Anthropometric Cutoffs for Increased Cardiometabolic Risk Among Lebanese Adults: A Cross-Sectional Study

Abstract

Background:

Obesity is associated with increased risk for metabolic syndrome (MetS). Anthropometric cutoffs derived for Caucasians may not be applicable to identify obesity in Middle Eastern populations. This study aims to (1) determine optimal cutoff values of body mass index (BMI), waist circumference (WC), and waist-to-height ratio (WHtR) for the prediction of MetS among Lebanese adults and (2) to evaluate the ability of the derived cutoffs in predicting MetS, in comparison with published cutoffs.

Methods:

A cross-sectional study involving adults aged ≥20 years (n = 305) with no history of chronic diseases was conducted. Data collection included sociodemographic characteristics, anthropometric measurements, and fasting blood samples. The International Diabetes Federation criteria were used to identify MetS. Receiver operating characteristic analyses were performed to determine optimal cutoff values. The ability of the derived cutoffs in predicting MetS was examined using multiple logistic regression analyses.

Results:

The derived cutoff values for men and women were 26.35 and 25.74 kg/m² for BMI, 94 and 83 cm for WC, and 0.54 and 0.53 for WHtR, respectively. The use of the derived cutoffs improved the prediction of MetS compared to reference published cutoffs. In men, abdominal adiposity indicators performed better than BMI in predicting MetS, while in women, BMI, WC, and WHtR were all strong predictors.

Conclusion:

The study identified, for three anthropometric indices, the optimal cutoff values that identify MetS among Lebanese adults, hence responding to the need for ethnic-and population-specific cutoffs. Of interest, the study results documented gender differences in the association between anthropometric indices and MetS.

Background

According to the World Health Organization (WHO), the number of obese adults worldwide has reached 650 million in 2016 [body mass index (BMI) ≥30 kg/m²],¹ and the burden of obesity continues to rise in both developed and developing countries. In Lebanon, a small Middle Eastern country, a sharp increase in the prevalence of adult obesity was observed during the past decade, with estimates rising from 17.4% to 28.2% between 1997 and 2009. Obesity contributes to several cardiometabolic abnormalities, including atherogenic dyslipidemia, hyperglycemia, and elevated blood pressure (BP), the constellation of which is referred to as the metabolic syndrome (MetS).²

Several anthropometric indices are used to assess obesity, with the most common being the BMI (weight in kilograms divided by the square of height in meters) and waist circumference (WC). Using these indices, the WHO defines obesity as BMI ≥30 kg/m² and abdominal adiposity as WC ≥94 cm for men and ≥80 cm for women.^3,4 However, these cutoffs were mainly derived from data pertinent to western populations, and an increasing body of evidence suggests that ethnic disparities in body composition patterns may limit the applicability of these cutoffs to other ethnicities.^5,6 For instance, Asians tend to have a higher prevalence of obesity-related morbidity and mortality at lower BMI and WC values compared to Caucasians.^7
–9 Besides BMI and WC, the waist-to-height ratio (WHtR), which accounts for height in addition to WC, was suggested as an additional indicator of abdominal adiposity with a cutoff of 0.5 as a predictor of cardiovascular risk.^9

–12 However, ethnic-specific disparities were also described, with the proposed WHtR cutoff values ranging between 0.47 and 0.63 in different countries and populations.^13

–16

Using anthropometric cutoff values that are not population-specific might lead to misclassification of weight status and could reduce the sensitivity and specificity of these indices in the diagnosis of obesity, hence jeopardizing prevention, early identification, and management of metabolic abnormalities. Accordingly, the WHO has recommended the development and use of population-specific anthropometric cutoff values.⁸ The evidence regarding the validity of various anthropometric cutoffs in identifying obesity is limited in the Middle East,⁶ with available studies producing conflicting results. Hence, the objectives of this study are (1) to determine, for three anthropometric indices (BMI, WC, WHtR), the optimal cutoff values that identify the MetS among Lebanese adults and (2) to evaluate the predictive power of the derived optimal cutoffs for MetS identification, in comparison with reference published cutoffs.

Methodology

Study design and subjects' recruitment

Data for this study were drawn from the Nutrition and Non-Communicable Disease Risk Factor Survey conducted in Lebanon between 2008 and 2009. Details about the design and data collection procedures for this survey are described in details elsewhere.¹⁷ In brief, a nationally representative sample of adults was selected using a door-to-door approach and a stratified cluster sampling technique. Households were the primary sampling units, and from each household, one adult was approached to participate in the survey. A total of 2680 adults (>20 years) were recruited (response rate: 90%). The obtained sample was proportional to the target population (Lebanese adults) based on the age-sex distribution of the Lebanese population.¹⁸ Subjects were interviewed using a multicomponent questionnaire that inquired about socioeconomic characteristics, lifestyle, dietary practices as well as medical history.¹⁸ Survey participants older than 20 years of age and with no previous diagnosis of chronic diseases or metabolic abnormalities (diabetes, hyperglycemia, hypertension, other cardiovascular diseases, dyslipidemia, and cancer) were recontacted to give blood samples (n = 1331). Only 363 subjects gave a blood sample (response rate: 27.3%). For the purpose of this study, data of adults aged between 20 and 65 years were included (n = 314). The protocol was approved by the Institutional Review Board of the American University of Beirut, and all subjects gave informed consent for their participation.

