Journal of Human Hypertension (2015), 1–7 © 2015 Macmillan Publishers Limited All rights reserved 0950-9240/15 www.nature.com/jhh
ORIGINAL ARTICLE
Hypertension outcomes in metabolically unhealthy normal-weight and metabolically healthy obese children and adolescents WQ Ding1,2,3, YK Yan1,2, MX Zhang1,2, H Cheng1, XY Zhao1, DQ Hou1 and J Mi1,2 Metabolically healthy obesity (MHO) begins in childhood and continues into adulthood. However, the association between MHO and the risk of developing hypertension remains controversial. A prospective cohort study was conducted to investigate the risk of hypertension in MHO and metabolically unhealthy normal-weight (MUNW) Chinese children and adolescents. A total of 1183 participants, 6–18 years old at baseline with normal blood pressure values, were studied using follow-up data from the cohort of the Beijing Child and Adolescent Metabolic Syndrome (BCAMS) study. The participants were classified according to the body mass index and the presence/absence of metabolic abnormality, which was defined by metabolic syndrome (MetS) or insulin resistance (IR). During the 6-year follow-up period, 239 (20.2%) participants developed incident hypertension. After adjusting for age, sex, physical activity, pubertal stage, dietary habits and family history of hypertension, an increased risk for hypertension was observed in the MHO individuals (risk ratio, RRMetS 5.42; 95% confidence interval (CI) 3.19–9.22 and RRIR 7.59; 95% CI 1.64–35.20) compared with their metabolically healthy normal-weight counterparts. Independent of the definition of metabolic abnormality, the MUNW subjects did not have an elevated incidence of hypertension. These results suggest that the risk of developing hypertension is increased in the MHO but not in the MUNW individuals. Journal of Human Hypertension advance online publication, 5 February 2015; doi:10.1038/jhh.2014.124
INTRODUCTION Obesity in children and adolescents has increased at an alarming rate in both developed and developing countries in recent decades and has become a major public health and social problem.1,2 In China, the prevalence of childhood obesity has rapidly increased from 0.2% in 1985 to 8.1% in 2010.3 The most important consequences of obesity include hypertension, type 2 diabetes mellitus and cardiovascular disease (CVD). However, not all obese subjects are at a similar risk for obesity-related metabolic abnormalities.4 Recent research has focused on two representative phenotypes of obesity. The metabolically unhealthy normal-weight (MUNW) phenotype presents as metabolic abnormalities with a normal weight.5 Another phenotype, termed metabolically healthy obese (MHO), presents as obesity with a normal metabolic profile.6 Conflicting evidence regarding the risk of CVD associated with the MHO and MUNW phenotypes exists. A prospective communitybased cohort study reported that the MHO and MUNW phenotypes appear to protect against hypertension.7 Moreover, MHO individuals might not benefit from energy restriction and exercise interventions.8,9 Therefore, there are potential benefits of distinguishing MHO from metabolically unhealthy obese individuals who are more likely to require early pharmacological treatment and lifestyle interventions. In contrast, a recent metaanalysis including eight studies and 61 386 participants reported that obese individuals are at an increased risk of cardiovascular events even in the absence of metabolic abnormalities.10
Hypertension is one of the most important predictors of CVD mortality. The identification of the association of MHO and MUNW phenotypes with the risk of hypertension might partially explain the inconsistent results regarding the risk of CVD events or mortality.7 A community-based longitudinal study suggested that the MHO phenotype begins in childhood and continues into adulthood.11 Given a rapidly increasing prevalence of obesity and hypertension in children and adolescents in China,3,12 a better understanding of the health consequences associated with distinct obese phenotypes would support more efficient strategies for hypertension prevention and management and would benefit both public health and clinical practice. Observations on the relationship between the obesity phenotype and the risk of cardiovascular events have been made in adult populations.7,10 However, to the best of our knowledge, no longitudinal studies have addressed the effects of the MHO and MUNW phenotypes on the risk of developing hypertension in children and adolescents. Therefore, the aims of the present study were to (1) examine the prevalence of the MHO and MUNW phenotypes; and (2) determine the associations of the MHO and MUNW phenotypes with the risk of hypertension using data from a cohort study of children and adolescents. SUBJECTS AND METHODS Subjects The subjects in the present study were part of the Beijing Child and Adolescent Metabolic Syndrome (BCAMS) study. The study sample and
1 Department of Epidemiology, Capital Institute of Pediatrics, Beijing, China; 2Graduate School, Peking Union Medical College, Beijing, China and 3Department of Maternal and Child Health Care, School of Public Health, Ningxia Medical University, Yinchuan, China. Correspondence: Professor J Mi, Department of Epidemiology, Capital Institute of Pediatrics, 2 Yabao Road, Beijing 100020, China. E-mail: [email protected] Received 21 July 2014; revised 13 November 2014; accepted 19 November 2014
Metabolically healthy obesity and hypertension WQ Ding et al
2 data collection have been described elsewhere.13 In total, 2661 children aged 6–14 years (55% boys, n = 1463; 45% girls, n = 1198) underwent a baseline examination in 2004. We included participants who had complete systolic and diastolic blood pressure (SBP and DBP, respectively) measurements and excluded 472 participants with hypertension at baseline. For longitudinal analysis, 1006 individuals who did not attend the follow-up examination at the 6-year interval in 2010 were excluded from the participant pool. The remaining 1183 participants were eligible for analysis. Signed informed consent was obtained from all of the participants and/or from their parents or guardians for all of the parts of the study. The study was approved by the ethics committee at Capital Institute of Pediatrics.
Baseline investigation Anthropometric and laboratorial data were collected using a standardized procedure by trained investigators. Fat mass percentage (FMP) was measured with the subject wearing lightweight clothing using an electronic weighing scale (Tanita TBF-300A Body Composition Analyzer, Tokyo, Japan). Body mass index (BMI) was calculated as the weight per height2 (kg m−2). Pubertal development was assessed by the Tanner stage of breast development in girls and by the testicular volume in boys.14 Participants also completed questionnaires related to their demographic information (age and sex), family history of hypertension, alcohol and cigarette consumption, dietary habits and physical activity. Venous blood samples were drawn by direct venipuncture after overnight fasting (longer than 12 h). Plasma glucose (glucose oxidase method) and serum lipids (enzymatic methods) were assayed using the Hitachi 7060 C (Tokyo, Japan) automatic biochemistry analysis system. The high-density lipoprotein (HDL) cholesterol and low-density lipoprotein cholesterol levels were measured directly. The serum insulin level was measured using monoclonal antibody-based sandwich enzyme-linked immunosorbent assays,15 developed in the Key Laboratory of Endocrinology, Peking Union Medical College Hospital with interassay coefficients of variation of o9%. The previously validated homeostasis model
Table 1.
assessment (HOMA) index was used to estimate the insulin resistance (IR) index,16 which was calculated using the following formula: HOMAIR = fasting insulin (μU ml − 1) × fasting glucose (mmol l − 1)/22.5.
Definitions We used the BMI to define normal weight, overweight and obesity (ageand gender-specific BMI cutoffs were as recommended by the Working Group on Obesity in China17). Metabolic abnormalities were defined by the presence of metabolic syndrome (MetS) components. The MetS components for this study included the following criteria:18 (1) central obesity defined as waist circumference (WC) ⩾ 90th percentile for age and sex; (2) elevated SBP and/or DBP ⩾ 90th percentile for age and sex; (3) hypertriglyceridemia, defined as triglyceride (TG) levels ⩾ 1.24 mmol l − 1; (4) low serum HDL cholesterol defined as ⩽ 1.03 mmol l − 1; and (5) impaired fasting glucose defined as ⩾ 5.6 mmol l − 1. Thus, we defined MHO as obesity with none of the five MetS components and MUNW as normal weight with one or more MetS components. In the secondary analyses, we used IR to define metabolic abnormalities. IR was defined as a HOMA-IR 43.0.19 Similarly, participants with a HOMA-IR ⩽ 3.0 and obesity were classified as MHO, and those with a HOMA-IR 43.0 and normal weight were defined as MUNW.
BP measurement and hypertension incidence At the baseline and follow-up visits, BP was measured by trained examiners according to a standardized protocol using a mercury sphygmomanometer with the subject in the sitting position.20 After a resting period of at least 15 min, BP measurements were performed three times, with each measurement taken every 1–2 min; an appropriately sized arm cuff was used on the right arm, and the mean values of the last two measurements were recorded. Korotkoff phase 1 (K1) and phase 4 (K4) sounds were used to define the SBP and DBP, respectively. Hypertension was defined at baseline if the SBP and (or) DBP were equal to or greater than the 95th percentile for age and gender.21 The same criteria were applied to define incident hypertension at the follow-up examination.
