Hostname: page-component-78c5997874-dh8gc Total loading time: 0 Render date: 2024-11-07T01:25:56.189Z Has data issue: false hasContentIssue false

All-cause mortality risk with different metabolic abdominal obesity phenotypes: the Rural Chinese Cohort Study

Published online by Cambridge University Press:  16 March 2023

Xiaoyan Wu
Affiliation:
Department of Cardio-Cerebrovascular Disease and Diabetes Prevention and Control, Shenzhen Center for Chronic Disease Control, Shenzhen, Guangdong, People’s Republic of China Department of General Practice, The Affiliated Luohu Hospital of Shenzhen University Medical School, Shenzhen, Guangdong, People’s Republic of China
Yang Zhao
Affiliation:
Department of General Practice, The Affiliated Luohu Hospital of Shenzhen University Medical School, Shenzhen, Guangdong, People’s Republic of China
Qionggui Zhou
Affiliation:
Department of General Practice, The Affiliated Luohu Hospital of Shenzhen University Medical School, Shenzhen, Guangdong, People’s Republic of China
Minghui Han
Affiliation:
Department of Epidemiology and Health Biostatistics, College of Public Health, Zhengzhou University, Zhengzhou, Henan, People’s Republic of China
Ranran Qie
Affiliation:
Department of Epidemiology and Health Biostatistics, College of Public Health, Zhengzhou University, Zhengzhou, Henan, People’s Republic of China
Pei Qin
Affiliation:
Department of Medical Record Management, Shenzhen Qianhai Shekou Free Trade Zone Hospital, Shenzhen, Guangdong, People’s Republic of China
Yanyan Zhang
Affiliation:
Department of General Practice, The Affiliated Luohu Hospital of Shenzhen University Medical School, Shenzhen, Guangdong, People’s Republic of China
Zelin Huang
Affiliation:
Department of Biostatistics and Epidemiology, Shenzhen University Medical School, Shenzhen, Guangdong, People’s Republic of China
Jiong Liu
Affiliation:
Department of Biostatistics and Epidemiology, Shenzhen University Medical School, Shenzhen, Guangdong, People’s Republic of China
Fulan Hu
Affiliation:
Department of Biostatistics and Epidemiology, Shenzhen University Medical School, Shenzhen, Guangdong, People’s Republic of China
Xinping Luo
Affiliation:
Department of Biostatistics and Epidemiology, Shenzhen University Medical School, Shenzhen, Guangdong, People’s Republic of China
Ming Zhang
Affiliation:
Department of Biostatistics and Epidemiology, Shenzhen University Medical School, Shenzhen, Guangdong, People’s Republic of China
Yu Liu
Affiliation:
Department of General Practice, The Affiliated Luohu Hospital of Shenzhen University Medical School, Shenzhen, Guangdong, People’s Republic of China
Xizhuo Sun*
Affiliation:
Department of General Practice, The Affiliated Luohu Hospital of Shenzhen University Medical School, Shenzhen, Guangdong, People’s Republic of China
Dongsheng Hu*
Affiliation:
Department of General Practice, The Affiliated Luohu Hospital of Shenzhen University Medical School, Shenzhen, Guangdong, People’s Republic of China
*
*Corresponding authors: Xizhuo Sun, email sunxz632@126.com; Dongsheng Hu, email dongshenghu563@126.com
*Corresponding authors: Xizhuo Sun, email sunxz632@126.com; Dongsheng Hu, email dongshenghu563@126.com
Rights & Permissions [Opens in a new window]

Abstract

We aimed to investigate the association of metabolic obesity phenotypes with all-cause mortality risk in a rural Chinese population. This prospective cohort study enrolled 15 704 Chinese adults (38·86 % men) with a median age of 51·00 (interquartile range: 41·00–60·00) at baseline (2007–2008) and followed up during 2013–2014. Obesity was defined by waist circumference (WC: ≥ 90 cm for men and ≥ 80 cm for women) or waist-to-height ratio (WHtR: ≥ 0·5). The hazard ratio (HR) and 95 % CI for the risk of all-cause mortality related to metabolic obesity phenotypes were calculated using the Cox hazards regression model. During a median follow-up of 6·01 years, 864 deaths were identified. When obesity was defined by WC, the prevalence of participants with metabolically healthy non-obesity (MHNO), metabolically healthy obesity (MHO), metabolically unhealthy non-obesity (MUNO) and metabolically unhealthy obesity (MUO) at baseline was 12·12 %, 2·80 %, 41·93 % and 43·15 %, respectively. After adjusting for age, sex, alcohol drinking, smoking, physical activity and education, the risk of all-cause mortality was higher with both MUNO (HR = 1·20, 95 % CI 1·14, 1·26) and MUO (HR = 1·20, 95 % CI 1·13, 1·27) v. MHNO, but the risk was not statistically significant with MHO (HR = 0·99, 95 % CI 0·89, 1·10). This result remained consistent when stratified by sex. Defining obesity by WHtR gave similar results. MHO does not suggest a greater risk of all-cause mortality compared to MHNO, but participants with metabolic abnormality, with or without obesity, have a higher risk of all-cause mortality. These results should be cautiously interpreted as the representation of MHO is small.