Data collection

Data collection took place in the participants' homes by trained field workers, including phlebotomists and dieticians.

During face-to-face interviews, participants completed a multicomponent questionnaire, addressing sociodemographic and lifestyle characteristics: age, sex, marital status, education level, and income. In addition to income, data regarding crowding index were collected and considered as a proxy of socioeconomic status. Crowding index is calculated as the ratio of the number of people living in the household over the number of rooms in the house used for sleeping. The lifestyle characteristics examined in this study were smoking, alcohol consumption, and physical activity. The latter was assessed using the Arabic version of the short form of the International Physical Activity Questionnaire (IPAQ). Three categories of physical activity (low, moderate, and high) were assigned on the basis of MET-min/week.¹⁹

Anthropometric measurements obtained included weight, height, and WC. For weight measurement, participants were instructed to be in light indoor clothing, barefoot, or wearing stockings. Height was measured using stadiometers without shoes. Weight and height were measured to the nearest 0.1 kg and 0.5 cm, respectively. WC was measured, to the nearest 0.5 cm, using a nonstretching measuring tape at the midpoint between the bottom of the rib cage and above the top of the iliac crest during minimal respiration. BP was measured using a standard mercury sphygmomanometer, after participants were seated and rested for 5 min.

For the biochemical assessment, subjects with no known diagnosis of chronic diseases were invited to provide fasting blood samples. Samples were analyzed for triglycerides (TG) and high-density lipoprotein cholesterol (HDL-C), by an enzymatic spectrophotometric technique using Vitros 350 analyzer (Ortho-Clinical Diagnostics, Johnson & Johnson, 50–100 Holmers Farm Way, High Wycombe, Buckinghamshire, HP12 4DP, United Kingdom). Details about the sensitivity and specificity of the assays used in biochemical assessments are published elsewhere.¹⁷

Definitions

In this study, BMI was calculated as weight (in kilogram) divided by the square of height (in meter) (kg/m²). WHtR was obtained by dividing the WC (in cm) over height (in cm). MetS was defined based on the harmonized definition from the IDF,²⁰ whereby participants were classified as having the MetS if they had three out of the five following cardiometabolic abnormalities: (1) elevated TG level (≥150 mg/dL); (2) low HDL-C level (<40 mg/dL for men and <50 mg/dL for women); (3) elevated BP (systolic BP ≥130 mmHg and/or diastolic BP ≥85 mmHg); (4) elevated fasting glucose level (≥100 mg/dL); and (5) elevated WC (≥94 cm for men, ≥80 cm for women).

Statistical analyses

Descriptive statistics were presented to summarize the study variables of interest as counts and percentages for the categorical variables and as means and standard deviations for the continuous ones. Chi-square and independent t-tests were used to chart comparisons of categorical and continuous variables between participants with and without MetS. Receiver operating characteristic (ROC) analyses were used to determine optimal cutoff values of BMI, WC, and WHtR in identifying those with MetS. The selection of the optimal cutoff point for each of the anthropometric indices was carried out to maximize sensitivity and specificity. For this purpose the Youden's index was calculated as [sensitivity + specificity −1]. The optimal cutoff point which was chosen in the article corresponded to the maximum value of Youden's index.²¹ Odds ratios (ORs) and their respective 95% confidence intervals for the association of the anthropometric measurements with MetS were computed based on multiple logistic regression analysis models. In these models, adjustment was made for sociodemographic and lifestyle variables that showed significant associations with MetS in univariate analysis. All reported P-values were based on two-sided tests and were compared with a significance level of 5%. The Statistical Package for the Social Sciences (SPSS) was used for all computations.²²

Results

Study participants with missing values for any anthropometric, BP measurements, or biochemical analyses were excluded from the analyses for this article (n = 9), resulting in a sample size of 305 subjects. The sociodemographic and lifestyle characteristics of study participants by MetS for males and females are presented in Table 1. Prevalence of MetS in the study population was 29.5% (39.6% among males and 21.1% among females). Age was significantly associated with MetS in both males and females, with higher prevalence rates of MetS among older subjects (P = 0.04 and P < 0.01, respectively). Among males, the prevalence of MetS was higher among married compared with unmarried subjects; however, among females, marital status was associated with lower prevalence of MetS. Educational status was associated with MetS, only among females, with lower proportions of MetS among subjects with higher levels of education. Among lifestyle factors, smoking was associated with higher prevalence of MetS among males only (Table 1).