Baseline characteristics of subjects by BMI and metabolic status as defined by MetS components
Variables
Metabolically healthy Normal weight
n Age (years) Sex (boys, %) Pubertal stage (prepuberty, %)a BMI (kg m − 2) WC (cm) FMP (%) SBP (mm Hg) DBP (mmHg) TG (mmol l − 1)b HDL-C (mmol l − 1) LDL-C (mmol l − 1) TC (mmol l − 1) Fasting glucose (mmol l − 1) Physical activity (%)c Frequent dietary consumption Fruits and vegetables (every day) High fat foods/sweetened beverages (⩾2 times/per 2 weeks)d Family history of hypertension (%)e
294 9.3 ± 1.9 46.6 70.0 16.2 ± 1.8 56.7 ± 5.3 15.4 ± 4.3 96 ± 9 61 ± 8 0.69 (0.55, 0.90) 1.63 ± 0.28 2.36 ± 0.74 4.11 ± 0.78 5.00 ± 0.41 85.7
Overweight
P-value
Metabolically unhealthy Obese (MHO)
76 172 9.7 ± 1.9 9.1 ± 1.7Δ 60.5* 55.2 64.2 69.2 20.9 ± 2.0* 23.7 ± 2.5*,Δ 69.3 ± 7.9* 75.9 ± 8.5*,Δ 23.9 ± 5.4* 29.2 ± 5.5*,Δ 102 ± 10 103 ± 7Δ 64 ± 7* 66 ± 6* 0.79 (0.59, 0.93) 0.89 (0.68, 1.08)*,Δ
Normal weight (MUNW) 212 10.8 ± 2.1* 48.1 57.1 17.3 ± 2.2* 60.6 ± 6.3* 17.6 ± 5.3* 101 ± 9* 65 ± 8* 0.92 (0.72, 1.33)*
Overweight
Obese
98 297 11.1 ± 2.0* 10.3 ± 1.9* 50.0 63.0* 52.9 64.9 22.4 ± 2.2* 25.5 ± 2.8* 73.3 ± 6.9* 80.6 ± 8.9* 26.9 ± 5.4* 31.4 ± 6.6* 104 ± 8 109 ± 8 66 ± 7* 69 ± 6* 1.28 (0.86, 1.59)* 1.33 (0.95, 1.70)*
o0.001 o0.001 NS o0.001 o0.001 o0.001 o0.001 o0.001 o0.001
1.52 ± 0.32* 2.45 ± 0.63 4.14 ± 0.67 5.08 ± 0.37 95.8*
1.42 ± 0.23*,Δ 2.53 ± 0.58* 4.10 ± 0.67 5.10 ± 0.36Δ 91.5Δ
1.55 ± 0.36* 2.23 ± 0.73 4.14 ± 0.8 5.67 ± 0.49* 88.1
1.34 ± 0.30* 2.41 ± 0.62 4.08 ± 0.66 5.60 ± 0.52* 85.1
1.31 ± 0.28* 2.51 ± 0.61* 4.21 ± 0.72 5.58 ± 0.99* 82.7
o0.001 o0.001 NS o0.001 0.021
59.3 94.0
63.4 90.3
59.5 93.4
57.1 95.4
54.6 93.8
56.4 93.3
NS NS
20.4
27.6
26.2
12.7*
19.4
15.5
0.003
Abbreviations: BMI, body mass index; DBP, diastolic blood pressure; FMP, fat mass percentage; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; MetS, metabolic syndrome; MHO, metabolically healthy obesity; MUNW, metabolically unhealthy normal weight; NS, not significant; SBP, systolic blood pressure; TC, total cholesterol; TG, triglyceride; WC, waist circumference. Normally distributed data are expressed as mean ± s.d. Median and interquartile range were used for skewed variables. *Po0.05 for post hoc pairwise comparison with metabolically healthy normal weight; ΔPo 0.05 for comparison between MHO and metabolically unhealthy obese. aPrepuberty: Tanner stage 1. bSkewed distributions were logarithmically transformed for statistical tests. cOver one time biweekly and over 30 min per time. dHigh fat food: consumption of at least one meat, fried food or western fast food. ePresence of at least one parent with hypertension.
Journal of Human Hypertension (2015), 1 – 7
© 2015 Macmillan Publishers Limited
Metabolically healthy obesity and hypertension WQ Ding et al
3 Statistical analyses All data are expressed as mean ± s.d., median (25th–75th percentiles) or as proportions. TG levels were not normally distributed, and their logtransformed values were analyzed. Analyses of continuous and categorical variables to assess differences in the baseline characteristics between the categories of BMI and metabolic status were determined using analysis of variance and a Bonferroni post hoc comparison test or the chi-squared test. The crude cumulative hypertension incidence was calculated as the number of subjects with incident hypertension/the number of subjects at risk of hypertension. Risk ratios (RRs) and 95% confidence intervals (CIs) from multivariate logistic regression models were used to estimate the relative risk for cumulative hypertension incidence with the MHO and MUNW phenotypes and were adjusted for confounders including age, sex, pubertal stage, physical activity, dietary habits and family history of hypertension. Statistical analyses were performed using SPSS13.0 (SPSS Inc., Chicago, Illinois, USA) with statistical significance set at Po0.05.
RESULTS Baseline characteristics of the participants The subjects (1183) had an average age of 10.0 ± 0.1 years, and 54.3% (642) were boys. As defined by the MetS components, the MHO phenotype was identified in 36.7% of the obese individuals, and the MUNW phenotype was identified in 41.9% of the normalweight individuals. Participant characteristics by BMI categories and metabolic status are presented in Table 1. Individuals with
Table 2.
MHO or MUNW had worse metabolic profiles, including BMI, WC, FMP, DBP, TG and HDL cholesterol, compared with the metabolically healthy normal-weight individuals. Compared with the metabolically unhealthy obese individuals, the MHO individuals had lower BMI, WC, FMP, SBP, TG and fasting glucose with higher HDL cholesterol and levels of physical activity. Participant characteristics by BMI categories and metabolic status, as defined by IR, are shown in Table 2. MHO phenotype was identified in 3.9% of the obese individuals, and MUNW phenotype was identified in 65.6% of the normal-weight individuals. Compared with metabolically healthy normal-weight individuals, the MHO and MUNW individuals had higher BMI and WC. Hypertension incidence During the 6-year follow-up, 239 (20.2%) participants developed incident hypertension. The hypertension incidence rates were affected by the presence of obesity more than by the presence of IR. The cumulative 6-year hypertension incidence was higher with higher BMI categories, irrespective of the presence of IR. MHO individuals had the highest incidence of hypertension (50%, n = 5). Similar results were observed across the groups in terms of BMI and metabolic status as defined by MetS components. The cumulative incidence rate of hypertension was 36% (n = 62) in MHO individuals (Figure 1).
Baseline characteristics of subjects by BMI and metabolic status as defined by IR
Variables
Metabolically healthy Normal weight
Overweight
P-value
Metabolically unhealthy Obese (MHO)
Normal weight (MUNW)
Overweight
Obese
n 93 11 10 178 80 244 Age (years) 8.3 ± 1.6 9.3 ± 1.6 8.8 ± 1.5 9.1 ± 1.5 9.2 ± 1.5* 9.0 ± 1.5* Sex (boys, %) 55.9 18.2 30.0 59.0* 55.0 37.7 81.5 72.7 100.0 59.4* 58.7* 65.7* Pubertal stage a (prepuberty, %) BMI (kg m − 2) 15.4 ± 1.4 20.4 ± 2.0* 23.3 ± 2.6* 16.4 ± 1.7* 20.4 ± 1.5* 23.9 ± 2.2* WC (cm) 54.5 ± 4.7 69.5 ± 6.8* 76.5 ± 8.8* 57.7 ± 5.3* 68.4 ± 7.6* 76.8 ± 8.1* FMP (%) 14.2 ± 3.7 24.2 ± 5.2* 31.1 ± 7.3* 15.8 ± 4.2 24.8 ± 4.3* 29.3 ± 5.4* SBP (mm Hg) 93 ± 9 98 ± 10 98 ± 9 95.9 ± 9.9 101 ± 9* 105 ± 7* DBP (mm Hg) 59 ± 7 61 ± 7 61 ± 6 59.2 ± 8.6 64 ± 8* 67 ± 6* 0.65 (0.50, 0.85) 0.79 (0.56, 0.98) 0.84 (0.54, 1.06) 0.73 (0.57, 0.98) 0.96 (0.72, 1.32)* 1.04 (0.74, 1.37)* TG (mmol l − 1)b 1.63 ± 0.34 1.59 ± 0.31 1.39 ± 0.24 1.57 ± 0.31 1.36 ± 0.33* 1.34 ± 0.27* HDL-C (mmol l − 1) 2.48 ± 0.66 2.90 ± 0.53 2.90 ± 0.53 2.53 ± 0.94 2.59 ± 0.63 2.62 ± 0.60 LDL-C (mmol l − 1) −1 4.14 ± 0.72 4.59 ± 0.55 4.36 ± 0.61 4.15 ± 1.00 4.12 ± 0.69 4.13 ± 0.69 TC (mmol l ) 4.78 ± 0.42 4.82 ± 0.50 4.83 ± 0.66 5.04 ± 0.45* 5.09 ± 0.46* 5.10 ± 0.39* Fasting glucose (mmol l − 1) 1.91 (1.48, 2.53) 2.48 (2.03, 2.69) 2.03 (1.31, 2.51)Δ 5.56 (4.02, 7.07) 7.99 (5.63, 10.50)* 9.92 (6.80, 13.22)* Fasting insulinb b Δ 0.40 (0.31, 0.55) 0.55 (0.39, 0.63) 0.44 (0.24, 0.57) 1.26 (0.89, 1.71) 1.78 (1.20, 2.42)* 2.18 (1.53, 2.92)* HOMA-IR 89.9 90.0 88.9 86.7 91.0 91.2 Physical activity (%)c Frequent dietary consumption Fruits and vegetables (every day) High fat foods/ sweetened beverages (⩾2 times/per 2 weeks)d Family history of hypertension (%)e
o0.001 o0.001 0.001 o0.001 o0.001 o0.001 o0.001 o0.001 o0.001 o0.001 o0.001 NS NS o0.001 o0.001 NS
53.3
63.6
90.0
56.6
55.8
53.1
NS
88.9
90.9
100
93.0
92.3
91.6
NS
26.9
27.3
40.0
35.4
45.0
35.7
NS
Abbreviations: BMI, body mass index; DBP, diastolic blood pressure; FMP, fat mass percentage; HDL-C, high-density lipoprotein cholesterol; HOMA-IR, homeostatic model assessment of insulin resistance; LDL-C, low-density lipoprotein cholesterol; MetS, metabolic syndrome; MHO, metabolically healthy obesity; MUNW, metabolically unhealthy normal weight; NS, not significant; SBP, systolic blood pressure; TC, total cholesterol; TG, triglyceride; WC, waist circumference. Of 1183 volunteers, 616 participants had their insulin measured. Normally distributed data are expressed as mean ± s.d. Median and interquartile range were used for skewed variables. *Po0.05 for post hoc pairwise comparison with metabolically healthy normal weight. ΔPo 0.05 for comparison between MHO and metabolically unhealthy obese. aPrepuberty: Tanner stage 1. bSkewed distributions were logarithmically transformed for statistical tests. cOver one time biweekly and over 30 min per time. dHigh fat food: consumption of at least one meat, fried food or western fast food. ePresence of at least one parent with hypertension.