Type
Research Article
Copyright
© The Author(s), 2023. Published by Cambridge University Press on behalf of The Nutrition Society

Over the past few decades, the prevalence of obesity has continued to increase such that it has become a serious public health issue worldwide(1). Obesity is associated with death(Reference Aune, Sen and Prasad2) and various chronic conditions such as hypertension(Reference Seravalle and Grassi3), cancer(Reference Deng, Lyon and Bergin4) and CVD(Reference Ortega, Lavie and Blair5). There is, however, heterogeneity among people with obesity which can be divided into two phenotypes: metabolically healthy and metabolically unhealthy(Reference Phillips6Reference Karelis, Faraj and Bastard8). People who were obese with favourable blood pressure, lipid profile, inflammation levels and insulin sensitivity are considered to have metabolically healthy obesity (MHO)(Reference Phillips6Reference Karelis, Faraj and Bastard8). Other metabolic obesity phenotypes include metabolically healthy non-obesity (MHNO), metabolically unhealthy non-obesity (MUNO) and metabolically unhealthy obesity (MUO)(Reference Zhao, Qin and Sun9Reference Doustmohamadian, Serahati and Barzin11). In population-based research, the association between different metabolic obesity phenotypes and death has received increasing attention. The association between these metabolic obesity phenotypes and the risk of mortality is inconsistent(Reference van der, Nooyens and van Duijnhoven10Reference Bo, Musso and Gambino16). Moreover, so far as we know, only two research examined the relation in the Chinese population(Reference Zhang, Dong and Wang17,Reference Tian, Wang and Zuo18) . These, however, were based on data from physical examination and hospital visit populations in older men, suggesting some bias and limited generalisability to the general population. Data on the relationship of metabolic obesity phenotypes with mortality in rural natural China in areas of relatively low-socioeconomic status are still lacking.

Most of the current studies linking mortality to metabolic obesity phenotypes were based on Western populations and used BMI to define obesity(Reference Al-Khalidi, Kimball and Kuk14,Reference Kuk, Rotondi and Sui15,Reference Guo and Garvey19) . Previous studies have shown that Asians are more inclined to abdominal obesity than Western populations(Reference Nazare, Smith and Borel20), and that increased waist circumference (WC) or waist-to-height ratio (WHtR) are better indicators of all-cause mortality risk independent of BMI(Reference Leitzmann, Moore and Koster21Reference Ashwell, Mayhew and Richardson23). Nevertheless, no study has investigated the association between metabolic obesity phenotypes and death with abdominal obesity as the focus rather than general obesity, which is defined by BMI in the rural Chinese population.

This study therefore prospectively explored the relationship of different metabolic obesity phenotypes with all-cause mortality risk by using WC and WHtR to define obesity on the basis of the Rural Chinese Cohort Study.

Materials and methods

Study participants

The Rural Chinese Cohort Study recruited 20 194 Chinese adults aged over 18 residing in a rural area in the middle of China from July to August 2007 and July to August 2008 at baseline examination(Reference Zhou, Wu and Zhang24). Two towns, Tiemen and Cijian in Xin’an County, were selected as representatives of the area’s geographical and rural economic status. The study participants were randomly recruited by a cluster sampling procedure, with villages as the sampling unit from the two towns. Details of the eligibility requirements for study participants have been previously described(Reference Zhao, Zhang and Luo25). The first follow-up survey was conducted from July to August 2013 and July to October 2014, with 17 265 individuals successfully followed up (response rate 85·5 %). For the current study, we excluded participants who had missing data for defining metabolic status (n 84), those with missing data for defining obesity (height or WC) (n 6), those who were underweight (BMI < 18·5 kg/m2) (n 562) and those with CVD and/or cancer (n 909) at baseline. Ultimately, a total of 15 704 participants were included in the final analysis (Fig. 1). This study was conducted according to the guidelines laid down in the Declaration of Helsinki, and all procedures involving human subjects were approved by the Medical Ethics Committee of Shenzhen University. All the participants gave written informed consent.

Fig. 1. Flow chart of the selection of participants.