Table 1.

Sociodemographic and Lifestyle Characteristics by Metabolic Syndrome, Among Males and Females in the Study Population (n = 305)

	Males (n = 139)		P	Females (n = 166)		P
	Without MetS (n = 84)	With MetS (n = 55)	P	Without MetS (n = 131)	With MetS (n = 35)	P
Age (years)
20–34	56 (69.1)	25 (30.9)	0.04	57 (89.1)	7 (10.9)	<0.01
35–49	18 (47.4)	20 (52.6)		65 (77.4)	19 (22.6)
≥50	10 (50)	10 (50)		9 (50)	9 (50)
Marital status
Not married	54 (69.2)	24 (30.8)	0.02	45 (91.8)	4 (8.2)	<0.01
Married	30 (49.2)	31 (50.8)	0.02	86 (96.6)	31 (3.4)	<0.01
Educational level
Primary or less	18 (69.2)	8 (30.8)	0.22	9 (60)	6 (40)	0.02
Primary to high school	77 (62.6)	46 (37.4)		67 (75.3)	22 (24.7)
College or higher	44 (59.5)	30 (40.5)		55 (88.7)	7 (11.3)
Household income per month
<3 millions L.L.	24 (70.6)	10 (29.4)	0.19	39 (83.0)	8 (17.0)	0.19
>3 millions L.L.	3 (37.5)	5 (62.5)		8 (100)	0 (0)
No answer	57 (58.8)	40 (41.2)		84 (75.7)	27 (24.3)
Crowding index
<1	31 (57.4)	23 (42.6)	0.56	43 (87.8)	6 (12.2)	0.08
≥1	53 (62.4)	32 (37.6)	0.56	87 (75.7)	28 (24.3)	0.08
Smoking
Nonsmoker	57 (68.7)	26 (31.3)	0.02	100 (78.1)	28 (21.9)	0.65
Current smoker	27 (48.2)	29 (51.8)	0.02	31 (81.6)	7 (18.4)	0.65
Alcohol consumption
Do not drink	36 (63.2)	21 (36.8)	0.67	87 (81.3)	20 (18.7)	0.93
Drink	44 (59.5)	30 (40.5)	0.67	42 (80.8)	10 (19.2)	0.93
Physical activity
Low	41 (61.2)	26 (38.8)	0.17	51 (76.1)	16 (23.9)	0.82
Moderate	17 (48.6)	18 (51.4)		46 (79.3)	12 (20.7)
High	26 (70.3)	11 (29.7)		30 (81.1)	7 (18.9)

Some numbers do not add up to the total count because of missing values.

MetS, metabolic syndrome; L.L., Lebanese Lira.

Descriptive statistics of anthropometric and biochemical measurements and their associations with MetS among males and females are presented in Table 2. As expected, the prevalence of MetS was significantly higher among subjects who were overweight or obese, had elevated WC, and WHtR (P < 0.01). Similarly, prevalence estimates of all metabolic abnormalities (fasting blood glucose, low HDL-C, elevated TG, and elevated BP) were significantly higher among those with MetS compared with those without (Table 2).

Table 2.

Anthropometric Measurements, Blood Pressure, and Biochemical Characteristics by Metabolic Syndrome, Among Males and Females in the Study Population (n = 305)

