Independent Impact of Body Mass Index and Metabolic Syndrome on the Risk of Type 2 Diabetes in Koreans

Abstract

Background:

Although obesity and metabolic syndrome have been associated with the risk of type 2 diabetes mellitus (T2DM), it is unclear whether obese or overweight people without metabolic syndrome are at increased risk for T2DM.

Methods:

Clinical and laboratory data were assessed in 8,748 subjects without diabetes (5,707 men, 3,041 women; age 20–79 years) who underwent voluntary medical check-ups at a 5-year interval. The subjects were categorized by body mass index (BMI) and metabolic syndrome status at baseline, and the incidence of diabetes over 5 years was assessed.

Results:

Of the 8,748 subjects, 308 (3.5%) developed T2DM over 5 years. Compared with normal weight (BMI <25.0 kg/m²) individuals without metabolic syndrome, the adjusted odds ratios (ORs) were 1.61 (1.13–2.29) and 4.93 (1.90–12.79) for overweight (BMI 25.0–29.9 kg/m²) and obese (BMI ≥30.0 kg/m²) individuals without metabolic syndrome, respectively, and 6.94 (5.08–9.47) and 10.61 (5.59–20.14) for overweight and obese individuals with metabolic syndrome, respectively. Using the lower BMI cutoff points for Asian populations, compared with subjects with BMI <23 kg/m² without metabolic syndrome, the adjusted ORs for subjects with BMI 23–27.4 kg/m² and BMI ≥27.5 kg/m² without metabolic syndrome were 2.64 (1.74–4.00) and 4.31 (2.36–7.86), respectively, and 10.11 (6.53–15.67) and 16.69 (10.40–26.77), respectively, for those with metabolic syndrome.

Conclusions:

Overweight/obesity and metabolic syndrome both are significant risk factors for development of T2DM in Koreans, and overweight or obesity without metabolic syndrome should not be considered a harmless condition. The lower BMI cutoffs for Asian populations can be useful in predicting risk of T2DM in Koreans.

Introduction

T he association between obesity and type 2 diabetes mellitus (T2DM) has been suggested to be mediated by the metabolic syndrome, a cluster of risk factors associated with insulin resistance.^1,2 On the other hand, the existence of a subgroup of obese individuals who do not show traditional risk factors associated with insulin resistance syndrome, despite having a high accumulation of body fat, has been reported. These “metabolically healthy but obese (MHO)” people have normal insulin sensitivity, no sign of hypertension, normal lipid levels, and a favorable inflammation profile,^3,4 and do not show increased risk for diabetes and cardiovascular disease or mortality.^1,5 Thus, there are claims that obesity without metabolic syndrome may be innocuous. In contrast, others have reported that overweight or obese people without metabolic syndrome were at increased risk for cardiovascular disease or diabetes and that overweight or obesity should not be considered a harmless condition.^6
–8

The associations of body mass index (BMI) with body composition and health outcomes may differ between Asian and European populations.⁹ For example, some studies have suggested that Asian populations have an elevated risk of T2DM at a relatively lower BMI.^9,10 However, this is still controversial. A previous study in middle-aged Koreans¹¹ reported that although obesity-related health risks begin to increase at much lower BMI levels than the existing cutoff point (25 kg/m²), the relative risks at a given point of BMI were not greater than those that have been reported for Caucasians. Moreover, longitudinal studies examining the impact of different BMI/metabolic syndrome combinations on the risk of T2DM in Asian populations are still lacking. Therefore, we investigated the associations between combinations of BMI and metabolic syndrome categories and risk of T2DM in Koreans.

Subjects and Methods

Subjects

We retrospectively assessed the clinical and laboratory data of 9,432 subjects (6,260 men and 3,172 women; ages 20–79 years) who visited the Health Promotion Center at Asan Medical Center (Seoul, Korea) for medical check-ups in 2005 and underwent follow-up examinations in 2010. Subjects with a self-reported history of physician-diagnosed diabetes mellitus or those who taking antihyperglycemic medications (n=332), fasting blood glucose concentrations ≥126 mg/dL (n=220), or glycosylated hemoglobin (HbA1c) ≥6.5% (n=132) at baseline were excluded. Data for the remaining 8,748 subjects (5,707 men and 3,041 women) were analyzed.