Data collection

We conducted face-to-face interviews, physical examinations and blood sample collection using the same procedures during the baseline and follow-up surveys. Detailed information on demographic characteristics and lifestyles was collected by interview with standardised questionnaires. Education level was dichotomised as high school or above and low education level. Smoking was defined as currently smoking and/or having smoked at least 100 cigarettes in a lifetime(Reference Bondy, Victor and Diemert26). Alcohol drinking was defined as having consumed alcohol twelve or more times during the last year(Reference Han, Liu and Sun27). Physical activity level was classified as low or moderate/high physical activity level according to the International Physical Activity Questionnaire(Reference Craig, Marshall and Sjöström28). With participants wearing light clothing, body weight was measured to the nearest 0·5 kg on a vertical weight scale. Height was measured to the nearest 0·1 cm with participants standing erect in bare feet. With participants gently breathing, WC was measured at the mid-point between the lowest rib and the iliac crest to the nearest 0·1 cm. WC, height and body weight were measured twice according to standard methods(29), with the average used in the analysis. WHtR was calculated as WC (metres)/height (metres). BMI was calculated as weight (kilograms) divided by the square of height (metres). In accordance with the standardised protocol of the American Heart Association, blood pressure was assessed three times on the right arm at 30-s intervals using an electronic sphygmomanometer (HEM-770A Fuzzy), with the mean of the three measurements used in the analysis. Fasting blood samples for biochemical analysis were collected after an overnight fast of at least 8 h. TAG, HDL-cholesterol and fasting plasma glucose were measured using a HITACHI automatic clinical analyzer (Model 7060, Tokyo). Detailed information about storage and measurement methods has been previously described(Reference Zhao, Zhang and Luo25). The same measurements as for the baseline examination were taken during the follow-up examination.

Definition of metabolic obesity phenotypes

Metabolically healthy individuals were defined as having zero metabolic risk factors among the following harmonised criteria by the Joint Interim Statement: (1) systolic blood pressure ≥ 130 mmHg or diastolic blood pressure ≥ 85 mmHg or anti-hypertensive drug treatment; (2) TAG level ≥ 1·7 mmol/l or drug treatment; (3) HDL-cholesterol level < 40 mg/dl (1·034 mmol/l) for men or < 50 mg/dl (1·293 mmol/l) for women or drug treatment and (4) fasting plasma glucose level ≥ 5·6 mmol/l or drug treatment(Reference Alberti, Eckel and Grundy30,Reference Scherer and Hill31) . Participants with one or more of the four metabolic risk factors were defined as metabolically unhealthy. Obesity was defined by WC (≥ 90 cm for men and ≥ 80 cm for women(Reference Alberti, Zimmet and Shaw32)) or WHtR (≥ 0·5(Reference Hsieh and Muto33)). Participants were divided into four metabolic obesity phenotypes: MHNO, MHO, MUNO and MUO.

Follow-up of mortality

Death information was collected through face-to-face interviews with participants’ family members, the village doctor or other health care providers during the follow-up survey. The information on death was further checked with the local Centers for Disease Control and Prevention. For conflicting data, we verified the information with relatives or local village doctors(Reference Liu, Chen and Liu34).

Statistical analyses

For baseline characteristics, continuous variables with skewed distribution are presented as median (interquartile range) and were compared using the Kruskal–Wallis test. Categorical variables are presented as number (percentage), with chi-square test used for comparison. The proportional hazard assumption was met and tested by the Kaplan–Meier Curve and Schoenfeld residuals. Cox proportional-hazards regression model was thus used to calculate the hazard ratio and 95 % CI for the risk of all-cause mortality associated with different metabolic obesity phenotypes. We chose MHNO as the reference group and adjusted for several potential confounders, including sex, age, alcohol drinking, smoking, physical activity level and education, in the final analyses. To examine the potential effects of known confounding factors, we conducted subgroup analyses stratified by sex (men or women) and age (< 60 or ≥ 60 years). To assess the robustness of the results, we performed sensitivity analyses that involved excluding participants with diabetes, those who were smokers at baseline, and those who died within the first year.

All statistical analyses were conducted with SAS v9.4 (SAS Inst.). Statistical significance was established as two-sided P < 0·05.

Results

Baseline characteristics

A total of 15 704 participants were eligible for inclusion, with a median age of 51·00 (interquartile range: 41·00–60·00). When obesity was defined by WC, the prevalence of participants with MHNO, MHO, MUNO and MUO at baseline was 12·12 %, 2·80 %, 41·93 % and 43·15 %, respectively (Table 1). The baseline characteristics including age, sex, education, smoking, alcohol drinking, physical activity, BMI, WC, WHtR, systolic blood pressure, diastolic blood pressure, fasting plasma glucose, TAG and HDL-cholesterol levels significantly differed by metabolism and obesity status as defined by WC (all P < 0·05) (Table 1). Metabolically unhealthy individuals are older compared to metabolically healthy individuals, with a median age of 45·00 (39·00–53·00) and 53·00 (44·00–60·00) for individuals with MHO and MUO, respectively. Individuals who are obese, especially MUO, are more likely to be women, have lower levels of education, be non-smokers, non-drinkers and be physically inactive. These individuals also have higher levels of BMI, WC, WHtR, systolic blood pressure, diastolic blood pressure, fasting plasma glucose and TG, but lower levels of HDL-cholesterol. When obesity was defined by WHtR, the prevalence was 10·23 %, 4·68 %, 28·24 % and 56·85 %, respectively. Individuals with different metabolic obesity phenotypes have similar characteristics to those defined as obese by WHtR.