	Males (n = 139)		P^a	Females (n = 166)		P^a
	Without MetS (n = 84)	With MetS (n = 55)	P^a	Without MetS (n = 131)	With MetS (n = 35)	P^a
Anthropometric measurements
Height (cm)	174.88 ± 7.02	175.90 ± 5.60	0.37	160.98 ± 6.57	159.50 ± 6.72	0.24
Weight (kg)	78.99 ± 13.59	91.39 ± 14.53	<0.001	64.36 ± 12.69	76.57 ± 14.16	<0.001
BMI (kg/m²)⁴	25.78 ± 4.03	29.55 ± 4.68	<0.001	24.85 ± 4.76	30.07 ± 5.10	<0.001
Underweight (<18.5)	0	0	<0.001	1 (100)	0 (0)	<0.001
Normal (18.5–24.9)	36 (78.3)	10 (21.7)		79 (84)	5 (6)
Overweight (25–34.9)	37 (66.1)	19 (33.9)		32 (65.3)	17 (34.7)
Obese (≥30)	11 (29.7)	26 (70.3)		19 (59.4)	13 (40.6)
WC (cm)²⁴	87.94 ± 10.93	99.40 ± 11.23	<0.001	80.05 ± 11.17	91.72 ± 11.55	<0.001
Normal	62 (80.5)	15 (19.5)	<0.001	73 (97.3)	2 (2.7)	<0.001
Elevated (males ≥94; females ≥80)	22 (35.5)	40 (64.5)		58 (63.7)	33 (36.3)
WHtR¹²	0.50 ± 0.06	0.57 ± 0.04	<0.001	0.50 ± 0.07	0.57 ± 0.07	<0.001
Normal	41 (78.8)	11 (21.2)	<0.001	76 (95)	4 (5)	<0.001
Elevated (≥0.5)	43 (49.4)	44 (50.6)		55 (64)	31 (36)
Biochemical measurements
FBG (mg/dL): mean ± SD	96.19 ± 14.50	107.05 ± 16.06	<0.001	94.44 ± 12.06	109.51 ± 16.10	<0.001
Normal	64 (78)	18 (22)	<0.001	98 (93.3)	7 (6.7)	<0.001
Elevated (≥100)²⁰	20 (35.1)	37 (64.9)		33 (54.1)	28 (45.9)
HDL cholesterol (mg/dL)	46.81 ± 10.50	37.69 ± 8.93	<0.001	59.45 ± 13.57	49.69 ± 13.14	<0.001
Normal	63 (74.1)	22 (25.9)	<0.001	107 (88.4)	14 (11.6)	<0.001
Low (males <40, females <50)²⁰	21 (38.9)	33 (61.1)		24 (53.3)	21 (46.7)
TG (mg/dL)	113.74 ± 48.52	215.04 ± 97.29	<0.001	102.20 ± 49.24	162.46 ± 73.19	<0.001
Normal	75 (87.2)	11 (12.8)	<0.001	117 (86)	19 (14)	<0.001
Elevated (≥150)²⁰	9 (17)	44 (83)		14 (46.7)	16 (53.3)
SBP (mmHg)	125.98 ± 12.73	135.07 ± 16.45	<0.001	113.967 ± 10.85	127.25 ± 17.53	<0.001
DBP (mmHg)	77.53 ± 9.66	82.05 ± 10.05	0.009	71.95 ± 7.92	79.80 ± 12.75	0.001
Normal BP	48 (80)	12 (20)	<0.001	117 (87.3)	17 (12.7)	<0.001
Elevated BP (SBP ≥130 and/or DBP ≥85)²⁰	36 (45.6)	43 (54.4)		14 (42.4)	18 (57.6)

Values in this table represent mean ± SD and n (%) for continuous and categorical variables, respectively.

P-values were derived from interdependent t-test and chi square analysis for continuous and categorical variables, respectively.

BMI, body mass index; BP, blood pressure; DBP, diastolic blood pressure; FBG, fasting blood glucose; HDL, high-density lipoprotein; SBP, systolic blood pressure; TG, triglyceride; WC, waist circumference; WHtR, waist-to-height ratio.

ROC curves are shown in Fig. 1a for males and Fig. 1b for females. For the anthropometric indices considered in this study (BMI, WC, and WHtR), the OR and their corresponding 95% confidence interval for their association with MetS are presented in Table 3. All the indices were significantly associated with odds of MetS in both males and females. The areas under the ROC curves were all significant at P = 0.01 for both males and females (Table 3). For BMI, the optimal cutoffs were 26.35 kg/m² for males and 25.74 kg/m² for females, with better sensitivity, specificity, and accuracy ratio among females than in males. For WC and WHtR, the optimal cutoffs were 94 cm and 0.54 for males and 82.62 cm and 0.53 among females, respectively. For these two anthropometric indices (WC and WHtR), the accuracy ratio was higher than 70% in both sexes indicating a good performance (Table 3). The negative predictive value was greater than 90% for all indices among females and greater than 77% among males. On the contrary, the positive predictive value estimates for the indicators were low among females, ranging between 41.1% for WC and 43.9% for WHtR and among males, ranging between 58.1% for BMI and 68.6% for WHtR (Table 3). A sensitivity analysis was conducted excluding the one underweight female subject. A comparison between the results with and without this subject showed minimal differences (in the range of 10⁻²–10⁻¹). The significance as well as the optimal cutoffs remained unchanged.

FIG. 1.

ROC curves for the prediction of MetS using BMI, WC, and WHtR for males (a) and females (b). BMI, body mass index; MetS, metabolic syndrome; ROC, receiver operating characteristic; WC, waist circumference; WHtR, waist-to-height ratio.

Table 3.

