood pressure measure at the morning of medical examination, containing systolic blood pressure (SBP) and diastolic blood pressure (DBP). HTN was referred to SBP ≥140 mmHg or DBP ≥90 mmHg (19). Hyperuricemia was referred to serum uric acid (UA) levels higher than 6 mg/dL in females and higher than 7.0 mg/dL in males (20). Additionally, past morbidities included medical histories of DM, HTN or hyperuricemia or taking corresponding medications from self-reports.
To monitor appendicular skeletal muscle mass (ASM; kg), every subject carried out BIA detection (InBody770; InBody Japan Inc., Tokyo, Japan). Then, skeletal muscle mass index (SMI; kg/m2) was computed through following formula: ASM(kg)/height2 (m2). Additionally, according to the AWGS 2019 Consensus(1), males with SMI lower than 7.0kg/m2 and females with SMI lower than 5.7kg/m2 were diagnosed as sarcopenia. Subjects with Body mass index (BMI) higher than 25 kg/m2 were regarded as with status of overweight.
The following parameters were measured at the morning of medical examination: TG, HDL-C, low density lipoprotein cholesterol (LDL-C), total cholesterol (TC), glycated hemoglobin (HbA1c), FPG, albumin, UA, hemoglobin (Hb), leukocyte count (WBC) and platelet (PLT).
Continuous and categorical factors were displayed as medians (ranges) and frequencies (percentages), respectively. Through receiver operating characteristic (ROC) curves, the diagnostic accuracies of lipid profile for sarcopenia were assessed. The optimal indicator was selected, based on area under curve (AUC) and Youden index. The subjects were grouped according to the selected indicator(high and low group). The differences of the continuous factors were compared with the Mann-Whitney test, while the differences of the categorical factors was compared with the χ2 test. Through univariate and multivariate logistic regression models, the effects of the factors on the risk of sarcopenia were evaluated. Backward stepwise selection with the Akaike information criterion(AIC) was performed to identify variables for the multivariable logistic regression models. Interaction term in the multivariable logistic regression analysis was also conducted to get rid of confound factor. In addition, individuals were stratified by quartiles, χ2 test was utilized to compare the different rate of sarcopenia.
R version 3.6.1 was utilized to perform statistical analyses (https://www.r-project.org/). All analyses were 2-sided, and a P value <0.05 was regarded as significance level.