| It is estimated by the World Health Organization,there are 600 million obese individuals worldwide.Obesity is a key risk factor for metabolic syndrome(MetS),type 2 diabetes,coronary heart disease,stroke and certain types of cancer.However,it is found that not all obese populations defined by body mass index(BMI)have MetS and insulin resistance.Nearly 30%of obese individuals present favorable metabolic profiles which are characterized by high levels of insulin sensitivity,normal blood pressue and fasting glucose,as well as favorable lipid fractions,and named as metabolically normal obese(MNO).MNO subjects have been widely studied in many countries.However,because of lacking unified definition of MNO phenotype,there are great differences of the prevalence of MNO phenotype among these studies.Due to rapid transition of dietary behaviors and lifestyles,China is experiencing an epidemic of obesity and metabolic disease in the last few decades.Obesity and related complications have been major public health issue in China.However,few study has examined the characteristics of the MNO phenotype and its influencing factors in the Chinese population.The reason and mechanisms of being MNO phenotype are not explicit.The currently used definitions of MNO phenotype are not included the criteria concerning to the distribution of body fat.The deposition of less visceral and ectopic adipose but more gluteal and femoral fat may be a major determinant of being MNO phenotype.Previous studies provided substantial evidence that upper-body obesity was closely associated with developing metabolic complications.However,there are inconsistent findings of the associations between lower-body fat stores and cardiometabolic risks.Given the difficulties in identifying the causal and effect relationship between lower-body fat accumulation and diseases risk through observational epidemiological studies,Mendelian randomization(MR)has great potential to confirm the relationships.MR is a method for making inferences about causal effects by using genetic instrumental variables(IVs).Because genetic variants are determined at conception and remain constant throughout life,MR is not influenced by reverse causation.Additionally,due to the random assortment of alleles during conception,the effects of genetic variants on exposures are randomly distributed in relation to potential confounders.Therefore,the MR analysis can be used to detect the associations between lower-body obesity and cardio-metabolic states.Previous intervention studies did not classify the obese subjects by their metabolic states,and these studies had proved that calorie restriction and high level physicalactivity were benefit for cardiometabolic factors.However,whether MNO subjects need to lose weight is not ascertained.Additionally,the effects of weight changes on metabolic states for MNO subjects are still not clear.To assess whether MNO subjects are truly healthy or not,the key point is to compare the diseases risks between MNO and the metabolically normal normal-weight(MN-NW)subjects.Many researchers had conducted population-based longitudinal studies to evaluate the associations between MNO phenotype and the risk of developing type 2 diabetes(T2D),CVD and all-cause mortality.But the conclusions are not consistent.Moreover,there are no studies reporting whether there are differences of risks between MNO and MAO subjects.There were many investigations of MNO prevalence reported in western countries,few studies had examined the characteristics of the MNO phenotype in the Chinese population.To systematically explore the epidemiological characteristics of MNO phenotype,the present study firstly analyzed the prevalence of MNO subjects in Chinese.Then the Mendelian Randomization method was used to explore the possible reason of being MNO phenotype.Next,we evaluated the effects of weight change on metabolic states in natural conditions.The fourth,we assessed the diseases and mortality risks for MNO and MAO phenotype by comparing with MN-NW.At last,we made a conclusion and discussed the public health implications for metabolic heterogeneous in obesity.The study included following four sections:Part Ⅰ The present study firstly analyzed the prevalence of MNO subjects among the urban residents in seven cities in China,and then the influencing factors of metabolic heterogeneous in obesity was analyzedPart Ⅱ The aim of this section was to explore a kind of mechanism of being MNO phenotype by using two-sample MR analysis to determine the effects of fat stores in lower body on lipid fractions and type 2 diabetes.Part Ⅲ In this section,the risk of developing metabolic abnormality of metabolically normal overweight/obese(MN-OW/OB)adults was observed by comparing with metabolically normal normal-weight.The impact of weight changes on developing metabolic abnormalities in MN-OW/OB subjects was evaluated.Part Ⅳ To assess the risks of developing T2D,cardiovascular(CV)events,cancer and all-cause mortality for MNO and MAO individuals by using systematical review and meta-analysis.Part Ⅰ The epidemic characteristics of metabolically normal obesity and its influencing factorsObjectives(1)To investigate and analyze the prevalence of MNO phenotype in the urban residents in seven cities,in China.