Table 1. Baseline characteristics of study participants by metabolic obesity phenotypes

Abbreviations: DBP, diastolic blood pressure; FPG, fasting plasma glucose; HDL-C, HDL-cholesterol; IQR, interquartile range; MHNO, metabolically healthy non-obesity; MHO, metabolically healthy obesity; MUNO, metabolically unhealthy non-obesity; MUO, metabolically unhealthy obesity; SBP, systolic blood pressure; WC, waist circumference; WHtR, waist-to-height ratio.

Data are median (interquartile range) or number (percentage). The P value was generated by Kruskal–Wallis H test or chi-square test for continuous variables and categorical variables.

Metabolic obesity phenotypes at baseline and risk of all-cause mortality

During the follow-up of 92 805·61 person-years (average follow-up of 6·01 years), we identified 864 deaths (all-cause mortality 9·31/1000 person-years).

When obesity was defined by WC, the all-cause mortality was 5·40, 4·48, 11·38 and 8·77/1000 person-years with MHNO, MHO, MUNO and MUO, respectively (Table 2). After adjusting for age, sex, alcohol drinking, smoking, physical activity level and education, the risk of all-cause mortality was higher with MUNO (adjusted hazard ratio (aHR) = 1·20, 95 % CI 1·14, 1·26) and MUO (aHR = 1·20, 95 % CI 1·13, 1·27) v. MHNO (Table 2), but the association was not statistically significant for MHO (aHR = 0·99, 95 % CI 0·89, 1·10) (Table 2).

Table 2. Association of metabolic obesity phenotypes at baseline with risk of all-cause mortality

Abbreviations: HR, hazard ratio; MHNO, metabolically healthy non-obesity; MHO, metabolically healthy obesity; MUNO, metabolically unhealthy non-obesity; MUO, metabolically unhealthy obesity; WC, waist circumference; WHtR, waist-to-height ratio.

* Per 1000 person-years.

Unadjusted model.

Adjusted for sex, age, alcohol drinking, smoking, physical activity level and education.

When obesity was defined by WHtR, all-cause mortality was 4·45, 6·98, 10·85 and 9·66/1000 person-years with MHNO, MHO, MUNO and MUO, respectively (Table 2). After adjusting for age, sex, smoking, physical activity level and education confounding factors, the risk of all-cause mortality was higher with both MUNO (aHR = 1·18, 95 % CI 1·12, 1·26) and MUO (aHR = 1·21, 95 % CI 1·14, 1·28) v. MHNO, but the association was not statistically significant for MHO (aHR = 1·00, 95 % CI 0·91, 1·09) (Table 2).

The results of the sensitivity analyses were all similar to the main analysis (online Supplementary Table 2). When obesity was defined by WC, the aHR (95 % CI) for all-cause mortality with MHO, MUNO, and MUO v. MHNO was 0·99 (0·89, 1·10), 1·20 (1·14, 1·26) and 1·20 (1·13, 1·27) after excluding participants who died within 1 year; the aHR (95 % CI) was 0·99 (0·89, 1·10), 1·19 (1·13, 1·25) and 1·18 (1·11, 1·24) after excluding participants with diabetes at baseline; the aHR (95 % CI) was 0·96 (0·85, 1·09), 1·18 (1·10, 1·26) and 1·16 (1·09, 1·24) after excluding participants who smoke. When obesity was defined by WHtR, the results were equally robust (online Supplementary Table 2).

Subgroup analyses

All subgroup analyses stratified by sex and age gave similar results for MHO, MUNO and MUO with obesity defined using WC or WHtR. In comparison with MHNO, MHO was not associated with the risk of all-cause mortality, while the risk with MUNO and MUO was higher by sex and age groups. Detailed results of subgroup analyses are shown in Fig. 2 and 3.

Fig. 2. Association of metabolic obesity phenotypes (by WC) at baseline with risk of all-cause mortality by sex and age. Abbreviations: HR, hazard ratio; MHNO, metabolically healthy non-obesity; MHO, metabolically healthy obesity; MUNO, metabolically unhealthy non-obesity; MUO, metabolically unhealthy obesity; WC, waist circumference. aPer 1000 person-years. bUnadjusted model. cAdjusted for sex, age, alcohol drinking, smoking, physical activity level and education. Each group adjusted for the other covariates except for itself.

Fig. 3. Association of metabolic obesity phenotypes (by WHtR) at baseline with risk of all-cause mortality by sex and age. Abbreviations: HR, hazard ratio; MHNO, metabolically healthy non-obesity; MHO, metabolically healthy obesity; MUNO, metabolically unhealthy non-obesity; MUO, metabolically unhealthy obesity; WHtR, waist-to-height ratio. aPer 1000 person-years. bUnadjusted model. cAdjusted for sex, age, alcohol drinking, smoking, physical activity level and education. Each group adjusted for the other covariates except for itself.