Anthropometric Indices and Metabolic Syndrome in the Study Population: Area Under the Curves of the Receiver Operating Characteristic, Selected Cutoffs, and Their Performance Among Male and Females

	OR (95% CI) ^a	P	AUC (95% CI)	P^b	Cutoff values	Sensitivity	Specificity	Accuracy ratio	PPV	NPV
Males
BMI (kg/m²)	1.23 (1.11–1.36)	<0.01	0.74 (0.66–0.83)	<0.01	26.35	78.2	63.1	69.0	58.1	81.5
WC (cm)	1.10 (1.05–1.14)	<0.01	0.77 (0.69–0.85)	<0.01	94	72.7	73.8	73.4	64.5	80.5
WHtR	1.16 (1.08–1.24)	<0.01	0.76 (0.68–0.84)	<0.01	0.54	63.6	81.0	74.1	68.6	77.3
Females
BMI (kg/m²)	1.18 (1.08–1.28)	<0.01	0.80 (0.73–0.87)	<0.01	25.74	85.7	69.5	72.9	42.9	94.8
WC (cm)	1.08 (1.03–1.12)	<0.01	0.80 (0.73–0.87)	<0.01	82.62	85.7	67.2	71.1	41.1	94.6
WHtR	1.12 (1.05–1.19)	<0.01	0.80 (0.73–0.87)	<0.01	0.53	82.9	71.8	74.1	43.9	94.0

OR were adjusted for variables that were significantly associated with MetS in Table 1.

These P-values were derived from comparing the AUCs obtained in this study with a baseline model of 50%.

AUC, area under the curve; CI, confidence interval; NPV, negative predictive value; OR, odds ratio; PPV, positive predictive value.

The associations of the anthropometric indices with MetS were examined using the derived optimal cutoffs as well as reference published cutoffs. Among males, using the derived cutoff, the greatest association with MetS was observed for WHtR and WC followed by BMI. In females, BMI was the strongest predictor of MetS, followed by WHtR and WC (Table 4). When comparing the generated cutoffs to published ones, the largest differences in ORs were for WHtR in men and BMI in women.

Table 4.

Association of Metabolic Syndrome with Anthropometric Indices Among Male and Females in the Study Population

	OR (95% CI) ^a	P
Males
BMI (kg/m²)
Optimal cutoff (>26.35)	5.81 (2.54–13.30)	<0.01
Published cutoff (>30)⁴	5.18 (2.16–12.43)	<0.01
WC (cm)
Optimal cutoff (>94)	6.79 (3.01–15.32)	<0.01
Published cutoff (>94)³	6.79 (3.01–15.32)	<0.01
WHtR
Optimal cutoff (>0.54)	7.30 (3.09–17.26)	<0.01
Published cutoff (>0.5)¹²	2.94 (1.24–7.01)	0.01
Females
BMI (kg/m²)
Optimal cutoff (>25.74)	9.83 (3.45–28.00)	<0.01
Published cutoff (>30)⁴	2.58 (1.01–6.57)	0.05
WC (cm)
Optimal cutoff (>82.62)	9.53 (3.32–27.32)	<0.01
Published cutoff (>80)²⁴	8.11 (2.62–25.14)	<0.01
WHtR
Optimal cutoff (>0.53)	9.75 (3.59–26.46)	<0.01
Published cutoff (>0.5)¹²	8.37 (2.68–26.11)	<0.01

OR were adjusted for variables that were significantly associated with MetS in Table 1.

Discussion

The present study determined optimal anthropometric cutoffs for the identification of MetS among 20–65-year-old adults in Lebanon, a country that is witnessing a sharp increase in obesity prevalence and associated comorbidities.²³ The optimal cutoff values for men and women were 26.35 and 25.74 kg/m² for BMI, 94 and 83 cm for WC, and 0.54 and 0.53 for WHtR, respectively. The study showed that, using the derived optimal cutoff values, WHtR and WC performed better than BMI in predicting MetS in men, while in women, BMI, WC, and WHtR were all strongly associated with the MetS.

The BMI cutoff value derived in our study (≈26 kg/m²) is lower than the internationally recommended WHO cutoff of 30 kg/m² for obesity identification.²⁴ Disparities in the applicability of the WHO BMI cutoff values have been previously described, with increasing evidence suggesting that ethnic-specific differences in body composition may limit their use in various ethnicities.^5,6 For instance, in studies conducted among South Asians, the BMI cutoffs for the identification of obesity and cardiometabolic risk were reported to range between 21.5 and 24 kg/m².^8,9 Studies conducted among Middle Eastern populations are rather scarce. The BMI cutoff derived in our study falls within the range reported by other countries in the region (24–27 kg/m² in Tunisia; 23.3–26.8 kg/m² in Oman; and 26–28.5 kg/m² in Iran).^25
–27 Taken together, these findings indicate that Middle Eastern adults may have a higher susceptibility to adverse metabolic profiles at a relatively lower BMI. There is a need for future studies examining body composition, fat deposition patterns, and their association with cardiometabolic abnormalities in populations of the Middle East.