This study was approved by the Institutional Review Board (IRB) of the Asan Medical Center, and all participants provided written informed consent.

Measurements

Height and weight were measured with subjects wearing light clothing without shoes. BMI was calculated as weight in kilograms divided by the square of height in meters. Waist circumference was measured midway between the costal margin and the iliac crest at the end of a normal expiration. Blood pressure was measured with a mercury sphygmomanometer on the right arm with subjects in a sitting position after a 5-min rest. We used average of two measurements at a 5-min interval. Blood samples were obtained in the morning after an overnight fast. Plasma glucose was measured by the hexokinase method using an autoanalyzer (Toshiba, Tokyo, Japan). Total cholesterol, high-density lipoprotein cholesterol (HDL-C), and triglycerides were also measured using an autoanalyzer (Toshiba).

At the baseline visit, each subject filled out a questionnaire on alcohol intake (average frequency per week), smoking habits (never, current, or former smoker), usual pattern of physical activity (average time per week and intensity as mild, moderate, or vigorous), medical history, family history of diabetes, and regular medications. Physical activity was analyzed as a three-category variable. “Sedentary” was defined as less than 3 h per week of light activity without sweating or dyspnea; “moderate” was defined as 3 h or more of light activity or 1–2 h of vigorous activity that caused sweating or dyspnea per week; “active” was defined as vigorous activity at least 3 h per week.

The subjects were categorized by BMI and metabolic syndrome status at baseline, and incidence of diabetes over 5 years was assessed. Diabetes was defined as fasting plasma glucose ≥126 mg/dL, HbA1c ≥6.5%, or treatment with antihyperglycemic medications. The participants who were nondiabetic at baseline and began to use the medications between 2006 and 2010 were regarded as incident diabetes.

Metabolic syndrome was defined according to the 2009 consensus criteria,¹² as subjects with any combination of three or more of the following risk factors: Fasting plasma glucose ≥100 mg/dL or on antidiabetic treatment; blood pressure ≥130/85 mmHg or on antihypertensive treatment; plasma triglycerides ≥150 mg/dL; plasma HDL-C <40 mg/dL in men and <50 mg/dL in women; and waist circumference ≥90 cm in men and ≥80 cm in women.

Statistics

Data are expressed as mean±standard deviation (SD). Variables not normally distributed, such as blood concentrations of glucose, triglycerides, and cholesterol, were log-transformed before analysis. Standardized incidence rates were calculated by indirect standardization method using the whole study population as a reference population (multiplying incidence rate of reference population by standardized incidence ratio). Multivariate logistic regression analyses were used to obtain odds ratios (ORs) for development of diabetes after adjusting for clinical and biochemical variables, such as age, gender, smoking, alcohol consumption, and physical activity. Statistical analyses were performed using the SPSS 17.0 software (SPSS Inc., Chicago, IL, USA). P values less than 0.05 were considered statistically significant.

Results

The baseline clinical characteristics of the study participants according to the different categories of BMI/metabolic syndrome are shown in Table 1. As expected, the prevalence of metabolic syndrome and cardiovascular risk factors increased with increasing BMI, and cardiovascular risk factors were markedly increased in subjects with metabolic syndrome within each BMI group. Among the 8,748 (5,707 men and 3,041 women; mean age at baseline, 47.7±8.2 years) participants who were nondiabetic at baseline, 308 (3.5%), including 242 men (4.2%) and 66 women (2.2%), developed T2DM over 5 years. The risk of diabetes was higher in overweight (BMI 25.0–29.9 kg/m²) and obese (BMI ≥30.0 kg/m²) subjects than in normal weight (BMI <25.0 kg/m²) individuals, regardless of metabolic syndrome (Table 2). Relative to normal weight individuals without metabolic syndrome, the adjusted ORs for T2DM were 1.61 [95% confidence interval (CI), 1.13–2.29] and 4.93 (1.90–12.79) in overweight and obese individuals without metabolic syndrome, respectively, and 6.94 (5.08–9.47) and 10.61 (5.59–20.14) for overweight and obese individuals with metabolic syndrome, respectively, after adjustment for baseline age, gender, smoking status, alcohol consumption, and physical activity. The association between BMI and the risk of T2DM remained significant after adjustment for metabolic syndrome (data not shown). Multiple logistic regression analysis showed no significant multiplicative interaction between BMI and metabolic syndrome on the risk of development of T2DM (P=0.143).