Discussion

In this large prospective cohort study, we included 15 704 adults with a median follow-up of 6·01 years to explore the association between different metabolic obesity phenotypes and the risk of all-cause mortality. Regardless of whether WC or WHtR was used to define obesity, after adjusting for potential confounding factors, the risk of all-cause mortality was higher with MUNO and MUO v. MHNO, with no significant association found for MHO. The results persisted in subgroup and sensitivity analyses.

The association between different metabolic obesity phenotypes and the risk of all-cause mortality remains controversial. Consistent with our results, some studies found that both MUNO and MUO were positively associated with the risk of all-cause mortality(Reference Doustmohamadian, Serahati and Barzin11,Reference Al-Khalidi, Kimball and Kuk14,Reference Kuk, Rotondi and Sui15) , with no association found for MHO(Reference Doustmohamadian, Serahati and Barzin11,Reference Al-Khalidi, Kimball and Kuk14,Reference Bo, Musso and Gambino16,Reference Zhang, Dong and Wang17,Reference Lee, Jeong and Kim35) compared with MHNO. One systematic review and meta-analysis(Reference Zheng, Zhou and Zhu36) that included eleven prospective studies (2705 deaths and 118 471 participants) did not find a positive association of MHO with all-cause mortality risk; however, other studies have questioned the benign health status of MHO(Reference van der, Nooyens and van Duijnhoven10,Reference Loprinzi and Frith12,Reference Hinnouho, Czernichow and Dugravot13) . A prospective cohort study that included 22 654 participants with an average follow-up time of 13·4 years found that(Reference van der, Nooyens and van Duijnhoven10), compared to MHNO, MHO defined by WC was associated with a higher risk of all-cause mortality, while another cohort study of 1758 individuals followed up for 30 years and with 788 deaths showed that MHO could increase the risk of all-cause mortality(Reference Arnlöv, Ingelsson and Sundström37).

Follow-up duration may be one of the factors explaining the inconsistent results. Kramer et al. included eight studies systematically evaluating the association of MHO and all-cause mortality or risk of cardiovascular events. The results suggested that MHO represented a similar risk to that shown in our results (hazard ratio = 1·19, 95 % CI 0·98, 1·38), but when the review included only four studies with a follow-up of > 10 years, MHO increased the risk (hazard ratio = 1·24, 95 % CI 1·02, 1·55)(Reference Kramer, Zinman and Retnakaran38). This finding may suggest that a longer follow-up is warranted to identify any increased risk associated with MHO(Reference Kramer, Zinman and Retnakaran38). Reis et al. deeply explored the association between obesity duration and coronary artery calcification, finding that the risk was significant among participants with > 10 years’ abdominal obesity defined by WC and > 20 years’ general obesity defined by BMI(Reference Reis, Loria and Lewis39). Bell et al. studied the natural course of MHO over 20 years, finding that after a 5-year follow-up, 31·8 % of MHO individuals changed to metabolically unhealthy and after a 20-year follow-up, 51·5 % of MHO individuals changed to metabolically unhealthy(Reference Bell, Hamer and Sabia40). This finding may also explain the importance of follow-up duration in the association between MHO and risk of all-cause mortality. Additionally, the inconsistent definition of MHO in different studies may lead to discrepant findings(Reference Doustmohamadian, Serahati and Barzin11,Reference Al-Khalidi, Kimball and Kuk14,Reference Zhang, Dong and Wang17,Reference Lee, Jeong and Kim35,Reference Caleyachetty, Thomas and Toulis41) . Some studies defined metabolic health by including one or two risk factors(Reference Doustmohamadian, Serahati and Barzin11,Reference Lee, Jeong and Kim35,Reference Cheng, Gao and Mitchell42) , while in the present study, we adopted a stricter definition (none of the metabolic abnormality indicators is defined as metabolic healthy), which can reduce the impact of metabolic abnormality factors on the outcome. However, using a strict definition resulted in a smaller sample size of metabolic health. Moreover, by using WC to define obesity, only 2·80 % of participants were classified as MHO, with relatively fewer deaths among them, resulting in a wide CI for risk estimates. Future research should therefore use a unified standard to define metabolic healthy when comparing the risk among different studies and populations.

Our study indicates that special attention should be paid to individuals with MUNO. Consistent with other studies(Reference Doustmohamadian, Serahati and Barzin11,Reference Al-Khalidi, Kimball and Kuk14,Reference Kuk, Rotondi and Sui15) , this group, similar to MUO, could be at increased risk of all-cause mortality. Itʼs mortality rate is higher than that of the MUO group in our study. It may represent the most severe subtype in the phenotype spectrum(Reference Kramer, Zinman and Retnakaran38). Because people in the MUNO group are not obese, this population is easily overlooked by the usual preventive healthcare strategies. Regardless of obesity, metabolic abnormalities could increase the risk of all-cause mortality. Compared with obesity, therefore, metabolic abnormalities may be more strongly associated with all-cause mortality risk, suggesting that people should maintain a metabolically healthy status. Regular evaluation of metabolic levels of blood glucose, blood lipid and blood pressure for people with obesity is essential for preventing all-cause mortality.