(2)To explore the influencing factors of metabolic status in obesity.Methods and materials(1)In 2010,a cross-sectional investigation was conducted in the urban residents in seven cities in China.The investigated region included Beijing,Taiyuan,Shenyang,Shanghai,Hangzhou,Chengdu and Guangzhou.One or two communities were randomly selected,and the adults in the age above 30 in the community were investigated.The survey included biochemical detection,physical examination and questionnaire investigation.The subjects were classified according to their obese and metabolic phenotypes.The data of the sixth population census in China was used to calculate the age-adjusted prevalence.ResultsThe prevalence of MNO was 3.9%in all the population,age-adjusted prevalence was 2.4%,and 27.9%in obese individuals.In men,the prevalence of MNO was 3.7%in all and 23.4%in obesity;while the prevalence was 4.1%in all and 32.9%in obesity for women.The prevalence of the MNO phenotype was significantly decreased with age in women(Ptrend<0.001),but not significantly in men(Ptrend = 0.349).In northern region,the prevalence of MNO was 4.9%in all and 22.3%in obesity;while the prevalence was 3.3%in all and 36.8%in obesity in southern urban region.The obese subjects with central obesity(odds ratio[OR]= 2.16,95%confidence interval[CI]= 1.27-3.67,P=0.004),longer sedentary time(OR = 1.97,95%CI =1.27-3.06,P=0.002)and family history of obesity related diseases(hypertension,diabetes,dyslipidemia)(OR = 1.85,95%CI = 1.26-2.71,P = 0.002)were significantly associated with having metabolic abnormality.Higher levels of physical activity and more fruit/vegetable intake had decreased ORs of 0.67(95%CI = 0.45-0.98,P = 0.039)and 0.44(95%CI = 0.28-0.70,P=0.001),respectively.SummaryIn Chinese population,the MNO prevalence was 3.9%in all and 27.9%in obese participants were in metabolic health.The proportion of metabolic health in obesity was higher in women than that in men.The prevalence of MNO was various in different regions.Central obesity,physical activity,sedentary time,fruits/vegetables intake and family history of diseases were associated with metabolic status in obesity.PART Ⅱ The causal effects of lower body obesity on lipid fractions and type 2 diabetesObjectivesThe aim of present study was to determine the causal effects of fat stores in lower body on lipid fractions and type 2 diabetes by using two-sample Mendelian Randomization(MR)analysis.Methods and materialsPublished data from genome-wide association studies were used to construct genetic instruments.Seven SNPs were employed as proxy of BMI increased for lower body obesity.Three different models(MR-Egger,inverse-variance weighted and weighted median)were used in the MR analysis.ResultsA 1-SD genetically instrumented increase in BMI for lower-body obesity was associated with lower triglyceride(TG)level(β=-0.81,95%CI =-1.19,-043,P = 0.006)and increased high-density lipoprotein cholesterol(HDL-C)(β = 0.85,95%CI = 0.39,1.31,P = 0.010).There were no causal estimates for total cholesterol and low-density lipoprotein cholesterol per SD increment in BMI for lower-body obesity.There was evidence for causal association of per SD increase in BMI for lower-body obesity with reduced risk of T2D(β =-1.97,95%CI =-2.72,-1.22,P=0.002).SummaryThis study added to the evidence of the protective effects of lower-body obesity on the lipid fractions and T2D.The fat accumulated in lower body might be one of the mechanisms of being MNO phenotype.PART Ⅲ Natural course of the metabolically normal obesity and the effects of weight changes on metabolic statesObjectives(1)The aim of this study was to prospectively observe the natural course of MN-OW/OB and assess the impact of weight changes on metabolic parameters for them.Methods and materialsIn 2009-2010,3853 individuals with age over 30 and residing in the city of Tongxiang and Zhoushan in Zhejiang province,China,were invited to participate in a health survey aimed at identifying risk factors for non-communicable diseases.The metabolic normal subjects were categorized into two groups as 133 overweight/obese and 392 normal