Neck Circumference as Predictor of Cardio-Metabolic Syndrome
Neck Circumference as Predictor of Cardio-Metabolic Syndrome
The aim of the present study was to determine whether neck circumference (NC) was associated with cardio-metabolic risks and contributed to the prediction beyond the classical anthropometric indices in a cross-sectional survey of 4201 individuals from China. We found that NC was significantly associated with cardio-metabolic risk factors and independently contributed to predicting the likelihood of cardio-metabolic risks both in males and females.
Based on previous studies, obesity is widely accepted to be associated with metabolic disorders and cardiovascular risk factors. Although BMI, WC and WHpR are widely used anthropometric indices to reflect obesity and predict cardio-metabolic risks, an increasing number of studies have suggested that NC is a simpler, more innovative and practical anthropometric parameter. WC measurements are easily affected by being full or hungry, respiratory movement and wearying heavy clothing, whereas these problems can be avoided in the NC measurement. Therefore, NC was a more reliable anthropometric indice to indicate central obesity. In the Framingham Heart Study, including 2732 subjects (mean age: 57 y), Preis et al found that NC was positively associated with risks of type 2 diabetes, hypertension, decreased HDL-C and increased TG. After further adjustments for BMI and WC, NC remained associated with type 2 diabetes. Similar results were observed in a Turkish Adult Cohort Study in 1912 middle-aged and elderly individuals. In our study, we found that NC was significantly correlated with all outcomes of cardio-metabolic risks in both genders, which was in line with the previous studies. NC might reflect fat deposition at an ectopic site such as observed in fatty liver. Therefore, the thicker the NC, the greater the risk of cardio-metabolic syndrome is.
Next, we calculated AUCs to evaluate the predictive values of NC and other anthropometric indices for cardio-metabolic risks. According to the ROC analyses between anthropometric indices and MS-rf, the NC presented a significantly large AUC (men: 0.683, women: 0.703), but the values were relatively lower than those of WHpR (0.716 and 0.766), WC (0.715 and 0.764) and BMI (0.699 and 0.723). Up to date, few studies have compared the effect size of NC with WC, BMI, WHtR or other anthropometric indices. Some studies, but not all, observed a less strong association in NC than those of BMI, WHpR and WC with metabolic abnormalities. From these results, we could not consider NC superior to WC, BMI and WHpR to predict cardio-metabolic risks.
It is well established that visceral fat (or central obesity) has greater effect on the development of metabolic abnormalities than subcutaneous fat (general obesity). WC, WHpR, sagittal abdominal diameter and visceral adiposity index are widely used markers of visceral fat. These indices are more closely correlated to visceral fat than BMI and NC. A variety of studies have demonstrated that these indices of visceral fat have stronger association with metabolic abnormalities or CVD risks than BMI. Therefore, it was likely that the inefficient predictive values of NC (vs. WC or WHtR) for MS in this study were due to its poor correlation with visceral fat.
We observed optimal NC cut-offs of ≥37 cm in men and ≥33 cm in women for the prediction of MS in this population. Greater values were found in men than in women, which was in consistent with previous study. However, much greater NC cut-offs in both gender were observed in Brazil study (>40 cm in men and > 36 cm in women) for the prediction of MS and insulin resistance. Larger cut-off values of NC for the prediction of MS were also noted in a Turkey population. Greater differences in body size might partially explain the heterogeneity in the optimal cut-offs of NC as WC among different populations. In this regard, ethnic-specific cut-offs of NC would be required for the prediction of cardio-metabolic abnormalities.
Furthermore, the cardio-metabolic risk factors were categorized as dichotomous variables, and their associations with NC were evaluated by logistic regression analysis. The ORs found that the associations between NC and cardio-metabolic risk factors were similar for both genders. Originally, NC was significantly associated with the likelihood of MS-rf and each component of cardio-metabolic syndrome in the model adjusted for age. Subsequently, when controlled for age, BMI, WC and WHpR, the measurement of NC showed significant independence of association regarding the likelihood of cardio-metabolic risk. Thus, we could conclude from these results that NC contributed to cardio-metabolic risk independently of other anthropometric indices in Chinese population, which was in accordance with a previous diabetic population-based study. A measurement of NC might yield additional information in terms of the identification of cardio-metabolic syndrome.
The present study has some limitations. First, this cross-sectional design study limited extension of its interpretation to the causality of associations. Second, all the participants were from the same health examination center, and a selective bias could not be excluded. Finally, because the survey was completed in a single visit, the inherent variability in laboratory tests and measurements could not be taken into account. Despite these limitations, our study has the advantage of introducing a simple and inexpensive method to predict cardio-metabolic risks in a large population. However, because the study was limited to the representativeness of the study sample and cross-sectional study design, further longitudinal studies in representative populations are required to obtain more conclusive results to establish NC as a basic criterion in the diagnosis of cardio-metabolic syndrome.