Our study has several strengths. To our knowledge, it is the first to use abdominal obesity (WC and WHtR) to explore the association of metabolic obesity phenotypes with the risk of all-cause mortality in a rural Chinese adult population. In addition, we adjusted for confounding factors, including demographic characteristics and behavioural factors, in the statistical model to test whether the metabolic obesity phenotypes were independently related to the risk of all-cause mortality. We also conducted subgroup and sensitivity analyses to test the robustness of the current findings. Nevertheless, our study had several limitations. First, there may still be some unmeasured confounding factors, such as anxiety, depression or stress, that are associated with mortality(Reference Russ, Stamatakis and Hamer43,Reference Batty, Russ and Stamatakis44) . Second, using a strict definition resulted in a smaller sample size for metabolic healthy, especially for the MHO, with relatively fewer deaths, resulting in a wide CI for risk estimates. In addition, we have had only one follow-up result so far; hence, we could not assess the association between dynamic changes in metabolic obesity phenotypes and the risk of all-cause mortality. More research in this area is needed in the future. Finally, the participants in our study were from a rural Chinese population which may not be a representative sample of a multi-ethnic, multi-centre cohort of Chinese adults.

Conclusions

Compared with MHNO, MUNO and MUO were positively associated with the risk of all-cause mortality at 6·01 years of follow-up among rural Chinese people, while MHO did not relate to the risk. The short follow-up period and small sample size for the healthy metabolic group, especially for the MHO, may indicate the need to interpret results with caution. Larger studies with longer follow-up periods are therefore needed to provide more information in this field. Our findings indicate that people with MUNO should also be included in routine preventive care. Additionally, the combined assessment of both obesity and metabolic status should be considered to predict the risk of all-cause mortality.

Acknowledgements

This study was supported by the National Natural Science Foundation of China (grant numbers 81402752 and 81673260); the Natural Science Foundation of Guangdong Province (grant no. 2019A1515011183) and the Science and Technology Development Foundation of Shenzhen (grant nos. JCYJ20170412110537191 and JCYJ20190808145805515).

The investigators thank the organisations that funded the research, dedicated participants and all research staff of the study.

X. W., Y. Z., and D. H. substantially contributed to the design and drafting of the study and the analysis and interpretation of the data. X. W. and Y. Z. wrote the manuscript. Q. Z., M. H., R. Q., P. Q., Y. Z., Z. H., J. L., F. H., X. L., M. Z., Y. L., X. S., and D. H. revised it critically for important intellectual content. All authors were involved in the collection of data and approved the final manuscript.

The authors declare that they have no competing interests.

Supplementary material

For supplementary material/s referred to in this article, please visit https://doi.org/10.1017/S0007114523000673