The study findings showed that the optimal cutoff values for WC were 94 cm in Lebanese men and 83 cm in women. These cutoffs are higher than those reported for Asian populations such as Chinese (85 cm in men and 75 cm in women)⁹ and Sri Lankan adults (76 cm in both genders),⁸ while being lower than those reported for North Americans (102 cm in men and 88 cm in women).^6,28
–30 The optimal cutoffs generated in our study were also lower than values reported by previous studies in Lebanon and Qatar^13,31 (99.5 cm in men and 91 cm in women) as well as in Jordan (98 cm in men and 96 cm in women),¹⁴ and Iraq (97 cm in men and 99 cm in women).³² Estimates provided by Tunisia and Oman, two other countries in the region, were relatively similar to our optimal cutoff in women (85 cm), but were lower in men (85 and 80 cm, respectively).^25,26 These differences in the generated cutoffs may be attributed to ethnic-specific variations, but may also result from the use of different criteria for identifying the MetS.¹³ Taken together, our results as well as those stemming from other countries in the Middle East confirm that the North American cutoffs for elevated WC (102 cm in men and 90 cm in women)^6,30 are not appropriate for adults in the region. The IDF has in fact recommended that Eastern Mediterranean and Middle East Arab populations use the Europids cutoff values of 94 cm in men and 80 cm in women until more specific data are available for this population.³³

In this study, the WHtR optimal cutoff was estimated at 0.54 and 0.53 in Lebanese men and women, respectively. This value is relatively higher than estimates reported among Asians, which tended to fall within the range of 0.45–0.52.^34

–37 Few studies in the Middle East have investigated WHtR cutoffs for the prediction of MetS. The generated WHtR cutoff in our study is lower than values reported by Al-Odat et al.¹⁴ (0.61) and Bener et al.¹³ (0.58–0.63) among Jordanian and Qatari adults, respectively.^13,14

In our study, the use of the derived optimal cutoffs improved the prediction of the MetS compared to the use of reference published cutoffs, among both sexes. Of interest, our results highlighted gender disparities in the ability of anthropometric indices to predict the MetS. Based on the derived optimal cutoffs, indicators of abdominal obesity were the highest predictors of MetS in men, with WHtR presenting the strongest association, followed by WC. Several studies conducted in various ethnic groups have also shown that indicators of abdominal adiposity in men were stronger predictors of cardiometabolic abnormalities, compared to other obesity indicators such as BMI.^{9,10,13,36,38} It is argued that WHtR (i.e., WC adjusted for height) may be more appropriate for the assessment of abdominal obesity compared to WC, as it was suggested that measuring WC alone as a surrogate for abdominal fat may lead to an overestimation of MetS risk in tall individuals and an underestimation in short ones.^39
–41 Unlike the results observed among men, BMI, WC, and WHtR were all strong predictors of MetS in women. Such gender disparities have been previously described in the literature. A cross-sectional survey among white and African American adults in the United States showed that BMI was inferior to WC in predicting MetS among men in general but not in women.⁴² Similarly, another study conducted among Chinese adults suggested that the best indicator of hypertension was BMI in women and WHtR in men.⁴³ Such differences may be explained by gender variations in body fat distribution.^44,45 Compared with women, men have greater volume of visceral fat and are more likely to accumulate adipose tissue in the upper body,⁴⁶ whereas women usually accumulate adipose tissue in their lower body.

The strengths of the study include the collection of anthropometric measurements and biochemical data using standardized protocols, which reduced measurement errors. Another strength is the recruitment of study subjects, which was based on random sampling rather than enrolling subjects who were seeking health checkups in medical centers, and who therefore may be more health conscious.⁹ In addition, data on potential confounders such as smoking, alcohol consumption, physical activity, and socioeconomic characteristics were collected by trained researchers, and statistical analyses were adjusted for these confounders. However, the study findings ought to be considered in light of the following limitations. First, the study adopted a cross-sectional design, which cannot be used to establish causality or temporal relationships.⁹ Second, the study is based on a relatively small sample of adult subjects (n = 305), given that the percentage of subjects who agreed to give blood samples (respondents) was low (27.3%). However, comparison of the sociodemographic characteristics between respondents and nonrespondents showed no significant differences between the two groups, except for marital status (62% of respondents are married vs. 50% of nonrespondents).¹⁷

In conclusion, this study determined optimal anthropometric cutoffs for the identification of MetS among adults in Lebanon. It showed that the optimal BMI cutoff was of ∼26 kg/m², hence lower than the internationally recommended WHO cutoff of 30 kg/m². As for the indicators of abdominal obesity, the optimal cutoff values in men and women were of 94 and 83 cm for WC, and 0.54 and 0.53 for WHtR, respectively. Of interest, the study results documented gender differences in the association between anthropometric indices and MetS. In a country that is witnessing an escalating trend in obesity and associated noncommunicable diseases, accurate identification of at risk individuals is an utmost priority. The adoption of the optimal cutoffs derived in our study may assist public health professionals and clinical practitioners in providing ethnic-specific preventive and curative strategies tailored to Lebanese adults. Future studies are needed to validate the generated cutoffs in other countries of the Middle East and in prospective cohorts, examining obesity indices in relationship to cardiometabolic death and morbidity.