© 2015 Macmillan Publishers Limited
Journal of Human Hypertension (2015), 1 – 7
Metabolically healthy obesity and hypertension WQ Ding et al
4
Figure 1. Hypertension incidence rates for different categories of BMI and metabolic status. (a) Metabolic status was defined by metabolic syndrome components. (b) Metabolic status was defined by insulin resistance. BMI, body mass index; MHO, metabolically healthy obesity; MUNW, metabolically unhealthy normal weight.
We analyzed the risk of incident hypertension caused by BMI and metabolic status after controlling for age, sex, pubertal stage, physical activity, dietary habits and family history of hypertension. For metabolic status defined by MetS components, the MHO individuals had an increased hypertension incidence (RR = 5.42, 95% CI 3.19–9.22). However, the MUNW phenotype was not related to hypertension incidence (RR = 1.23, 95% CI 0.65–2.34). Using IR to define the metabolic status, the risk of developing hypertension was higher for the higher BMI categories in the metabolically healthy and unhealthy individuals compared with the metabolically healthy normal-weight individuals. Hypertension incidence was elevated in the MHO subjects (RR = 7.59, 95% CI 1.64–35.20), but was not elevated significantly in the MUNW subjects (RR = 1.39, 95% CI 0.57–3.39; Table 3). DISCUSSION This study yielded two key findings. First, the MHO and MUNW phenotypes, defined by MetS components, were common in Chinese children and adolescents. Second, an increased risk of hypertension was observed in the MHO but not in the MUNW individuals, independent of the definition of metabolic status. Although there is no consensus on the definition of the MHO phenotype, a few studies have reported that 6.3–31.5% of overweight or obese children and adolescents are classified as metabolically healthy.22–24 Because different definitions of Journal of Human Hypertension (2015), 1 – 7
metabolic abnormalities are used in these studies, the prevalence is not easily compared among studies. Nonetheless, our MHO phenotype defined by MetS and IR prevalence of 36.7% and 3.9% among obese individuals is approximately in agreement with the above studies. A cohort study performed in 40- to 69-year-old adults reported that the hypertension incidence rates were influenced by the presence of obesity more than by the presence of metabolic status, and metabolically unhealthy obese individuals have the highest incidence of hypertension.5 Similarly, our study indicates that obese individuals are at a higher incidence of hypertension before adjustment for potential confounding factors, regardless of the metabolic status. The MHO individuals had a much higher cumulative incidence of hypertension than the corresponding metabolically unhealthy obese individuals. This finding should be interpreted with caution because the MHO group had fewer individuals and events, and therefore, the findings had lower precision. In a 17.4-year follow-up study, Hinnouho et al.25 found that MHO individuals had higher risk of incident CVD compared with metabolically healthy normal-weight individuals, regardless of the metabolic health definition used, with the exception of HOMA. Similarly, a prospective study of Koreans reported that MHO phenotype had a significant impact on the development of hypertension.7 A more recent meta-analysis of two communitybased cohort studies indicated that the absence of IR and other © 2015 Macmillan Publishers Limited
Metabolically healthy obesity and hypertension WQ Ding et al
5 Table 3.
Incident hypertension by BMI and metabolic status as defined by MetS components or IR Number at risk
Number with incident hypertension
Model 1
Model 2
RR
95% CI
RR
95% CI
Defined by MetS components Metabolically healthy normal weight Metabolically healthy overweight MHO MUNW Metabolically unhealthy overweight Metabolically unhealthy obese
294 76 172 212 98 297
24 13 62 21 21 91
1.00 (Reference) 2.32 6.34 1.24 3.07 4.97
1.12–4.81 3.76–10.67 0.66–2.29 1.62–5.81 3.06–8.07
1.00 (Reference) 2.25 5.42 1.23 3.11 4.65
1.07–4.74 3.19–9.22 0.65–2.34 1.59–6.07 2.82–7.66
Defined by IR Metabolically healthy normal weight Metabolically healthy overweight MHO MUNW Metabolically unhealthy overweight Metabolically unhealthy obese
93 11 10 178 80 244
8 2 5 18 19 83
1.00 (Reference) 2.36 10.62 1.19 3.31 5.47
0.43–12.86 2.53–44.66 0.49–2.86 1.36–8.05 2.53–11.85
1.00 (Reference) 2.46 7.59 1.39 3.99 5.14
0.44–13.89 1.64–35.20 0.57–3.39 1.59–10.00 2.33–11.32
Abbreviations: BMI, body mass index; CI, confidence interval; IR, insulin resistance; MetS, metabolic syndrome; MHO, metabolically healthy obesity; MUNW, metabolically unhealthy normal weight; RR, relative risk. Model 1: unadjusted for confounding factors Model 2: adjusted for age, sex, pubertal stage, physical activity, dietary habits and family history of hypertension.
metabolic derangements in overweight or obese individuals provided no protection against the development of hypertension.26 In line with these previous studies, our findings indicated that the MHO phenotype was positively associated with the development of hypertension and that the role of obesity was greater than that of metabolic abnormalities in the development of hypertension. However, several cohort studies have reported that MHO individuals were not at an increased risk for CVD, regardless of the use an MHO definition based on a combination of two or more metabolic risk factors.27,28 The mechanisms underlying the medical consequences of the MHO phenotype are uncertain, and limited data are available. It has been reported that dietary and lifestyle factors play important roles in the development of IR and can lead to an increased risk of associated cardiometabolic abnormalities.29 Compared with metabolically unhealthy obese individuals, MHO individuals have lower WC and moderate-tohigh levels of physical activity, despite having similar BMI values.30 Phillips et al. recently demonstrated that dietary macronutrient composition, dietary quality assessed using the DASH (Dietary Approaches to Stop Hypertension) score and physical activity were similar between the metabolically healthy and unhealthy individuals regardless of BMI. Higher dietary quality was positively associated with metabolic health in obese and non-obese individuals. In addition, a moderate-to-high level of physical activity was positively associated with the MHO phenotype, as defined by IR.31 In line with the above studies, we also found that the dietary consumption patterns (including fruits and vegetables and high fat foods/sweetened beverages) did not differ significantly between the metabolically healthy and unhealthy participants regardless of BMI. However, in the present study, although the MHO individuals had lower BMI, lower WC, FMP, SBP, TG, fasting glucose levels, higher HDL cholesterol, better dietary habits and physical activity levels than their metabolically unhealthy obese counterparts, the MHO individuals still showed an increased risk of hypertension. These discrepancies between the present study and previous studies might be explained by the fact that there is a long-term effect of metabolic abnormalities on the risk of incident hypertension. The results of a longitudinal study are relatively more generalizable than the results of a cross-sectional study. This © 2015 Macmillan Publishers Limited
is supported by a prospective cohort study that found that an increased risk for incident hypertension was not apparent after 4 years of follow-up, although an increased risk was apparent after 8 years.32 Our study had a follow-up period of 6 years. Another difference between the studies is that metabolic health was defined in different ways.33 In recent years, the MUNW phenotype has received much attention. The results of the Third National Health and Nutrition Examination Survey (NHANES III) in the USA demonstrated that 0.1% of adolescents were normal-weight individuals with MetS.23 A recent meta-analysis showed that all of the unhealthy metabolic status phenotypes presented an increased risk for CVD, regardless of whether the subjects were normal weight, overweight or obese.10 An association between IR and the development of hypertension has been demonstrated in normal-weight individuals.34 However, in the present study, relative to the metabolically healthy normal-weight participants, a higher risk of incident hypertension was not observed in the MUNW participants. Our results are supported by Lytsy et al. who reported that normal weight with IR is not related to the development of hypertension.30 Our study has several limitations. First, because Chinese criteria were used to define hypertension, the results are not applicable to other ethnicities. Second, we did not assess some risk factors of hypertension, such as dietary salt intake. Third, the small subgroups made the study underpowered for comparing several phenotype groups. Despite these limitations, there are several strengths to this study. We recruited children and adolescents, in which incident hypertension has not been examined previously with regard to the MHO and MUNW phenotypes using prospective cohorts. Furthermore, because of the lack of a uniform definition of the MUNW and MHO phenotypes, we defined metabolic status in two ways. The MHO phenotype, as defined by MetS components in obese children and adolescents, is common. These findings add to the evidence demonstrating that MHO individuals have an increased risk of hypertension and MUNW individuals are not associated with a higher incidence of hypertension. The potential benefits of differentiating the MHO and MUNW phenotypes in clinical practice in children and adolescents appear limited. Journal of Human Hypertension (2015), 1 – 7
Metabolically healthy obesity and hypertension WQ Ding et al
6 To date, there is no standard definition of metabolic health. Different inclusion criteria, including IR, MetS components, inflammation and multiple cardiometabolic health features have been used to discriminate metabolically healthy from metabolically unhealthy individuals.35 Thus, it is difficult to make comparisons among studies. Given the lack of comprehensiveness regarding the use of a single metabolic marker, combined metabolic markers should be recommended for use in future studies. More prospective and intervention studies are needed to clarify which abnormal metabolic profile is a better definition of metabolic health in children and adolescents. To date, it is difficult to use HOMA-IR to define the metabolic health widely because the methodology to measure insulin has not yet been standardized. In addition, it has been demonstrated that inflammation is secondary to obesity and/or IR in humans. Therefore, the definition of metabolic health based on five components of the MetS, including BP, fasting plasma glucose, TG levels, HDL cholesterol and WC, would be feasible for use in large population-based studies that would help to understand the phenomenon of MHO more comprehensively in future. This also would allow consistent comparison among diverse populations. Moreover, such a definition may represent a first step toward a more detailed definition, including other metabolic markers such as IR and inflammation.35 In addition, further studies exploring the potential mechanisms underlying metabolically healthy obesity and hypertension are warranted. What is known about the topic: ● The MHO phenotype begins in childhood and continues into adulthood. Observations on the relationship between this obesity phenotype and the risk of cardiovascular events have been made in adult populations. ● The association between the MHO phenotype and the risk of developing hypertension is controversial. ● No longitudinal studies have addressed the effects of the MHO and MUNW phenotypes on the risk of developing hypertension in children and adolescents. What this study adds ● The MHO phenotype, as defined by metabolic syndrome components in obese children and adolescents, was common in children and adolescents. ● After adjusting for age, sex, physical activity, pubertal stage, dietary consumption patterns and family history of hypertension, an increased risk for hypertension was observed in the MHO individuals. Independent of the definition of metabolic abnormality, the MUNW subjects did not have an elevated incidence of hypertension. ● The results suggest that the potential benefits of differentiating the MHO and MUNW phenotypes in children and adolescents in clinical practice appear limited.