Footnotes

These authors contributed equally to this work

References

Collaboration NCDRF (2017) Worldwide trends in body-mass index, underweight, overweight, and obesity from 1975 to 2016: a pooled analysis of 2416 population-based measurement studies in 128·9 million children, adolescents, and adults. Lancet 390, 26272642.CrossRefGoogle Scholar
Aune, D, Sen, A, Prasad, M, et al. (2016) BMI and all cause mortality: systematic review and non-linear dose-response meta-analysis of 230 cohort studies with 3·74 million deaths among 30.3 million participants. BMJ 353, i2156.CrossRefGoogle ScholarPubMed
Seravalle, G & Grassi, G (2017) Obesity and hypertension. Pharmacol Res 122, 17.CrossRefGoogle ScholarPubMed
Deng, T, Lyon, CJ, Bergin, S, et al. (2016) Obesity, inflammation, and cancer. Annu Rev Pathol 11, 421449.CrossRefGoogle ScholarPubMed
Ortega, FB, Lavie, CJ & Blair, SN (2016) Obesity and cardiovascular disease. Circ Res 118, 17521770.CrossRefGoogle ScholarPubMed
Phillips, CM (2013) Metabolically healthy obesity: definitions, determinants and clinical implications. Rev Endocr Metab Disord 14, 219227.CrossRefGoogle ScholarPubMed
Primeau, V, Coderre, L, Karelis, AD, et al. (2011) Characterizing the profile of obese patients who are metabolically healthy. Int J Obes 35, 971981.CrossRefGoogle ScholarPubMed
Karelis, AD, Faraj, M, Bastard, JP, et al. (2005) The metabolically healthy but obese individual presents a favorable inflammation profile. J Clin Endocrinol Metab 90, 41454150.CrossRefGoogle ScholarPubMed
Zhao, Y, Qin, P, Sun, H, et al. (2020) Metabolically healthy general and abdominal obesity are associated with increased risk of hypertension. Br J Nutr 123, 583591.CrossRefGoogle ScholarPubMed
van der, AD, Nooyens, AC, van Duijnhoven, FJ, et al. (2014) All-cause mortality risk of metabolically healthy abdominal obese individuals: the EPIC-MORGEN study. Obesity 22, 557564.CrossRefGoogle Scholar
Doustmohamadian, S, Serahati, S, Barzin, M, et al. (2017) Risk of all-cause mortality in abdominal obesity phenotypes: Tehran Lipid and Glucose Study. Nutr Metab Cardiovasc Dis 27, 241248.CrossRefGoogle ScholarPubMed
Loprinzi, PD & Frith, E (2017) Cardiometabolic healthy obesity paradigm and all-cause mortality risk. Eur J Intern Med 43, 4245.CrossRefGoogle ScholarPubMed
Hinnouho, GM, Czernichow, S, Dugravot, A, et al. (2013) Metabolically healthy obesity and risk of mortality: does the definition of metabolic health matter? Diabetes Care 36, 22942300.CrossRefGoogle ScholarPubMed
Al-Khalidi, B, Kimball, SM, Kuk, JL, et al. (2019) Metabolically healthy obesity, vitamin D, and all-cause and cardiometabolic mortality risk in NHANES III. Clin Nutr 38, 820828.CrossRefGoogle ScholarPubMed
Kuk, JL, Rotondi, M, Sui, X, et al. (2018) Individuals with obesity but no other metabolic risk factors are not at significantly elevated all-cause mortality risk in men and women. Clin Obes 8, 305312.CrossRefGoogle Scholar
Bo, S, Musso, G, Gambino, R, et al. (2012) Prognostic implications for insulin-sensitive and insulin-resistant normal-weight and obese individuals from a population-based cohort. Am J Clin Nutr 96, 962969.CrossRefGoogle ScholarPubMed
Zhang, R, Dong, SY, Wang, WM, et al. (2018) Obesity, metabolic abnormalities, and mortality in older men. J Geriatr Cardiol 15, 422427.Google ScholarPubMed
Tian, Q, Wang, A, Zuo, Y, et al. (2020) All-cause mortality in metabolically healthy individuals was not predicted by overweight and obesity. JCI Insight 5, e136982.CrossRefGoogle Scholar
Guo, F & Garvey, WT (2016) Cardiometabolic disease risk in metabolically healthy and unhealthy obesity: stability of metabolic health status in adults. Obesity 24, 516525.CrossRefGoogle ScholarPubMed
Nazare, JA, Smith, JD, Borel, AL, et al. (2012) Ethnic influences on the relations between abdominal subcutaneous and visceral adiposity, liver fat, and cardiometabolic risk profile: the International Study of Prediction of Intra-Abdominal Adiposity and Its Relationship With Cardiometabolic Risk/Intra-Abdominal Adiposity. Am J Clin Nutr 96, 714726.CrossRefGoogle Scholar
Leitzmann, MF, Moore, SC, Koster, A, et al. (2011) Waist circumference as compared with body-mass index in predicting mortality from specific causes. PloS One 6, e18582.CrossRefGoogle ScholarPubMed
Bigaard, J, Frederiksen, K, Tjønneland, A, et al. (2005) Waist circumference and body composition in relation to all-cause mortality in middle-aged men and women. Int J Obes 29, 778784.CrossRefGoogle ScholarPubMed
Ashwell, M, Mayhew, L, Richardson, J, et al. (2014) Waist-to-height ratio is more predictive of years of life lost than body mass index. PloS One 9, e103483.CrossRefGoogle ScholarPubMed
Zhou, Q, Wu, X, Zhang, D, et al. (2020) Age and sex differences in the association between sleep duration and general and abdominal obesity at 6-year follow-up: the rural Chinese cohort study. Sleep Med 69, 7177.CrossRefGoogle ScholarPubMed
Zhao, Y, Zhang, M, Luo, X, et al. (2016) Association of obesity categories and high blood pressure in a rural adult Chinese population. J Hum Hypertens 30, 613618.CrossRefGoogle Scholar
Bondy, SJ, Victor, JC & Diemert, LM (2009) Origin and use of the 100 cigarette criterion in tobacco surveys. Tobacco Contr 18, 317323.CrossRefGoogle ScholarPubMed
Han, C, Liu, Y, Sun, X, et al. (2017) Prediction of a new body shape index and body adiposity estimator for development of type 2 diabetes mellitus: the Rural Chinese Cohort Study. Br J Nutr 118, 771776.CrossRefGoogle ScholarPubMed
Craig, CL, Marshall, AL, Sjöström, M, et al. (2003) International physical activity questionnaire: 12-country reliability and validity. Med Sci Sport Exerc 35, 13811395.CrossRefGoogle ScholarPubMed
The WHO Monica Project (1988) Geographical variation in the major risk factors of coronary heart disease in men and women aged 35–64 years. World Health Stat Q 41, 115140.Google Scholar
Alberti, KG, Eckel, RH, Grundy, SM, et al. (2009) Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation 120, 16401645.CrossRefGoogle Scholar
Scherer, PE & Hill, JA (2016) Obesity, diabetes, and cardiovascular diseases: a compendium. Circ Res 118, 17031705.CrossRefGoogle ScholarPubMed
Alberti, KG, Zimmet, P, Shaw, J, et al. (2005) The metabolic syndrome--a new worldwide definition. Lancet 366, 10591062.CrossRefGoogle ScholarPubMed
Hsieh, SD & Muto, T (2005) The superiority of waist-to-height ratio as an anthropometric index to evaluate clustering of coronary risk factors among non-obese men and women. Prev Med 40, 216220.CrossRefGoogle ScholarPubMed
Liu, L, Chen, X, Liu, Y, et al. (2019) The association between fasting plasma glucose and all-cause and cause-specific mortality by gender: the rural Chinese cohort study. Diabetes/Metab Res Rev 35, e3129.CrossRefGoogle ScholarPubMed
Lee, SH, Jeong, MH, Kim, JH, et al. (2018) Influence of obesity and metabolic syndrome on clinical outcomes of ST-segment elevation myocardial infarction in men undergoing primary percutaneous coronary intervention. J Cardiol 72, 328334.CrossRefGoogle ScholarPubMed
Zheng, R, Zhou, D & Zhu, Y (2016) The long-term prognosis of cardiovascular disease and all-cause mortality for metabolically healthy obesity: a systematic review and meta-analysis. J Epidemiol Community Health 70, 10241031.CrossRefGoogle ScholarPubMed
Arnlöv, J, Ingelsson, E, Sundström, J, et al. (2010) Impact of body mass index and the metabolic syndrome on the risk of cardiovascular disease and death in middle-aged men. Circulation 121, 230236.CrossRefGoogle ScholarPubMed
Kramer, CK, Zinman, B & Retnakaran, R (2013) Are metabolically healthy overweight and obesity benign conditions? A systematic review and meta-analysis. Ann Intern Med 159, 758769.CrossRefGoogle ScholarPubMed
Reis, JP, Loria, CM, Lewis, CE, et al. (2013) Association between duration of overall and abdominal obesity beginning in young adulthood and coronary artery calcification in middle age. JAMA 310, 280288.CrossRefGoogle ScholarPubMed
Bell, JA, Hamer, M, Sabia, S, et al. (2015) The natural course of healthy obesity over 20 years. J Am Coll Cardiol 65, 101102.CrossRefGoogle ScholarPubMed
Caleyachetty, R, Thomas, GN, Toulis, KA, et al. (2017) Metabolically healthy obese and incident cardiovascular disease events among 3·5 million men and women. J Am Coll Cardiol 70, 14291437.CrossRefGoogle ScholarPubMed
Cheng, FW, Gao, X, Mitchell, DC, et al. (2016) Metabolic health status and the obesity paradox in older adults. J Nutr Gerontol Geriatr 35, 161176.CrossRefGoogle ScholarPubMed
Russ, TC, Stamatakis, E, Hamer, M, et al. (2012) Association between psychological distress and mortality: individual participant pooled analysis of 10 prospective cohort studies. BMJ 345, e4933.CrossRefGoogle ScholarPubMed
Batty, GD, Russ, TC, Stamatakis, E, et al. (2017) Psychological distress in relation to site specific cancer mortality: pooling of unpublished data from 16 prospective cohort studies. BMJ 356, j108.CrossRefGoogle ScholarPubMed
Figure 0