Table 1.

Baseline Clinical Characteristics of the Study Participants According to BMI/Metabolic Syndrome Categories

	Normal weight		Overweight		Obese
	MetS(−)	MetS(+)	MetS(−)	MetS(+)	MetS(−)	MetS(+)
Number, n (%)	5540 (93)	419 (7)	1834 (69)	811 (31)	59 (41)	85 (59)
Male/female (%)	57/43	66/34	81/19	84/16	81/19	81/19
Age (years)	47.2±8.2	50.9±8.3	48.2±7.9	49.4±8.6	46.7±7.3	45.5±8.5
BMI (kg/m²)	22.2±1.8	23.5±1.3	26.4±1.2	27.0±1.3	31.5±1.9	31.6±1.6
Systolic blood pressure (mmHg)	116±14	131±15	121±12	131±15	123±10	134±15
Diastolic blood pressure (mmHg)	72±9	81±9	75±8	82±9	78±8	84±11
Waist circumference (cm)	78.0±7.1	84.3±5.8	88.1±5.5	91.6±5.1	98.8±7.7	99.5±6.7
FPG (mg/dL)	92±9	102±9	94±9	102±10	95±9	101±9
HbA1c (%)	5.4±0.3	5.6±0.4	5.5±0.4	5.6±0.4	5.5±0.3	5.7±0.3
Cholesterol (mg/dL)	188±32	198±33	196±31	198±33	202±40	204±29
Triglyceride (mg/dL)	106±59	202±86	128±66	205±97	134±54	239±135
HDL-cholesterol (mg/dL)	58±13	46±11	52±11	45±10	50±10	44±9
Smoking (%)
Current	22.1	27.7	27.6	31.0	42.0	35.7
Ex-smoker	26.5	35.6	43.0	40.2	34.0	30.0
Never	51.4	36.7	29.4	28.8	24.0	34.3
Alcohol (%)
≤1 ×/week	61.0	51.1	43.8	41.9	32.1	37.5
2–4 ×/week	28.0	29.9	39.6	35.6	43.4	37.5
≥5 ×/week	11.0	19.0	16.6	22.5	24.5	25.0
Physical activity (%)
Sedentary	36.1	32.4	29.0	29.0	33.3	35.9
Moderate	26.9	28.3	29.0	35.1	29.4	34.4
Active	37.0	39.2	42.1	35.9	37.3	29.7

Data are mean±standard deviation (SD) or percent (%).

Normal weight, BMI <25.0 kg/m²; overweight, BMI 25.0–29.9 kg/m²; obese, BMI ≥30.0 kg/m².

BMI, body mass index; MetS, metabolic syndrome; FPG, fasting plasma glucose; HbA1c, glycosylated hemoglobin; HDL, high-density lipoprotein.

Table 2.

Odds Ratio for Development of Type 2 Diabetes over 5 Years According to BMI/Metabolic Syndrome Categories

	Normal weight		Overweight		Obese
	MetS(−)	MetS(+)	MetS(−)	MetS(+)	MetS(−)	MetS(+)
Number at risk	5540	419	1834	811	59	85
Number of diabetes mellitus cases	95	41	54	100	5	13
Diabetes mellitus incidence (%/5 years)
Crude	1.7	9.8	2.9	12.3	8.5	15.3
Standardized for age, BMI	2.1	8.4	3.4	12.6	9.7	16.8
Crude OR	Referent	6.22	1.74	8.06	5.31	10.35
		(4.25–9.10)	(1.24–2.44)	(6.02–10.79)	(2.08–13.56)	(5.54–19.32)
Adjusted OR^a	Referent	5.40	1.61	6.94	4.93	10.61
		(3.62–8.08)	(1.13–2.29)	(5.08–9.47)	(1.90–12.79)	(5.59–20.14)

Normal weight, BMI <25.0 kg/m²; overweight, BMI 25.0–29.9 kg/m²; obese, BMI ≥30.0 kg/m².

Adjusted for age, gender, smoking, alcohol consumption, and physical activity.

BMI, body mass index; MetS, metabolic syndrome; OR, odds ratio.