CONFLICT OF INTEREST The authors declare no conflicts of interest.
ACKNOWLEDGEMENTS This study was supported by a National Natural Science Foundation of China grant (81172746), the Beijing Municipal and Technology Commission program (D111100000611002), and the National Science and Technology program funded by the Science and Technology Committee of Beijing (2012BAI03B00).
REFERENCES 1 Singh GK, Kogan MD, van Dyck PC. Changes in state-specific childhood obesity and overweight prevalence in the United States from 2003 to 2007. Arch Pediatr Adolesc Med 2010; 164: 598–607.
Journal of Human Hypertension (2015), 1 – 7
2 Wang Y, Monteiro C, Popkin BM. Trends of obesity and underweight in older children and adolescents in the United States, Brazil, China, and Russia. Am J Clin Nutr 2002; 75: 971–977. 3 Song Y, Wang HJ, Ma J, Wang Z. Secular trends of obesity prevalence in urban Chinese children from 1985 to 2010: gender disparity. PLoS ONE 2013; 8: e53069. 4 Genelhu VA, Celoria BM, Duarte SF, Cabello PH, Francischetti EA. Not all obese subjects of multiethnic origin are at similar risk for developing hypertension and type 2 diabetes. Eur J Intern Med 2009; 20: 289–295. 5 Lee SH, Ha HS, Park YJ, Lee JH, Yim HW, Yoon KH et al. Identifying metabolically obese but normal-weight (MONW) individuals in a nondiabetic Korean population: the Chungju Metabolic disease Cohort (CMC) study. Clin Endocrinol (Oxf) 2011; 75: 475–481. 6 Brochu M, Tchernof A, Dionne IJ, Sites CK, Eltabbakh GH, Sims EA et al. What are the physical characteristics associated with a normal metabolic profile despite a high level of obesity in postmenopausal women?. J Clin Endocrinol Metab 2001; 86: 1020–1025. 7 Lee SK, Kim SH, Cho GY, Baik I, Lim HE, Park CG et al. Obesity phenotype and incident hypertension: a prospective community-based cohort study. J Hypertens 2013; 31: 145–151. 8 Karelis AD, Messier V, Brochu M, Rabasa-Lhoret R. Metabolically healthy but obese women: effect of an energy-restricted diet. Diabetologia 2008; 51: 1752–1754. 9 Arsenault BJ, Cote M, Cartier A, Lemieux I, Despres JP, Ross R et al. Effect of exercise training on cardiometabolic risk markers among sedentary, but metabolically healthy overweight or obese post-menopausal women with elevated blood pressure. Atherosclerosis 2009; 207: 530–533. 10 Kramer CK, Zinman B, Retnakaran R. Are metabolically healthy overweight and obesity benign conditions? A systematic review and meta-analysis. Ann Intern Med 2013; 159: 758–769. 11 Li S, Chen W, Srinivasan SR, Xu J, Berenson GS. Relation of childhood obesity/ cardiometabolic phenotypes to adult cardiometabolic profile: the Bogalusa Heart Study. Am J Epidemiol 2012; 176(Suppl 7): S142–S149. 12 Xi B, Liang Y, Mi J. Hypertension trends in Chinese children in the national surveys, 1993 to 2009. Int J Cardiol 2013; 165: 577–579. 13 Shan XY, Xi B, Cheng H, Hou DQ, Wang Y, Mi J. Prevalence and behavioral risk factors of overweight and obesity among children aged 2-18 in Beijing, China. Int J Pediatr Obes 2010; 5: 383–389. 14 Chen FF, Mi J, Wang TY, Li H, Hou DQ, Cheng H et al. Relationship between precosious puberty and obese status in Beijing children and adolescents. Chin J Evid Based Pediatr 2007; 12: 14–20. 15 Li M, Wu C, Song A, Zhang K. Development and preliminary application of enzyme-linked immunosorbent assay for human net insulin in serum. Chin J Endocrinol Metab 1997; 13: 214–217. 16 Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28: 412–419. 17 Ji CY. Working Group on Obesity in China Report on childhood obesity in China (1)—body mass index reference for screening overweight and obesity in Chinese school-age children. Biomed Environ Sci 2005; 18: 390–400. 18 Wan NJ, Mi J, Wang TY, Duan JL, Li M, Gong CX et al. Metabolic syndrome in overweight and obese schoolchildren in Beijing. Zhonghua Er Ke Za Zhi 2007; 45: 417–421. 19 Yin J, Li M, Xu L, Wang Y, Cheng H, Zhao X et al. Insulin resistance determined by homeostasis model assessment (HOMA) and associations with metabolic syndrome among Chinese children and teenagers. Diabetol Metab Syndr 2013; 5: 71. 20 Lawes CM, Vander Hoorn S, Law MR, Elliott P, MacMahon S, Rodgers A. Blood pressure and the global burden of disease 2000. Part 1: estimates of blood pressure levels. J Hypertens 2006; 24: 413–422. 21 Mi J, Wang TY, Meng LH, Zhu GJ, Han SM, Zhong Y et al. Development of blood pressure standards for Chinese childen and adolescents. Chin J Evid Based Pediatr 2010; 5: 4–14. 22 Prince RL, Kuk JL, Ambler KA, Dhaliwal J, Ball GD. Predictors of metabolically healthy obesity in children. Diabetes Care 2014; 37: 1462–1468. 23 Cook S, Weitzman M, Auinger P, Nguyen M, Dietz WH. Prevalence of a metabolic syndrome phenotype in adolescents: findings from the third National Health and Nutrition Examination Survey, 1988-1994. Arch Pediatr Adolesc Med 2003; 157: 821–827. 24 Bokor S, Frelut ML, Vania A, Hadjiathanasiou CG, Anastasakou M, Malecka-Tendera E et al. Prevalence of metabolic syndrome in European obese children. Int J Pediatr Obes 2008; 3(Suppl 2): 3–8. 25 Hinnouho GM, Czernichow S, Dugravot A, Nabi H, Brunner EJ, Kivimaki M et al. Metabolically healthy obesity and the risk of cardiovascular disease and type 2 diabetes: the Whitehall II cohort study. Eur Heart J (e-pub ahead of print 26 March 2014; doi:10.1093/eurheartj/ehu123).
© 2015 Macmillan Publishers Limited
Metabolically healthy obesity and hypertension WQ Ding et al
7 26 Lytsy P, Ingelsson E, Lind L, Arnlov J, Sundstrom J. Interplay of overweight and insulin resistance on hypertension development. J Hypertens 2014; 32: 834–839. 27 Hamer M, Stamatakis E. Metabolically healthy obesity and risk of all-cause and cardiovascular disease mortality. J Clin Endocrinol Metab 2012; 97: 2482–2488. 28 Calori G, Lattuada G, Piemonti L, Garancini MP, Ragogna F, Villa M et al. Prevalence, metabolic features, and prognosis of metabolically healthy obese Italian individuals: the Cremona Study. Diabetes Care 2011; 34: 210–215. 29 Hotamisligil GS. Inflammation and metabolic disorders. Nature 2006; 444: 860–867. 30 Appleton SL, Seaborn CJ, Visvanathan R, Hill CL, Gill TK, Taylor AW et al. Diabetes and cardiovascular disease outcomes in the metabolically healthy obese phenotype: a cohort study. Diabetes Care 2013; 36: 2388–2394. 31 Phillips CM, Dillon C, Harrington JM, McCarthy VJ, Kearney PM, Fitzgerald AP et al. Defining metabolically healthy obesity: role of dietary and lifestyle factors. PLoS ONE 2013; 8: e76188.
© 2015 Macmillan Publishers Limited
32 Arsenault BJ, Cote M, Cartier A, Lemieux I, Despres JP, Ross R. Effect of exercise training on cardiometabolic risk markers among sedentary, but metabolically healthy overweight or obese post-menopausal women with elevated blood pressure. Atherosclerosis 2009; 207: 530–533. 33 Hinnouho GM, Czernichow S, Dugravot A, Batty GD, Kivimaki M, Singh-Manoux A. Metabolically healthy obesity and risk of mortality: does the definition of metabolic health matter? Diabetes Care 2013; 36: 2294–2300. 34 Arnlov J, Pencina MJ, Nam BH, Meigs JB, Fox CS, Levy D et al. Relations of insulin sensitivity to longitudinal blood pressure tracking: variations with baseline age, body mass index, and blood pressure. Circulation 2005; 112: 1719–1727. 35 Rey-Lopez JP, de Rezende LF, Pastor-Valero M, Tess BH. The prevalence of metabolically healthy obesity: a systematic review and critical evaluation of the definitions used. Obes Rev 2014; 15: 781–790.