weight based on criteria of BMI of 24.0 at baseline.All the subjects were followed-up after five years and administered another health check-up.Weight change categories were defined as weight-loss(percent change of weight<-3%comparing with the level at baseline),weight-maintenance(≥-3%to≤3%)and weight-gain(>3%),respectively.Changes of metabolic parameters from baseline to follow-up were assessed by using mixed effects models with compound symmetry accounting for within-subjects correlation(multiple observation from a participant)with adjusting for the covariates.ResultsAfter 5 years,75.2%of MN-OW/OB and 55.6%of MN-NW adults developed one or more metabolic abnormalities.The MN-OW/OB group was at a significantly elevated risk for pre-diabetes(RR = 1.93,95%CI = 1.43-2.47,P<0.001),high SBP(RR = 1.40,95%CI = 1.05-1.77,P = 0.024),low HDL-C(RR = 1.67,95%CI = 1.18-2.24,P = 0.004)and any one or more metabolic abnormalities(RR = 1.27,95%CI =1.09-1.43,P=0.005).In weight loss group,there was no significant difference in the changes of these metabolic parameters between MN-NW and MN-OW/OB adults.In the weight maintenance category,MN-OW/OB subjects performed significantly larger increase in TG(0.34 vs.0.16 mmol/L,P = 0.030),systolic blood pressure(12.4 vs.7.4 mmHg,P =0.009),diastolic blood pressure(5.6 vs.1.9 mmHg,P = 0.002)and decrease in HDL-C(-0.21 vs.-0.08 mmol/L,P = 0.003)compared with MN-NW peers.In weight gain category,significantly more deleterious changes in fasting plasma glucose,SBP,DBP and HDL-C were observed in MN-OW/OB subjects.SummaryThis study verified that metabolic normal subjects with more fat accumulation still had higher risk of developing metabolic abnormalities than healthy normal-weight subjects.After 5 years,75.2%of the MN-OW/OB subjects had developed one or more metabolic abnormalities.Compared with MN-NW,MN-OW/OB subjects had added 27%risk of developing metabolic abnormalities.The MN-OW/OB subjects still need to lose weight,because weight loss can reduce the risk of developing metabolic abnormality for MN-OW/OB.PART Ⅳ The prognosis of obesity-related diseases and all-cause mortality for metabolically normal obesityObjectivesThe present section was to assess the risks of developing MetS,T2D,cardiovascular(CV)events,cancer,and all-cause mortality for MNO and MAO subjects by using meta-analysis.Methods and materialsA systematic literature search for the published studies was performed using the database of CNKI,Medline,EMBASE,Web of Science and Cochrane library to assess the risks of developing obesity-related diseases for MNO and MAO subjects.ResultsForty-one studies were included in the meta-analysis,and 9 studies evaluated the MetS risk,14 studies for T2D,21 for CV events,three for cancer and 12 for all-cause mortality.Compared with MN-NW subjects,MNO subjects were still present increased risk for MetS(RR = 1.95,95%CI = 1.54-2.47,P<0.001),T2D(RR = 2.73,95%CI=2.15-3.47,P<0.001),and CV events(RR = 1.48,95%CI = 1.23-1.78,P<0.001),but not for cancer(RR = 0.92,95%CI = 0.82-1.03,P = 0.133)and all-cause mortality(RR=1.14,95%CI = 0.86-1.50,P = 0.370).But compared with MAO subjects,MNO had reduced risk for T2D(RR-0.30,95%CI = 0.26-0.35,P<0.001),CV events(RR = 0.59,95%CI = 0.53-0.66,P = 0.001),cancer(RR = 0.57,95%CI = 0.50-0.65,P<0.001),and all-cause mortality(RR = 0.52,95%CI = 0.43-0.63,P<0.001).SummaryCompared with the MN-NW subjects,MNO had increased risk for developing T2D,CV events,but they did not have increased risk for cancer and all-cause mortality.In addition,MNO subjects had 40%-70%lower risks of diseases and mortality than MAO subjects.ConclusionsBased on the aforementioned four parts,we obtained the following conclusions:(1)Obese individuals had different metabolic states.The prevalence of MNO in Chinese was 3.9%,and 27.9%in obese individuals.The prevalence of MNO in Chinese was relatively lower than other countries.Fat distribution,diet,physical activity and family history of obesity-related diseases might be influencing factors for metabolic states in obesity.(2)By conducting MR analysis,we proved that fat accumulation in lower-body was protected for dyslipidemia and type 2 diabetes.It was consistent with the characteristics of MNO phenotype,therefore,the lower-body obesity might be one of the causes of being MNO phenotype.(3)The MNO subjects were in higher risk of developing metabolic abnormalities,and they still needed to loss their weight.Weight loss could lower the risk of developing metabolic abnormalities for MNO subjects.(4)The MNO subjects presented higher risks of developing obesity-related diseases.However,comparing with MAO subjects,MNO had significantly lower risks of developing diseases and mortality. |