Fig. 1. Flow chart of the selection of participants.

Figure 1

Table 1. Baseline characteristics of study participants by metabolic obesity phenotypes

Figure 2

Table 2. Association of metabolic obesity phenotypes at baseline with risk of all-cause mortality

Figure 3

Fig. 2. Association of metabolic obesity phenotypes (by WC) at baseline with risk of all-cause mortality by sex and age. Abbreviations: HR, hazard ratio; MHNO, metabolically healthy non-obesity; MHO, metabolically healthy obesity; MUNO, metabolically unhealthy non-obesity; MUO, metabolically unhealthy obesity; WC, waist circumference. aPer 1000 person-years. bUnadjusted model. cAdjusted for sex, age, alcohol drinking, smoking, physical activity level and education. Each group adjusted for the other covariates except for itself.

Figure 4

Fig. 3. Association of metabolic obesity phenotypes (by WHtR) at baseline with risk of all-cause mortality by sex and age. Abbreviations: HR, hazard ratio; MHNO, metabolically healthy non-obesity; MHO, metabolically healthy obesity; MUNO, metabolically unhealthy non-obesity; MUO, metabolically unhealthy obesity; WHtR, waist-to-height ratio. aPer 1000 person-years. bUnadjusted model. cAdjusted for sex, age, alcohol drinking, smoking, physical activity level and education. Each group adjusted for the other covariates except for itself.

Supplementary material: File

Wu et al. supplementary material

Tables S1 and S2

Download Wu et al. supplementary material(File)
File 26.4 KB