Journal of Human Hypertension (2015), 1 – 7
ORIGINAL ARTICLE
Hypertension outcomes in metabolically unhealthy normal-weight and metabolically healthy obese children and adolescents WQ Ding1,2,3, YK Yan1,2, MX Zhang1,2, H Cheng1, XY Zhao1, DQ Hou1 and J Mi1,2 Metabolically healthy obesity (MHO) begins in childhood and continues into adulthood. However, the association between MHO and the risk of developing hypertension remains controversial. A prospective cohort study was conducted to investigate the risk of hypertension in MHO and metabolically unhealthy normal-weight (MUNW) Chinese children and adolescents. A total of 1183 participants, 6–18 years old at baseline with normal blood pressure values, were studied using follow-up data from the cohort of the Beijing Child and Adolescent Metabolic Syndrome (BCAMS) study. The participants were classified according to the body mass index and the presence/absence of metabolic abnormality, which was defined by metabolic syndrome (MetS) or insulin resistance (IR). During the 6-year follow-up period, 239 (20.2%) participants developed incident hypertension. After adjusting for age, sex, physical activity, pubertal stage, dietary habits and family history of hypertension, an increased risk for hypertension was observed in the MHO individuals (risk ratio, RRMetS 5.42; 95% confidence interval (CI) 3.19–9.22 and RRIR 7.59; 95% CI 1.64–35.20) compared with their metabolically healthy normal-weight counterparts. Independent of the definition of metabolic abnormality, the MUNW subjects did not have an elevated incidence of hypertension. These results suggest that the risk of developing hypertension is increased in the MHO but not in the MUNW individuals. Journal of Human Hypertension advance online publication, 5 February 2015; doi:10.1038/jhh.2014.124
INTRODUCTION Obesity in children and adolescents has increased at an alarming rate in both developed and developing countries in recent decades and has become a major public health and social problem.1,2 In China, the prevalence of childhood obesity has rapidly increased from 0.2% in 1985 to 8.1% in 2010.3 The most important consequences of obesity include hypertension, type 2 diabetes mellitus and cardiovascular disease (CVD). However, not all obese subjects are at a similar risk for obesity-related metabolic abnormalities.4 Recent research has focused on two representative phenotypes of obesity. The metabolically unhealthy normal-weight (MUNW) phenotype presents as metabolic abnormalities with a normal weight.5 Another phenotype, termed metabolically healthy obese (MHO), presents as obesity with a normal metabolic profile.6 Conflicting evidence regarding the risk of CVD associated with the MHO and MUNW phenotypes exists. A prospective communitybased cohort study reported that the MHO and MUNW phenotypes appear to protect against hypertension.7 Moreover, MHO individuals might not benefit from energy restriction and exercise interventions.8,9 Therefore, there are potential benefits of distinguishing MHO from metabolically unhealthy obese individuals who are more likely to require early pharmacological treatment and lifestyle interventions. In contrast, a recent metaanalysis including eight studies and 61 386 participants reported that obese individuals are at an increased risk of cardiovascular events even in the absence of metabolic abnormalities.10
Hypertension is one of the most important predictors of CVD mortality. The identification of the association of MHO and MUNW phenotypes with the risk of hypertension might partially explain the inconsistent results regarding the risk of CVD events or mortality.7 A community-based longitudinal study suggested that the MHO phenotype begins in childhood and continues into adulthood.11 Given a rapidly increasing prevalence of obesity and hypertension in children and adolescents in China,3,12 a better understanding of the health consequences associated with distinct obese phenotypes would support more efficient strategies for hypertension prevention and management and would benefit both public health and clinical practice. Observations on the relationship between the obesity phenotype and the risk of cardiovascular events have been made in adult populations.7,10 However, to the best of our knowledge, no longitudinal studies have addressed the effects of the MHO and MUNW phenotypes on the risk of developing hypertension in children and adolescents. Therefore, the aims of the present study were to (1) examine the prevalence of the MHO and MUNW phenotypes; and (2) determine the associations of the MHO and MUNW phenotypes with the risk of hypertension using data from a cohort study of children and adolescents. SUBJECTS AND METHODS Subjects The subjects in the present study were part of the Beijing Child and Adolescent Metabolic Syndrome (BCAMS) study. The study sample and
1 Department of Epidemiology, Capital Institute of Pediatrics, Beijing, China; 2Graduate School, Peking Union Medical College, Beijing, China and 3Department of Maternal and Child Health Care, School of Public Health, Ningxia Medical University, Yinchuan, China. Correspondence: Professor J Mi, Department of Epidemiology, Capital Institute of Pediatrics, 2 Yabao Road, Beijing 100020, China. E-mail: [email protected] Received 21 July 2014; revised 13 November 2014; accepted 19 November 2014
Metabolically healthy obesity and hypertension WQ Ding et al
2 data collection have been described elsewhere.13 In total, 2661 children aged 6–14 years (55% boys, n = 1463; 45% girls, n = 1198) underwent a baseline examination in 2004. We included participants who had complete systolic and diastolic blood pressure (SBP and DBP, respectively) measurements and excluded 472 participants with hypertension at baseline. For longitudinal analysis, 1006 individuals who did not attend the follow-up examination at the 6-year interval in 2010 were excluded from the participant pool. The remaining 1183 participants were eligible for analysis. Signed informed consent was obtained from all of the participants and/or from their parents or guardians for all of the parts of the study. The study was approved by the ethics committee at Capital Institute of Pediatrics.
Baseline investigation Anthropometric and laboratorial data were collected using a standardized procedure by trained investigators. Fat mass percentage (FMP) was measured with the subject wearing lightweight clothing using an electronic weighing scale (Tanita TBF-300A Body Composition Analyzer, Tokyo, Japan). Body mass index (BMI) was calculated as the weight per height2 (kg m−2). Pubertal development was assessed by the Tanner stage of breast development in girls and by the testicular volume in boys.14 Participants also completed questionnaires related to their demographic information (age and sex), family history of hypertension, alcohol and cigarette consumption, dietary habits and physical activity. Venous blood samples were drawn by direct venipuncture after overnight fasting (longer than 12 h). Plasma glucose (glucose oxidase method) and serum lipids (enzymatic methods) were assayed using the Hitachi 7060 C (Tokyo, Japan) automatic biochemistry analysis system. The high-density lipoprotein (HDL) cholesterol and low-density lipoprotein cholesterol levels were measured directly. The serum insulin level was measured using monoclonal antibody-based sandwich enzyme-linked immunosorbent assays,15 developed in the Key Laboratory of Endocrinology, Peking Union Medical College Hospital with interassay coefficients of variation of o9%. The previously validated homeostasis model
Table 1.
assessment (HOMA) index was used to estimate the insulin resistance (IR) index,16 which was calculated using the following formula: HOMAIR = fasting insulin (μU ml − 1) × fasting glucose (mmol l − 1)/22.5.
Definitions We used the BMI to define normal weight, overweight and obesity (ageand gender-specific BMI cutoffs were as recommended by the Working Group on Obesity in China17). Metabolic abnormalities were defined by the presence of metabolic syndrome (MetS) components. The MetS components for this study included the following criteria:18 (1) central obesity defined as waist circumference (WC) ⩾ 90th percentile for age and sex; (2) elevated SBP and/or DBP ⩾ 90th percentile for age and sex; (3) hypertriglyceridemia, defined as triglyceride (TG) levels ⩾ 1.24 mmol l − 1; (4) low serum HDL cholesterol defined as ⩽ 1.03 mmol l − 1; and (5) impaired fasting glucose defined as ⩾ 5.6 mmol l − 1. Thus, we defined MHO as obesity with none of the five MetS components and MUNW as normal weight with one or more MetS components. In the secondary analyses, we used IR to define metabolic abnormalities. IR was defined as a HOMA-IR 43.0.19 Similarly, participants with a HOMA-IR ⩽ 3.0 and obesity were classified as MHO, and those with a HOMA-IR 43.0 and normal weight were defined as MUNW.
BP measurement and hypertension incidence At the baseline and follow-up visits, BP was measured by trained examiners according to a standardized protocol using a mercury sphygmomanometer with the subject in the sitting position.20 After a resting period of at least 15 min, BP measurements were performed three times, with each measurement taken every 1–2 min; an appropriately sized arm cuff was used on the right arm, and the mean values of the last two measurements were recorded. Korotkoff phase 1 (K1) and phase 4 (K4) sounds were used to define the SBP and DBP, respectively. Hypertension was defined at baseline if the SBP and (or) DBP were equal to or greater than the 95th percentile for age and gender.21 The same criteria were applied to define incident hypertension at the follow-up examination.