Discussion
The aim of the present study was to determine whether neck circumference (NC) was associated with cardio-metabolic risks and contributed to the prediction beyond the classical anthropometric indices in a cross-sectional survey of 4201 individuals from China. We found that NC was significantly associated with cardio-metabolic risk factors and independently contributed to predicting the likelihood of cardio-metabolic risks both in males and females.
Neck Circumference and Cardio-metabolic Risks
Based on previous studies, obesity is widely accepted to be associated with metabolic disorders and cardiovascular risk factors. Although BMI, WC and WHpR are widely used anthropometric indices to reflect obesity and predict cardio-metabolic risks, an increasing number of studies have suggested that NC is a simpler, more innovative and practical anthropometric parameter. WC measurements are easily affected by being full or hungry, respiratory movement and wearying heavy clothing, whereas these problems can be avoided in the NC measurement. Therefore, NC was a more reliable anthropometric indice to indicate central obesity. In the Framingham Heart Study, including 2732 subjects (mean age: 57 y), Preis et al found that NC was positively associated with risks of type 2 diabetes, hypertension, decreased HDL-C and increased TG. After further adjustments for BMI and WC, NC remained associated with type 2 diabetes. Similar results were observed in a Turkish Adult Cohort Study in 1912 middle-aged and elderly individuals. In our study, we found that NC was significantly correlated with all outcomes of cardio-metabolic risks in both genders, which was in line with the previous studies. NC might reflect fat deposition at an ectopic site such as observed in fatty liver. Therefore, the thicker the NC, the greater the risk of cardio-metabolic syndrome is.
Comparison of Associations Among the Anthropometric Indices
Next, we calculated AUCs to evaluate the predictive values of NC and other anthropometric indices for cardio-metabolic risks. According to the ROC analyses between anthropometric indices and MS-rf, the NC presented a significantly large AUC (men: 0.683, women: 0.703), but the values were relatively lower than those of WHpR (0.716 and 0.766), WC (0.715 and 0.764) and BMI (0.699 and 0.723). Up to date, few studies have compared the effect size of NC with WC, BMI, WHtR or other anthropometric indices. Some studies, but not all, observed a less strong association in NC than those of BMI, WHpR and WC with metabolic abnormalities. From these results, we could not consider NC superior to WC, BMI and WHpR to predict cardio-metabolic risks.
It is well established that visceral fat (or central obesity) has greater effect on the development of metabolic abnormalities than subcutaneous fat (general obesity). WC, WHpR, sagittal abdominal diameter and visceral adiposity index are widely used markers of visceral fat. These indices are more closely correlated to visceral fat than BMI and NC. A variety of studies have demonstrated that these indices of visceral fat have stronger association with metabolic abnormalities or CVD risks than BMI. Therefore, it was likely that the inefficient predictive values of NC (vs. WC or WHtR) for MS in this study were due to its poor correlation with visceral fat.
Optimal Cut-off Points and Independent Contribution of Neck Circumference
We observed optimal NC cut-offs of ≥37 cm in men and ≥33 cm in women for the prediction of MS in this population. Greater values were found in men than in women, which was in consistent with previous study. However, much greater NC cut-offs in both gender were observed in Brazil study (>40 cm in men and > 36 cm in women) for the prediction of MS and insulin resistance. Larger cut-off values of NC for the prediction of MS were also noted in a Turkey population. Greater differences in body size might partially explain the heterogeneity in the optimal cut-offs of NC as WC among different populations. In this regard, ethnic-specific cut-offs of NC would be required for the prediction of cardio-metabolic abnormalities.
Furthermore, the cardio-metabolic risk factors were categorized as dichotomous variables, and their associations with NC were evaluated by logistic regression analysis. The ORs found that the associations between NC and cardio-metabolic risk factors were similar for both genders. Originally, NC was significantly associated with the likelihood of MS-rf and each component of cardio-metabolic syndrome in the model adjusted for age. Subsequently, when controlled for age, BMI, WC and WHpR, the measurement of NC showed significant independence of association regarding the likelihood of cardio-metabolic risk. Thus, we could conclude from these results that NC contributed to cardio-metabolic risk independently of other anthropometric indices in Chinese population, which was in accordance with a previous diabetic population-based study. A measurement of NC might yield additional information in terms of the identification of cardio-metabolic syndrome.
Limitations
The present study has some limitations. First, this cross-sectional design study limited extension of its interpretation to the causality of associations. Second, all the participants were from the same health examination center, and a selective bias could not be excluded. Finally, because the survey was completed in a single visit, the inherent variability in laboratory tests and measurements could not be taken into account. Despite these limitations, our study has the advantage of introducing a simple and inexpensive method to predict cardio-metabolic risks in a large population. However, because the study was limited to the representativeness of the study sample and cross-sectional study design, further longitudinal studies in representative populations are required to obtain more conclusive results to establish NC as a basic criterion in the diagnosis of cardio-metabolic syndrome.