Because Asian populations have been reported to have different associations between BMI and health risks than European populations, we reanalyzed the data according to the BMI cutoff points for Asian populations proposed by the World Health Organization (WHO).⁹ Compared with subjects having a BMI <23 kg/m², those with BMI 23–27.4 kg/m² and BMI ≥27.5 kg/m², with or without metabolic syndrome, were at higher risk of developing diabetes (Table 3). The adjusted ORs for each BMI category, with or without metabolic syndrome, were similar to or slightly higher than those obtained using the BMI criteria for European populations.

Table 3.

Odds Ratio for Development of Type 2 Diabetes over 5 Years According to BMI Categories for Asian Population Recommended by the WHO

	BMI <23.0 kg/m ²		BMI 23.0–27.4 kg/m ²		BMI ≥27.5 kg/m ²
	MetS(−)	MetS(+)	MetS(−)	MetS(+)	MetS(−)	MetS(+)
Number at risk	3300	122	3702	830	431	363
Number of diabetes mellitus cases	33	9	102	89	19	56
Diabetes mellitus incidence (%/5 years)
Crude	1.0	7.4	2.8	10.7	4.4	15.4
Standardized for age, BMI	1.2	7.0	3.7	12.0	3.8	14.2
Crude OR	Referent	7.88	2.81	11.89	4.57	18.06
		(3.68–16.87)	(1.89–4.16)	(7.91–17.87)	(2.57–8.10)	(11.56–28.20)
Adjusted OR^a	Referent	7.34	2.64	10.11	4.31	16.69
		(3.38–15.93)	(1.74–4.00)	(6.53– 15.67)	(2.36– 7.86)	(10.40–26.77)

Adjusted for age, gender, smoking, alcohol consumption, and physical activity.

BMI, body mass index (kg/m²); WHO, World Health Organization; MetS, metabolic syndrome; OR, odds ratio.

Discussion

We have shown here that increased BMI and metabolic syndrome individually affect the risk of developing T2DM in Koreans, as well as providing further evidence that overweight and obesity without metabolic syndrome are not harmless conditions. Moreover, our results that the ORs of developing diabetes in increased BMI categories defined by lower cutoff points for Asians were similar to or slightly higher than those obtained using BMI criteria for European populations suggested that the Asian criteria could be suitable for predicting the risk of T2DM in Koreans.

Although T2DM has shown separate associations with BMI and metabolic syndrome, it has been unclear whether the risk associated with elevated BMI is largely mediated by its combination with metabolic syndrome or is independent of metabolic syndrome. In the Framingham Offspring Study, overweight or obese individuals without metabolic syndrome did not show an increased risk for T2DM, whereas participants with metabolic syndrome were at substantially higher risk for diabetes regardless of BMI status.¹ In contrast, a recent study⁷ reported that overweight or obese middle-aged men without metabolic syndrome were at increased risk for diabetes, opposing the concept of “metabolically healthy obese” phenotype.^1,5 Our results support this finding, while extending it to Korean men and women across a broad age range.

Our results are also in agreement with those of the Tromso Study,¹³ which showed that only 53% of incident cases of T2DM fulfilled three or more metabolic syndrome criteria at baseline, thereby the “metabolic syndrome,” used as a dichotomously defined exposure variable, missed nearly half of the cases in predicting development of T2DM. Furthermore, BMI was a strong independent predictor of T2DM in both groups with and without metabolic syndrome. These findings suggest that additional factors associated with overweight/obesity, not included in the components of metabolic syndrome, further account for the risk of diabetes. These additional factors may include chronic subclinical inflammation, endothelial dysfunction, or adipokines such as adiponectin.¹ However, the presence of metabolic risk factors associated with obesity augmented the risk for T2DM, independently of obesity status. Indeed, we found that the OR for diabetes in normal weight individuals with metabolic syndrome was greater than that in overweight or obese subjects without metabolic syndrome, underscoring the independent role of metabolic syndrome in predicting risk of diabetes. Recent studies have suggested that a more comprehensive measure of health that considers obesity-related co-morbidity and functional status may be a better predictor of obesity related risk than BMI alone.^14,15

Most studies that have evaluated the association between BMI and metabolic risks have been conducted in populations of European origin. Therefore, the definitions of overweight and obesity are based essentially on data from European populations.¹⁶ It has been suggested that the associations of BMI and health outcomes may differ between Asian and European populations.⁹ Asians have been shown to have a higher percentage of body fat than do Europeans for a given BMI, and have an elevated risk of T2DM at a relatively low BMI.⁹ Our results showed that the ORs of developing diabetes in increased BMI categories defined by lower cutoff points for Asians were similar to or slightly higher than those obtained using BMI criteria for European populations. Therefore, the lower BMI cutoff points for Asian populations can be suitable for predicting the risk of T2DM in Koreans. The ethnic difference may be related to the genetic factors, lifestyle factors including dietary habits, or body composition.