Baseline characteristics of subjects by BMI and metabolic status as defined by MetS components
Variables
Metabolically healthy Normal weight
n Age (years) Sex (boys, %) Pubertal stage (prepuberty, %)a BMI (kg m − 2) WC (cm) FMP (%) SBP (mm Hg) DBP (mmHg) TG (mmol l − 1)b HDL-C (mmol l − 1) LDL-C (mmol l − 1) TC (mmol l − 1) Fasting glucose (mmol l − 1) Physical activity (%)c Frequent dietary consumption Fruits and vegetables (every day) High fat foods/sweetened beverages (⩾2 times/per 2 weeks)d Family history of hypertension (%)e
294 9.3 ± 1.9 46.6 70.0 16.2 ± 1.8 56.7 ± 5.3 15.4 ± 4.3 96 ± 9 61 ± 8 0.69 (0.55, 0.90) 1.63 ± 0.28 2.36 ± 0.74 4.11 ± 0.78 5.00 ± 0.41 85.7
Overweight
P-value
Metabolically unhealthy Obese (MHO)
76 172 9.7 ± 1.9 9.1 ± 1.7Δ 60.5* 55.2 64.2 69.2 20.9 ± 2.0* 23.7 ± 2.5*,Δ 69.3 ± 7.9* 75.9 ± 8.5*,Δ 23.9 ± 5.4* 29.2 ± 5.5*,Δ 102 ± 10 103 ± 7Δ 64 ± 7* 66 ± 6* 0.79 (0.59, 0.93) 0.89 (0.68, 1.08)*,Δ
Normal weight (MUNW) 212 10.8 ± 2.1* 48.1 57.1 17.3 ± 2.2* 60.6 ± 6.3* 17.6 ± 5.3* 101 ± 9* 65 ± 8* 0.92 (0.72, 1.33)*
Overweight
Obese
98 297 11.1 ± 2.0* 10.3 ± 1.9* 50.0 63.0* 52.9 64.9 22.4 ± 2.2* 25.5 ± 2.8* 73.3 ± 6.9* 80.6 ± 8.9* 26.9 ± 5.4* 31.4 ± 6.6* 104 ± 8 109 ± 8 66 ± 7* 69 ± 6* 1.28 (0.86, 1.59)* 1.33 (0.95, 1.70)*
o0.001 o0.001 NS o0.001 o0.001 o0.001 o0.001 o0.001 o0.001
1.52 ± 0.32* 2.45 ± 0.63 4.14 ± 0.67 5.08 ± 0.37 95.8*
1.42 ± 0.23*,Δ 2.53 ± 0.58* 4.10 ± 0.67 5.10 ± 0.36Δ 91.5Δ
1.55 ± 0.36* 2.23 ± 0.73 4.14 ± 0.8 5.67 ± 0.49* 88.1
1.34 ± 0.30* 2.41 ± 0.62 4.08 ± 0.66 5.60 ± 0.52* 85.1
1.31 ± 0.28* 2.51 ± 0.61* 4.21 ± 0.72 5.58 ± 0.99* 82.7
o0.001 o0.001 NS o0.001 0.021
59.3 94.0
63.4 90.3
59.5 93.4
57.1 95.4
54.6 93.8
56.4 93.3
NS NS
20.4
27.6
26.2
12.7*
19.4
15.5
0.003
Abbreviations: BMI, body mass index; DBP, diastolic blood pressure; FMP, fat mass percentage; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; MetS, metabolic syndrome; MHO, metabolically healthy obesity; MUNW, metabolically unhealthy normal weight; NS, not significant; SBP, systolic blood pressure; TC, total cholesterol; TG, triglyceride; WC, waist circumference. Normally distributed data are expressed as mean ± s.d. Median and interquartile range were used for skewed variables. *Po0.05 for post hoc pairwise comparison with metabolically healthy normal weight; ΔPo 0.05 for comparison between MHO and metabolically unhealthy obese. aPrepuberty: Tanner stage 1. bSkewed distributions were logarithmically transformed for statistical tests. cOver one time biweekly and over 30 min per time. dHigh fat food: consumption of at least one meat, fried food or western fast food. ePresence of at least one parent with hypertension.
Journal of Human Hypertension (2015), 1 – 7
© 2015 Macmillan Publishers Limited
Metabolically healthy obesity and hypertension WQ Ding et al
3 Statistical analyses All data are expressed as mean ± s.d., median (25th–75th percentiles) or as proportions. TG levels were not normally distributed, and their logtransformed values were analyzed. Analyses of continuous and categorical variables to assess differences in the baseline characteristics between the categories of BMI and metabolic status were determined using analysis of variance and a Bonferroni post hoc comparison test or the chi-squared test. The crude cumulative hypertension incidence was calculated as the number of subjects with incident hypertension/the number of subjects at risk of hypertension. Risk ratios (RRs) and 95% confidence intervals (CIs) from multivariate logistic regression models were used to estimate the relative risk for cumulative hypertension incidence with the MHO and MUNW phenotypes and were adjusted for confounders including age, sex, pubertal stage, physical activity, dietary habits and family history of hypertension. Statistical analyses were performed using SPSS13.0 (SPSS Inc., Chicago, Illinois, USA) with statistical significance set at Po0.05.
RESULTS Baseline characteristics of the participants The subjects (1183) had an average age of 10.0 ± 0.1 years, and 54.3% (642) were boys. As defined by the MetS components, the MHO phenotype was identified in 36.7% of the obese individuals, and the MUNW phenotype was identified in 41.9% of the normalweight individuals. Participant characteristics by BMI categories and metabolic status are presented in Table 1. Individuals with
Table 2.
MHO or MUNW had worse metabolic profiles, including BMI, WC, FMP, DBP, TG and HDL cholesterol, compared with the metabolically healthy normal-weight individuals. Compared with the metabolically unhealthy obese individuals, the MHO individuals had lower BMI, WC, FMP, SBP, TG and fasting glucose with higher HDL cholesterol and levels of physical activity. Participant characteristics by BMI categories and metabolic status, as defined by IR, are shown in Table 2. MHO phenotype was identified in 3.9% of the obese individuals, and MUNW phenotype was identified in 65.6% of the normal-weight individuals. Compared with metabolically healthy normal-weight individuals, the MHO and MUNW individuals had higher BMI and WC. Hypertension incidence During the 6-year follow-up, 239 (20.2%) participants developed incident hypertension. The hypertension incidence rates were affected by the presence of obesity more than by the presence of IR. The cumulative 6-year hypertension incidence was higher with higher BMI categories, irrespective of the presence of IR. MHO individuals had the highest incidence of hypertension (50%, n = 5). Similar results were observed across the groups in terms of BMI and metabolic status as defined by MetS components. The cumulative incidence rate of hypertension was 36% (n = 62) in MHO individuals (Figure 1).
Baseline characteristics of subjects by BMI and metabolic status as defined by IR
Variables
Metabolically healthy Normal weight
Overweight
P-value
Metabolically unhealthy Obese (MHO)
Normal weight (MUNW)
Overweight
Obese
n 93 11 10 178 80 244 Age (years) 8.3 ± 1.6 9.3 ± 1.6 8.8 ± 1.5 9.1 ± 1.5 9.2 ± 1.5* 9.0 ± 1.5* Sex (boys, %) 55.9 18.2 30.0 59.0* 55.0 37.7 81.5 72.7 100.0 59.4* 58.7* 65.7* Pubertal stage a (prepuberty, %) BMI (kg m − 2) 15.4 ± 1.4 20.4 ± 2.0* 23.3 ± 2.6* 16.4 ± 1.7* 20.4 ± 1.5* 23.9 ± 2.2* WC (cm) 54.5 ± 4.7 69.5 ± 6.8* 76.5 ± 8.8* 57.7 ± 5.3* 68.4 ± 7.6* 76.8 ± 8.1* FMP (%) 14.2 ± 3.7 24.2 ± 5.2* 31.1 ± 7.3* 15.8 ± 4.2 24.8 ± 4.3* 29.3 ± 5.4* SBP (mm Hg) 93 ± 9 98 ± 10 98 ± 9 95.9 ± 9.9 101 ± 9* 105 ± 7* DBP (mm Hg) 59 ± 7 61 ± 7 61 ± 6 59.2 ± 8.6 64 ± 8* 67 ± 6* 0.65 (0.50, 0.85) 0.79 (0.56, 0.98) 0.84 (0.54, 1.06) 0.73 (0.57, 0.98) 0.96 (0.72, 1.32)* 1.04 (0.74, 1.37)* TG (mmol l − 1)b 1.63 ± 0.34 1.59 ± 0.31 1.39 ± 0.24 1.57 ± 0.31 1.36 ± 0.33* 1.34 ± 0.27* HDL-C (mmol l − 1) 2.48 ± 0.66 2.90 ± 0.53 2.90 ± 0.53 2.53 ± 0.94 2.59 ± 0.63 2.62 ± 0.60 LDL-C (mmol l − 1) −1 4.14 ± 0.72 4.59 ± 0.55 4.36 ± 0.61 4.15 ± 1.00 4.12 ± 0.69 4.13 ± 0.69 TC (mmol l ) 4.78 ± 0.42 4.82 ± 0.50 4.83 ± 0.66 5.04 ± 0.45* 5.09 ± 0.46* 5.10 ± 0.39* Fasting glucose (mmol l − 1) 1.91 (1.48, 2.53) 2.48 (2.03, 2.69) 2.03 (1.31, 2.51)Δ 5.56 (4.02, 7.07) 7.99 (5.63, 10.50)* 9.92 (6.80, 13.22)* Fasting insulinb b Δ 0.40 (0.31, 0.55) 0.55 (0.39, 0.63) 0.44 (0.24, 0.57) 1.26 (0.89, 1.71) 1.78 (1.20, 2.42)* 2.18 (1.53, 2.92)* HOMA-IR 89.9 90.0 88.9 86.7 91.0 91.2 Physical activity (%)c Frequent dietary consumption Fruits and vegetables (every day) High fat foods/ sweetened beverages (⩾2 times/per 2 weeks)d Family history of hypertension (%)e
o0.001 o0.001 0.001 o0.001 o0.001 o0.001 o0.001 o0.001 o0.001 o0.001 o0.001 NS NS o0.001 o0.001 NS
53.3
63.6
90.0
56.6
55.8
53.1
NS
88.9
90.9
100
93.0
92.3
91.6
NS
26.9
27.3
40.0
35.4
45.0
35.7
NS
Abbreviations: BMI, body mass index; DBP, diastolic blood pressure; FMP, fat mass percentage; HDL-C, high-density lipoprotein cholesterol; HOMA-IR, homeostatic model assessment of insulin resistance; LDL-C, low-density lipoprotein cholesterol; MetS, metabolic syndrome; MHO, metabolically healthy obesity; MUNW, metabolically unhealthy normal weight; NS, not significant; SBP, systolic blood pressure; TC, total cholesterol; TG, triglyceride; WC, waist circumference. Of 1183 volunteers, 616 participants had their insulin measured. Normally distributed data are expressed as mean ± s.d. Median and interquartile range were used for skewed variables. *Po0.05 for post hoc pairwise comparison with metabolically healthy normal weight. ΔPo 0.05 for comparison between MHO and metabolically unhealthy obese. aPrepuberty: Tanner stage 1. bSkewed distributions were logarithmically transformed for statistical tests. cOver one time biweekly and over 30 min per time. dHigh fat food: consumption of at least one meat, fried food or western fast food. ePresence of at least one parent with hypertension.