Footnotes

Authors' Contributions

L.N. led the write-up of the article and supervised interpretation of the results; F.N. led the conception, study design, data analysis, and contributed to the writing of the article. N.B. contributed to the interpretation of the results and the write-up of the article; S.K. was responsible for the statistical analysis; M.C.C. supervised participants' recruitment and data collection. N.H. and A.S. were responsible for the conceptualization of the study objectives, provided valuable input for data interpretation, and critically reviewed the article. All authors reviewed the results and approved the final version of the article.

Author Disclosure Statement

No conflicting financial interests exist.

Funding Information

The original survey was funded by the World Health Organization (N.H. grant number TSA/08/00150. EM0882149) and the Lebanese National Council for Scientific Research (LNCSR) (N.H. and L.N., grant number 03-12-08). Internal funding from the American University of Beirut supported the data analysis included in this research.

References

WHO. Obesity and overweight. 2018. Accessed at www.who.int/en/news-room/fact-sheets/detail/obesity-and-overweight on January 1, 2019.

Moller

, Kaufman

. Metabolic syndrome: A clinical and molecular perspective. Annu Rev Med, 2005; 56:45–62.

WHO. Obesity: Preventing and managing the global epidemic. 2000. Accessed at www.who.int/nutrition/publications/obesity/WHO_TRS_894/en on February 1, 2019.

WHO Obesity. Accessed at www.who.int/topics/obesity/en on January 1, 2019.

Wang

, Ma

, Si

. Optimal cut-off values and population means of waist circumference in different populations. Nutr Res Rev, 2010; 23:191–199.

Lear

, James

, Ko

, et al. Appropriateness of waist circumference and waist-to-hip ratio cutoffs for different ethnic groups. Eur J Clin Nutr, 2010; 64:42.

WHO. The Asia-Pacific Perspective: Redifining Obesity and Its treatment: Health Communication Australia; 2000.

Katulanda

, Jayawardena

, Sheriff

, et al. Derivation of anthropometric cut-off levels to define CVD risk in Sri Lankan adults. Br J Nutr, 2011; 105:1084–1090.

Zeng

, He

, Dong

, et al. Optimal cut-off values of BMI, waist circumference and waist: Height ratio for defining obesity in Chinese adults. Br J Nutr, 2014; 112:1735–1744.

10.

Lee

CMY

, Huxley

, Wildman

, et al. Indices of abdominal obesity are better discriminators of cardiovascular risk factors than BMI: A meta-analysis. J Clin Epidemiol, 2008; 61:646–653.

11.

Ashwell

, Gunn

, Gibson

. Waist-to-height ratio is a better screening tool than waist circumference and BMI for adult cardiometabolic risk factors: Systematic review and meta-analysis. Obes Rev, 2012; 13:275–286.

12.

Browning

, Hsieh

, Ashwell

. A systematic review of waist-to-height ratio as a screening tool for the prediction of cardiovascular disease and diabetes: 0 · 5 could be a suitable global boundary value. Nutr Res Rev, 2010; 23:247–269.

13.

Bener

, Yousafzai

, Darwish

, et al. Obesity index that better predict metabolic syndrome: Body mass index, waist circumference, waist hip ratio, or waist height ratio. J Obes, 2013; 2013:269038.

14.

Al-Odat

, Ahmad

, Haddad

. References of anthropometric indices of central obesity and metabolic syndrome in Jordanian men and women. Diabetes Metab Syndr, 2012; 6:15–21.

15.

Raposo

, Severo

, Santos

. Adiposity cut-off points for cardiovascular disease and diabetes risk in the Portuguese population: The PORMETS study. PLoS One, 2018; 13:e0191641.

16.

Hastuti

, Kagawa

, Byrne

, et al. Determination of new anthropometric cut-off values for obesity screening in Indonesian adults. Asia Pac J Clin Nutr, 2017; 26:650.

17.

Naja

, Nasreddine

, Itani

, et al. Association between dietary patterns and the risk of metabolic syndrome among Lebanese adults. Eur J Nutr, 2013; 52:97–105.

18.

UNDP Living Conditions of Households: The National Survey of Household Living Conditions, 2004. Lebanese Republic Ministry of Social Affairs. Central Administration of Statistics, United Nations Development Program, Lebanon. 2006.

19.

Committee IR. Guidelines for data processing and analysis of the International Physical Activity Questionnaire (IPAQ)-short and long forms. 2005. Accessed at www.ipaq.ki.se/scoring.pdf on January 1, 2019.

20.

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.

21.

Reiser

. Measuring the effectiveness of diagnostic markers in the presence of measurement error through the use of ROC curves. Stat Med, 2000; 19:2115–2129.