The strength of this study was its performance in a large number of well-characterized participants. Moreover, by measuring HbA1c concentration, we increased the rate of detection of undiagnosed diabetes, and waist circumference was measured in all subjects for an exact definition of metabolic syndrome. However, our study also had several limitations. First, our study participants were those who came in for voluntary medical check-ups and are therefore not representative of the general population. Second, this was a retrospective study, and we did not consider changes in body weight and metabolic parameters during the follow-up period. Third, we could not assess indices of insulin resistance such as fasting insulin level or homeostasis model assessment index. Therefore, we could not verify the general hypothesis that the deleterious metabolic features associated with obesity are largely related to the presence of insulin resistance.⁵

In conclusion, overweight/obesity and metabolic syndrome are independent risk factors for development of T2DM in Koreans. Therefore, overweight or obesity without metabolic syndrome should not be considered a harmless condition. In addition, the lower BMI cutoff points for Asian populations can be useful in predicting the risk of T2DM in Koreans.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Meigs

, Wilson

, Fox

et al. Body mass index, metabolic syndrome, and risk of type 2 diabetes or cardiovascular disease. J Clin Endocrinol Metab, 2006; 91:2906–2912.

Cameron

. The metabolic syndrome: Validity and utility of clinical definitions for cardiovascular disease and diabetes risk prediction. Maturitas, 2010; 65:117–121.

Brochu

, Tchernof

, Dionne

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.

Karelis

, Faraj

, Bastard

et al. The metabolically healthy but obese individual presents a favorable inflammation profile. J Clin Endocrinol Metab, 2005; 90:4145–4150.

Calori

, Lattuada

, Piemonti

et al. Prevalence, metabolic features, and prognosis of metabolically healthy obese Italian individuals: The Cremona Study. Diabetes Care, 2011; 34:210–215.

Arnlov

, Ingelsson

, Sundstrom

et al. Impact of body mass index and the metabolic syndrome on the risk of cardiovascular disease and death in middle-aged men. Circulation, 2010; 121:230–236.

Arnlov

, Sundstrom

, Ingelsson

et al. Impact of BMI and the metabolic syndrome on the risk of diabetes in middle-aged men. Diabetes Care, 2011; 34:61–65.

Kuk

, Ardern

. Are metabolically normal but obese individuals at lower risk for all-cause mortality? Diabetes Care, 2009; 32:2297–2299.

WHO Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet, 2004; 363:157–163.

10.

Stevens

, Truesdale

, Katz

et al. Impact of body mass index on incident hypertension and diabetes in Chinese Asians, American Whites, and American Blacks: the People's Republic of China Study and the Atherosclerosis Risk in Communities Study. Am J Epidemiol, 2008; 167:1365–1374.

11.

, Shin

, Yun

et al. Cut-off point of BMI and obesity-related comorbidities and mortality in middle-aged Koreans. Obes Res, 2004; 12:2031–2040.

12.

Alberti

, Eckel

, Grundy

et al. 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, 2009; 120:1640–1645.

13.

Joseph

, Svartberg

, Njolstad

et al. Risk factors for type 2 diabetes in groups stratified according to metabolic syndrome: A 10-year follow-up of The Tromso Study. Eur J Epidemiol, 2011; 26:117–124.

14.

Padwal

, Pajewski

, Allison

et al. Using the Edmonton obesity staging system to predict mortality in a population-representative cohort of people with overweight and obesity. CMAJ, 2011; 183:E1059–E1066.

15.

Kuk

, Ardern

, Church

et al. Edmonton Obesity Staging System: Association with weight history and mortality risk. Appl Physiol Nutr Metab, 2011; 36:570–576.

16.

Zheng

, McLerran

, Rolland

et al. Association between body-mass index and risk of death in more than 1 million Asians. N Engl J Med, 2011; 364:719–729.