© 2015 Macmillan Publishers Limited
Journal of Human Hypertension (2015), 1 – 7
Metabolically healthy obesity and hypertension WQ Ding et al
4
Figure 1. Hypertension incidence rates for different categories of BMI and metabolic status. (a) Metabolic status was defined by metabolic syndrome components. (b) Metabolic status was defined by insulin resistance. BMI, body mass index; MHO, metabolically healthy obesity; MUNW, metabolically unhealthy normal weight.
We analyzed the risk of incident hypertension caused by BMI and metabolic status after controlling for age, sex, pubertal stage, physical activity, dietary habits and family history of hypertension. For metabolic status defined by MetS components, the MHO individuals had an increased hypertension incidence (RR = 5.42, 95% CI 3.19–9.22). However, the MUNW phenotype was not related to hypertension incidence (RR = 1.23, 95% CI 0.65–2.34). Using IR to define the metabolic status, the risk of developing hypertension was higher for the higher BMI categories in the metabolically healthy and unhealthy individuals compared with the metabolically healthy normal-weight individuals. Hypertension incidence was elevated in the MHO subjects (RR = 7.59, 95% CI 1.64–35.20), but was not elevated significantly in the MUNW subjects (RR = 1.39, 95% CI 0.57–3.39; Table 3). DISCUSSION This study yielded two key findings. First, the MHO and MUNW phenotypes, defined by MetS components, were common in Chinese children and adolescents. Second, an increased risk of hypertension was observed in the MHO but not in the MUNW individuals, independent of the definition of metabolic status. Although there is no consensus on the definition of the MHO phenotype, a few studies have reported that 6.3–31.5% of overweight or obese children and adolescents are classified as metabolically healthy.22–24 Because different definitions of Journal of Human Hypertension (2015), 1 – 7
metabolic abnormalities are used in these studies, the prevalence is not easily compared among studies. Nonetheless, our MHO phenotype defined by MetS and IR prevalence of 36.7% and 3.9% among obese individuals is approximately in agreement with the above studies. A cohort study performed in 40- to 69-year-old adults reported that the hypertension incidence rates were influenced by the presence of obesity more than by the presence of metabolic status, and metabolically unhealthy obese individuals have the highest incidence of hypertension.5 Similarly, our study indicates that obese individuals are at a higher incidence of hypertension before adjustment for potential confounding factors, regardless of the metabolic status. The MHO individuals had a much higher cumulative incidence of hypertension than the corresponding metabolically unhealthy obese individuals. This finding should be interpreted with caution because the MHO group had fewer individuals and events, and therefore, the findings had lower precision. In a 17.4-year follow-up study, Hinnouho et al.25 found that MHO individuals had higher risk of incident CVD compared with metabolically healthy normal-weight individuals, regardless of the metabolic health definition used, with the exception of HOMA. Similarly, a prospective study of Koreans reported that MHO phenotype had a significant impact on the development of hypertension.7 A more recent meta-analysis of two communitybased cohort studies indicated that the absence of IR and other © 2015 Macmillan Publishers Limited
Metabolically healthy obesity and hypertension WQ Ding et al
5 Table 3.
Incident hypertension by BMI and metabolic status as defined by MetS components or IR Number at risk
Number with incident hypertension
Model 1
Model 2
RR
95% CI
RR
95% CI
Defined by MetS components Metabolically healthy normal weight Metabolically healthy overweight MHO MUNW Metabolically unhealthy overweight Metabolically unhealthy obese
294 76 172 212 98 297
24 13 62 21 21 91
1.00 (Reference) 2.32 6.34 1.24 3.07 4.97
1.12–4.81 3.76–10.67 0.66–2.29 1.62–5.81 3.06–8.07
1.00 (Reference) 2.25 5.42 1.23 3.11 4.65
1.07–4.74 3.19–9.22 0.65–2.34 1.59–6.07 2.82–7.66
Defined by IR Metabolically healthy normal weight Metabolically healthy overweight MHO MUNW Metabolically unhealthy overweight Metabolically unhealthy obese
93 11 10 178 80 244
8 2 5 18 19 83
1.00 (Reference) 2.36 10.62 1.19 3.31 5.47
0.43–12.86 2.53–44.66 0.49–2.86 1.36–8.05 2.53–11.85
1.00 (Reference) 2.46 7.59 1.39 3.99 5.14
0.44–13.89 1.64–35.20 0.57–3.39 1.59–10.00 2.33–11.32
Abbreviations: BMI, body mass index; CI, confidence interval; IR, insulin resistance; MetS, metabolic syndrome; MHO, metabolically healthy obesity; MUNW, metabolically unhealthy normal weight; RR, relative risk. Model 1: unadjusted for confounding factors Model 2: adjusted for age, sex, pubertal stage, physical activity, dietary habits and family history of hypertension.
metabolic derangements in overweight or obese individuals provided no protection against the development of hypertension.26 In line with these previous studies, our findings indicated that the MHO phenotype was positively associated with the development of hypertension and that the role of obesity was greater than that of metabolic abnormalities in the development of hypertension. However, several cohort studies have reported that MHO individuals were not at an increased risk for CVD, regardless of the use an MHO definition based on a combination of two or more metabolic risk factors.27,28 The mechanisms underlying the medical consequences of the MHO phenotype are uncertain, and limited data are available. It has been reported that dietary and lifestyle factors play important roles in the development of IR and can lead to an increased risk of associated cardiometabolic abnormalities.29 Compared with metabolically unhealthy obese individuals, MHO individuals have lower WC and moderate-tohigh levels of physical activity, despite having similar BMI values.30 Phillips et al. recently demonstrated that dietary macronutrient composition, dietary quality assessed using the DASH (Dietary Approaches to Stop Hypertension) score and physical activity were similar between the metabolically healthy and unhealthy individuals regardless of BMI. Higher dietary quality was positively associated with metabolic health in obese and non-obese individuals. In addition, a moderate-to-high level of physical activity was positively associated with the MHO phenotype, as defined by IR.31 In line with the above studies, we also found that the dietary consumption patterns (including fruits and vegetables and high fat foods/sweetened beverages) did not differ significantly between the metabolically healthy and unhealthy participants regardless of BMI. However, in the present study, although the MHO individuals had lower BMI, lower WC, FMP, SBP, TG, fasting glucose levels, higher HDL cholesterol, better dietary habits and physical activity levels than their metabolically unhealthy obese counterparts, the MHO individuals still showed an increased risk of hypertension. These discrepancies between the present study and previous studies might be explained by the fact that there is a long-term effect of metabolic abnormalities on the risk of incident hypertension. The results of a longitudinal study are relatively more generalizable than the results of a cross-sectional study. This © 2015 Macmillan Publishers Limited
is supported by a prospective cohort study that found that an increased risk for incident hypertension was not apparent after 4 years of follow-up, although an increased risk was apparent after 8 years.32 Our study had a follow-up period of 6 years. Another difference between the studies is that metabolic health was defined in different ways.33 In recent years, the MUNW phenotype has received much attention. The results of the Third National Health and Nutrition Examination Survey (NHANES III) in the USA demonstrated that 0.1% of adolescents were normal-weight individuals with MetS.23 A recent meta-analysis showed that all of the unhealthy metabolic status phenotypes presented an increased risk for CVD, regardless of whether the subjects were normal weight, overweight or obese.10 An association between IR and the development of hypertension has been demonstrated in normal-weight individuals.34 However, in the present study, relative to the metabolically healthy normal-weight participants, a higher risk of incident hypertension was not observed in the MUNW participants. Our results are supported by Lytsy et al. who reported that normal weight with IR is not related to the development of hypertension.30 Our study has several limitations. First, because Chinese criteria were used to define hypertension, the results are not applicable to other ethnicities. Second, we did not assess some risk factors of hypertension, such as dietary salt intake. Third, the small subgroups made the study underpowered for comparing several phenotype groups. Despite these limitations, there are several strengths to this study. We recruited children and adolescents, in which incident hypertension has not been examined previously with regard to the MHO and MUNW phenotypes using prospective cohorts. Furthermore, because of the lack of a uniform definition of the MUNW and MHO phenotypes, we defined metabolic status in two ways. The MHO phenotype, as defined by MetS components in obese children and adolescents, is common. These findings add to the evidence demonstrating that MHO individuals have an increased risk of hypertension and MUNW individuals are not associated with a higher incidence of hypertension. The potential benefits of differentiating the MHO and MUNW phenotypes in clinical practice in children and adolescents appear limited. Journal of Human Hypertension (2015), 1 – 7
Metabolically healthy obesity and hypertension WQ Ding et al
6 To date, there is no standard definition of metabolic health. Different inclusion criteria, including IR, MetS components, inflammation and multiple cardiometabolic health features have been used to discriminate metabolically healthy from metabolically unhealthy individuals.35 Thus, it is difficult to make comparisons among studies. Given the lack of comprehensiveness regarding the use of a single metabolic marker, combined metabolic markers should be recommended for use in future studies. More prospective and intervention studies are needed to clarify which abnormal metabolic profile is a better definition of metabolic health in children and adolescents. To date, it is difficult to use HOMA-IR to define the metabolic health widely because the methodology to measure insulin has not yet been standardized. In addition, it has been demonstrated that inflammation is secondary to obesity and/or IR in humans. Therefore, the definition of metabolic health based on five components of the MetS, including BP, fasting plasma glucose, TG levels, HDL cholesterol and WC, would be feasible for use in large population-based studies that would help to understand the phenomenon of MHO more comprehensively in future. This also would allow consistent comparison among diverse populations. Moreover, such a definition may represent a first step toward a more detailed definition, including other metabolic markers such as IR and inflammation.35 In addition, further studies exploring the potential mechanisms underlying metabolically healthy obesity and hypertension are warranted. What is known about the topic: ● The MHO phenotype begins in childhood and continues into adulthood. Observations on the relationship between this obesity phenotype and the risk of cardiovascular events have been made in adult populations. ● The association between the MHO phenotype and the risk of developing hypertension is controversial. ● No longitudinal studies have addressed the effects of the MHO and MUNW phenotypes on the risk of developing hypertension in children and adolescents. What this study adds ● The MHO phenotype, as defined by metabolic syndrome components in obese children and adolescents, was common in children and adolescents. ● After adjusting for age, sex, physical activity, pubertal stage, dietary consumption patterns and family history of hypertension, an increased risk for hypertension was observed in the MHO individuals. Independent of the definition of metabolic abnormality, the MUNW subjects did not have an elevated incidence of hypertension. ● The results suggest that the potential benefits of differentiating the MHO and MUNW phenotypes in children and adolescents in clinical practice appear limited.
