Study population
All the participants were recruited from the Health Management Center in a teaching hospital from January 1, 2013 to December 31, 2018. All adult (≥18 years old) individuals receiving routine health checkup at our health management center were eligible for the study. An initial recruitment resulted in an identification of 109,410 subjects. Metabolically healthy were defined as participants without history of metabolic diseases but with normal blood pressure, fasting blood glucose (FBG), glycated hemoglobin A1c (HbA1c) level, and lipid profile at baseline (26). To recruit participants with metabolically healthy, we performed a sequential recruitment as following. First, we excluded participants with history of a series of metabolic diseases and cancer (n = 9,651), and those with baseline metabolic abnormalities [high blood pressure (n = 37,134), impaired glucose regulation (n = 14,133), and dyslipidemia (n = 14,363)]. Then, we excluded participants who lost to follow up (n = 13,378) and with missing data (n = 8,083). Finally, we excluded participants with low baseline BMI (≤18.4 kg/m2, n = 2,674) (24), aged participants (≥65 years, n = 79), with CAP at baseline (n = 78), and with declined estimated glomerular filtration (eGFR < 60 ml/min/1.73 m2, n = 1). Included were 9,836 (4,085 males and 5,751 females) Chinese adults with mean age of 35.8±9.0 years (Figure e-1). Participants included in the study tended to be younger, have higher proportion of women and lower baseline BMI, level of HbA1c, FBG, and blood pressure, compared with those who were out of the study (Table e-1). The study protocol was approved by the Ethical Committee of Ren Ji Hospital, School of Medicine, Shanghai Jiao Tong University. As a de-identified secondary data analysis, patients’ consent was waived.
Blood pressure (systolic blood pressure and diastolic blood pressure), FBG, HbA1c, lipid profile [total cholesterol (TC), total triglyceride (TG), low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C)], and carotid artery B ultrasonography were annually assessed throughout the study (2013–2018).
Body weight (to the nearest 0.5 kg) and height (to the nearest 0.5 cm) was measured in standing position without shoes and in light clothing, using an electronic scale (SK-CK, Shuang Jia Company, Shanghai, China). BMI was calculated by body weight (kg) divided by height square (m2). Blood pressure was measured twice using an automatic blood-pressure meter (HBP-9020, OMRON (China) Co., Ltd.) after participants were seated for at least 10 min. The average of two measurements was recorded for further analysis.
Venous blood samples were drawn and transfused into vacuum tubes containing EDTA in the morning after participants were fasted for at least six hours. An automatic analyzer (Roche 701 Bioanalyzer, Roche, UK) was used to measure FBG with the hexokinase/glucose-6-phosphate dehydrogenase method. The level of HbA1c was measured by high performance liquid chromatography, using the fully automated VARIANT™ II Hemoglobin Testing System (Bio-Rad, U.S). TC, TG, LDL-C, and HDL-C were measured by an automatic biochemical analyzer (Roche 701 Bioanalyzer, Roche, UK). The eGFR was calculated using the Chronic Kidney Disease Epidemiology Collaboration 2-level race Eq. (27). The concentration of high sensitivity CRP (hs-CRP) was measured by immune-tubidimetric method. All the measurements were completed in the Clinical Laboratory of our hospital.
The history of hypertension, diabetes/impaired glucose regulation, dyslipidemia, cardiovascular diseases (stroke, hemorrhage, coronary artery bypass grafting, stent surgery, and ischemic infarction), was collected via a self-report questionnaire.
The transition from metabolically healthy status to metabolic abnormality (exposure)
The transition was deemed if any of the following abnormalities was confirmed during five-year follow up: high blood pressure (systolic blood pressure≥130 mmHg or diastolic blood pressure≥80 mmHg) (28); impaired glucose regulation (FBG ≥5.6 mmol/L or HbA1c≥5.7%) (29); high TC (TC≥5.72 mmol/L); high TG (TG≥1.7 mmol/L); high LDL-C (LDL-C≥3.4 mmol/L); low HDL-C (HDL-C < 0.9 mmol/L in men and < 1.0 mmol/L in women). Dyslipidemia was defined as any of the four lipid parameters (TC, TG, LDL-C, and HDL-C) was confirmed abnormal based on above-mentioned references. In the secondary analysis, we evaluated the associated between baseline body weight status (metabolically healthy normal weight vs. metabolically healthy overweight) and the risk of transition to metabolic abnormalities. Participants were classified into normal weight (18.5≤BMI < 24.0 kg/m2) or overweight (BMI≥24.0 kg/m2) groups based on the criteria for Chinese adults (30).
Ultrasound B-mode imaging was performed annually to detect CAP during five-year follow-up (Philips HDI 5000 ultrasound system equipped with a 7.5 MHz probe). Intima-media thickness was measured at the point approximately 1.5 cm away from the distal part of the bifurcation of common carotid artery. CAP is defined as a focal region with a thickness > 1.5 mm as measured from the media adventitia interface to the lumen-intima interface or as the presence of focal wall thickening that is at least 50% greater than that of the surrounding vessel wall (31).
Statistical analysis
Data were presented as mean±standard deviation if it was in normal distribution and medium and quartile range if it was in abnormal distribution. We completed all statistical analyses by SAS version 9.4 (SAS Institute, Inc, Cary, NC). Formal hypothesis testing will be two-sided with a significant level of 0.05. Because the conversion was confirmed at least twice, we determined the person-time of follow-up for each participant from January 1, 2014 to either the first onset date of the conversion, or the end of follow-up (December 31, 2018), whichever came first.
We used the proportional Cox regression model to evaluate the association between the transition and future risk of CAP. We adjusted the potential confounders in different models: model 1, adjusting for age (y) and sex; and model 2, further adjusting for baseline systolic blood pressure (mmHg), diastolic blood pressure (mmHg), FBG (mmol/L), HbA1c (%), TC(mmol/L), TG (mmol/L), LDL-C (mmol/L), HDL-C (mmol/L), eGFR (ml/min/1.73 m2) and hs-CRP (mg/L).
Likelihood ratio tests were conducted to examine statistical interactions between the transition and sex, and age, in relation to the CAP by comparing − 2 log likelihood χ2 between nested models with and without the cross-product terms.
To test the robustness of the main results, we conducted two sensitivity analyses in model 2. First, we censored participants whose baseline level of hs-CRP was 10.0 mg/L or more (32). Then, we censored participants who was confirmed with metabolic abnormalities once to lower the possibility of misclassification (26).