22.

SPSS. Statistical package for windows, 2003, version 24. Chicago, IL: SPSS Inc.

23.

Sibai

, Nasreddine

, Mokdad

, et al. Nutrition transition and cardiovascular disease risk factors in Middle East and North Africa countries: Reviewing the evidence. Ann Nutr Metab, 2010; 57:193–203.

24.

WHO Obesity: Preventing and Managing the Global Epidemic. Report of a WHO Consultation Group on Obesity. Geneva, Switzerland: WHO.

25.

Bouguerra

, Alberti

, Smida

, et al. Waist circumference cut-off points for identification of abdominal obesity among the Tunisian adult population. Diabetes Obes Metab, 2007; 9:859–868.

26.

Al-Lawati

, Jousilahti

. Body mass index, waist circumference and waist-to-hip ratio cut-off points for categorisation of obesity among Omani Arabs. Public Health Nutr, 2008; 11:102–108.

27.

Aghababaie

Predictors of metabolic syndrome in the Iranian population: Waist circumference, body mass index, or waist to hip ratio?. Cholesterol, 2013:6.

28.

Lean

, Han

, Morrison

. Waist circumference as a measure for indicating need for weight management. BMJ, 1995; 311:158–161.

29.

Lean

, Han

, Seidell

. Impairment of health and quality of life in people with large waist circumference. Lancet, 1998; 351:853–856.

30.

National Cholesterol Education Program. Executive summary of the third report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III). JAMA, 2001; 285:2486.

31.

Chedid

, Gannagé-Yared

M-H

, Khalifé S, et al. Impact of different metabolic syndrome classifications on the metabolic syndrome prevalence in a young Middle Eastern population. Metabolism, 2009; 58:746–752.

32.

Mansour

, Al-Hassan

, Al-Jazairi

. Cut-off values for waist circumference in rural Iraqi adults for the diagnosis of metabolic syndrome. Rural Remote Health, 2007; 7:765.

33.

International Diabetes Federation. The IDF Consensus Worldwide Definition of the Metabolic Syndrome, 2006. Brussels, Belgium: International Diabetes Federation (IDF).

34.

Tseng

C-H

, Chong

C-K

, Chan

T-T

, et al. Optimal anthropometric factor cutoffs for hyperglycemia, hypertension and dyslipidemia for the Taiwanese population. Atherosclerosis, 2010; 210:585–589.

35.

S-Y

, Lam

T-H

, Janus

. Waist to stature ratio is more strongly associated with cardiovascular risk factors than other simple anthropometric indices. Ann Epidemiol, 2003; 13:683–691.

36.

Lin

, Lee

, Chen

, et al. Optimal cut-off values for obesity: Using simple anthropometric indices to predict cardiovascular risk factors in Taiwan. Int J Obes, 2002; 26:1232.

37.

, Zhai

, Ma

, et al. Abdominal obesity and the prevalence of diabetes and intermediate hyperglycaemia in Chinese adults. Public Health Nutr, 2009; 12:1078–1084.

38.

Park

S-H

, Choi

S-J

, Lee

K-S

, et al. Waist circumference and waist-to-height ratio as predictors of cardiovascular disease risk in Korean adults. Circ J, 2009; 73:1643–1650.

39.

Hsieh

, Yoshinaga

. Do people with similar waist circumference share similar health risks irrespective of height?. Tohoku J Exp Med, 1999; 188:55–60.

40.

Ashwell

, Gibson S Waist to height ratio is a simple and effective obesity screening tool for cardiovascular risk factors: Analysis of data from the British National Diet and Nutrition Survey of adults aged 19–64 years. Obesity Facts, 2009; 2:97–103.

41.

Hsieh

, Ashwell

, Muto

, et al. Urgency of reassessment of role of obesity indices for metabolic risks. Metabolism, 2010; 59:834–840.

42.

Beydoun

, Kuczmarski

MTF

, Wang

, et al. Receiver-operating characteristics of adiposity for metabolic syndrome: The Healthy Aging in Neighborhoods of Diversity across the Life Span (HANDLS) study. Public Health Nutr, 2011; 14:77–92.

43.

Zhou

, Hu

, Chen

. Association between obesity indices and blood pressure or hypertension: Which index is the best?. Public Health Nutr, 2009; 12:1061–1071.

44.

Krotkiewski

, Björntorp

, Sjöström

, et al. Impact of obesity on metabolism in men and women. Importance of regional adipose tissue distribution. J Clin Invest, 1983; 72:1150–1162.

45.

Tchernof

, Després

J-P

. Pathophysiology of human visceral obesity: An update. Physiol Rev, 2013; 93:359–404.

46.

Demerath

, Sun

, Rogers

, et al. Anatomical patterning of visceral adipose tissue: Race, sex, and age variation. Obesity, 2007; 15:2984–2993.