CONFLICT OF INTEREST The authors declare no conflicts of interest.
ACKNOWLEDGEMENTS This study was supported by a National Natural Science Foundation of China grant (81172746), the Beijing Municipal and Technology Commission program (D111100000611002), and the National Science and Technology program funded by the Science and Technology Committee of Beijing (2012BAI03B00).
REFERENCES 1 Singh GK, Kogan MD, van Dyck PC. Changes in state-specific childhood obesity and overweight prevalence in the United States from 2003 to 2007. Arch Pediatr Adolesc Med 2010; 164: 598–607.
Journal of Human Hypertension (2015), 1 – 7
2 Wang Y, Monteiro C, Popkin BM. Trends of obesity and underweight in older children and adolescents in the United States, Brazil, China, and Russia. Am J Clin Nutr 2002; 75: 971–977. 3 Song Y, Wang HJ, Ma J, Wang Z. Secular trends of obesity prevalence in urban Chinese children from 1985 to 2010: gender disparity. PLoS ONE 2013; 8: e53069. 4 Genelhu VA, Celoria BM, Duarte SF, Cabello PH, Francischetti EA. Not all obese subjects of multiethnic origin are at similar risk for developing hypertension and type 2 diabetes. Eur J Intern Med 2009; 20: 289–295. 5 Lee SH, Ha HS, Park YJ, Lee JH, Yim HW, Yoon KH et al. Identifying metabolically obese but normal-weight (MONW) individuals in a nondiabetic Korean population: the Chungju Metabolic disease Cohort (CMC) study. Clin Endocrinol (Oxf) 2011; 75: 475–481. 6 Brochu M, Tchernof A, Dionne IJ, Sites CK, Eltabbakh GH, Sims EA et al. What are the physical characteristics associated with a normal metabolic profile despite a high level of obesity in postmenopausal women?. J Clin Endocrinol Metab 2001; 86: 1020–1025. 7 Lee SK, Kim SH, Cho GY, Baik I, Lim HE, Park CG et al. Obesity phenotype and incident hypertension: a prospective community-based cohort study. J Hypertens 2013; 31: 145–151. 8 Karelis AD, Messier V, Brochu M, Rabasa-Lhoret R. Metabolically healthy but obese women: effect of an energy-restricted diet. Diabetologia 2008; 51: 1752–1754. 9 Arsenault BJ, Cote M, Cartier A, Lemieux I, Despres JP, Ross R et al. Effect of exercise training on cardiometabolic risk markers among sedentary, but metabolically healthy overweight or obese post-menopausal women with elevated blood pressure. Atherosclerosis 2009; 207: 530–533. 10 Kramer CK, Zinman B, Retnakaran R. Are metabolically healthy overweight and obesity benign conditions? A systematic review and meta-analysis. Ann Intern Med 2013; 159: 758–769. 11 Li S, Chen W, Srinivasan SR, Xu J, Berenson GS. Relation of childhood obesity/ cardiometabolic phenotypes to adult cardiometabolic profile: the Bogalusa Heart Study. Am J Epidemiol 2012; 176(Suppl 7): S142–S149. 12 Xi B, Liang Y, Mi J. Hypertension trends in Chinese children in the national surveys, 1993 to 2009. Int J Cardiol 2013; 165: 577–579. 13 Shan XY, Xi B, Cheng H, Hou DQ, Wang Y, Mi J. Prevalence and behavioral risk factors of overweight and obesity among children aged 2-18 in Beijing, China. Int J Pediatr Obes 2010; 5: 383–389. 14 Chen FF, Mi J, Wang TY, Li H, Hou DQ, Cheng H et al. Relationship between precosious puberty and obese status in Beijing children and adolescents. Chin J Evid Based Pediatr 2007; 12: 14–20. 15 Li M, Wu C, Song A, Zhang K. Development and preliminary application of enzyme-linked immunosorbent assay for human net insulin in serum. Chin J Endocrinol Metab 1997; 13: 214–217. 16 Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28: 412–419. 17 Ji CY. Working Group on Obesity in China Report on childhood obesity in China (1)—body mass index reference for screening overweight and obesity in Chinese school-age children. Biomed Environ Sci 2005; 18: 390–400. 18 Wan NJ, Mi J, Wang TY, Duan JL, Li M, Gong CX et al. Metabolic syndrome in overweight and obese schoolchildren in Beijing. Zhonghua Er Ke Za Zhi 2007; 45: 417–421. 19 Yin J, Li M, Xu L, Wang Y, Cheng H, Zhao X et al. Insulin resistance determined by homeostasis model assessment (HOMA) and associations with metabolic syndrome among Chinese children and teenagers. Diabetol Metab Syndr 2013; 5: 71. 20 Lawes CM, Vander Hoorn S, Law MR, Elliott P, MacMahon S, Rodgers A. Blood pressure and the global burden of disease 2000. Part 1: estimates of blood pressure levels. J Hypertens 2006; 24: 413–422. 21 Mi J, Wang TY, Meng LH, Zhu GJ, Han SM, Zhong Y et al. Development of blood pressure standards for Chinese childen and adolescents. Chin J Evid Based Pediatr 2010; 5: 4–14. 22 Prince RL, Kuk JL, Ambler KA, Dhaliwal J, Ball GD. Predictors of metabolically healthy obesity in children. Diabetes Care 2014; 37: 1462–1468. 23 Cook S, Weitzman M, Auinger P, Nguyen M, Dietz WH. Prevalence of a metabolic syndrome phenotype in adolescents: findings from the third National Health and Nutrition Examination Survey, 1988-1994. Arch Pediatr Adolesc Med 2003; 157: 821–827. 24 Bokor S, Frelut ML, Vania A, Hadjiathanasiou CG, Anastasakou M, Malecka-Tendera E et al. Prevalence of metabolic syndrome in European obese children. Int J Pediatr Obes 2008; 3(Suppl 2): 3–8. 25 Hinnouho GM, Czernichow S, Dugravot A, Nabi H, Brunner EJ, Kivimaki M et al. Metabolically healthy obesity and the risk of cardiovascular disease and type 2 diabetes: the Whitehall II cohort study. Eur Heart J (e-pub ahead of print 26 March 2014; doi:10.1093/eurheartj/ehu123).
© 2015 Macmillan Publishers Limited
Metabolically healthy obesity and hypertension WQ Ding et al
7 26 Lytsy P, Ingelsson E, Lind L, Arnlov J, Sundstrom J. Interplay of overweight and insulin resistance on hypertension development. J Hypertens 2014; 32: 834–839. 27 Hamer M, Stamatakis E. Metabolically healthy obesity and risk of all-cause and cardiovascular disease mortality. J Clin Endocrinol Metab 2012; 97: 2482–2488. 28 Calori G, Lattuada G, Piemonti L, Garancini MP, Ragogna F, Villa M et al. Prevalence, metabolic features, and prognosis of metabolically healthy obese Italian individuals: the Cremona Study. Diabetes Care 2011; 34: 210–215. 29 Hotamisligil GS. Inflammation and metabolic disorders. Nature 2006; 444: 860–867. 30 Appleton SL, Seaborn CJ, Visvanathan R, Hill CL, Gill TK, Taylor AW et al. Diabetes and cardiovascular disease outcomes in the metabolically healthy obese phenotype: a cohort study. Diabetes Care 2013; 36: 2388–2394. 31 Phillips CM, Dillon C, Harrington JM, McCarthy VJ, Kearney PM, Fitzgerald AP et al. Defining metabolically healthy obesity: role of dietary and lifestyle factors. PLoS ONE 2013; 8: e76188.
© 2015 Macmillan Publishers Limited
32 Arsenault BJ, Cote M, Cartier A, Lemieux I, Despres JP, Ross R. Effect of exercise training on cardiometabolic risk markers among sedentary, but metabolically healthy overweight or obese post-menopausal women with elevated blood pressure. Atherosclerosis 2009; 207: 530–533. 33 Hinnouho GM, Czernichow S, Dugravot A, Batty GD, Kivimaki M, Singh-Manoux A. Metabolically healthy obesity and risk of mortality: does the definition of metabolic health matter? Diabetes Care 2013; 36: 2294–2300. 34 Arnlov J, Pencina MJ, Nam BH, Meigs JB, Fox CS, Levy D et al. Relations of insulin sensitivity to longitudinal blood pressure tracking: variations with baseline age, body mass index, and blood pressure. Circulation 2005; 112: 1719–1727. 35 Rey-Lopez JP, de Rezende LF, Pastor-Valero M, Tess BH. The prevalence of metabolically healthy obesity: a systematic review and critical evaluation of the definitions used. Obes Rev 2014; 15: 781–790.
Journal of Human Hypertension (2015), 1 – 7
