Study Notes on Cardiovascular Risk Assessment

Indices of Abdominal Obesity vs. Body Mass Index in Cardiovascular Risk Factors

Authors and Affiliations

• Crystal Man Ying Lee, Nutrition and Lifestyle Division, The George Institute for International Health, University of Sydney, Australia

• Rachel R. Huxley, Department of Epidemiology and Population Health, Albert Einstein College of Medicine, New York, USA

• Rachel P. Wildman, Department of Epidemiology and Population Health, Albert Einstein College of Medicine, New York, USA

• Mark Woodward, Department of Medicine, Mount Sinai Medical Center, New York, USA

Abstract

Objective

The main aim is to determine which obesity index best discriminates cardiovascular risk factors (CRF).

Study Design and Setting

A meta-analysis was conducted using studies that employed receiver-operating characteristic (ROC) curve analysis.

• Inclusion Criteria: Studies involving ROC curve analysis published on obesity indices correlating with hypertension, type-2 diabetes, and/or dyslipidemia.

Results

• Key Finding: Body mass index (BMI) proved to be the least effective discriminator of cardiovascular risk factors compared to waist-to-height ratio (WHtR), which performed best across both sexes.

• Pooled AUC Values: WHtR showed an area under the curve (AUC) of $(0.67 ext{ to } 0.73)$ in males and $(0.68 ext{ to } 0.76)$ in females for various cardiovascular risk factors.

Conclusion

The evidence strongly supports using measures of central obesity, particularly WHtR, over BMI in detecting cardiovascular risk factors.

Introduction

Background

• A growing interest has emerged regarding which measure of overweight and obesity can best indicate those at increased cardiovascular risk.

• Body mass index (BMI) is traditionally endorsed by the World Health Organization to classify obesity severity but has limitations.

• Measures of central adiposity like waist circumference (WC) and waist-to-hip ratio (WHR) are increasingly recognized as superior predictors of obesity-related cardiovascular risk.

Purpose

This meta-analysis aims to systematically compare the ability of four indices of obesity—BMI, WC, WHR, and WHtR—in identifying major cardiovascular risk factors: hypertension, type-2 diabetes, and dyslipidemia.

Methodology

2.1 Study Identification

• Database: MEDLINE

• Date Range: Studies published from 1966 to December 1, 2006.

• Keywords Used: "receiver operating characteristic curve," "anthropometry," "diabetes," "hypertension," "dyslipidaemia," "obesity," "body mass index," "waist circumference," "waist-hip ratio," and "waist-height ratio."

2.2 Inclusion Criteria

• Studies involving adults (aged over 18 years) with ROC analysis comparing BMI, WC, WHR, and WHtR against at least one cardiovascular risk factor (hypertension, type-2 diabetes, dyslipidemia) were included.

• Studies were incorporated regardless of measurement method differences for WC.

2.3 Data Analysis

• The AUC for each obesity index for the selected risk factors was pooled using a random-effects model.

• AUC Interpretation:

  • An AUC of 1 denotes perfect discrimination.

  • An AUC of 0.5 indicates discrimination equal to chance.

• Analyses were conducted separately for males and females and tested for statistical heterogeneity.

Study Selection

• Out of 25 studies identified, 10 studies met inclusion criteria; these were predominantly cross-sectional studies conducted between 1990 and 2004.

• Sample sizes varied widely, accumulating data from a total of 88,514 subjects, of which 54% were female.

Results

3.1 Hypertension

• Eight studies addressed hypertension outcomes.

• Pooled AUC Values:

  • WHtR: $(0.68)$ for males vs. $(0.64)$ for BMI

  • WHtR: $(0.73)$ for females vs. $(0.69)$ for BMI

• WHtR demonstrated statistically significant superiority over BMI in males (P = 0.04).

3.2 Type-2 Diabetes

• Nine studies provided data on type-2 diabetes.

• Pooled AUC Findings:

  • Highest AUC for WHtR: $(0.73)$ for males and $(0.76)$ for females

  • Lowest AUC for BMI: $(0.67)$ for males and $(0.69)$ for females

• Statistically significant differences noted between BMI and WHtR (P < 0.01) only in males.

3.3 Dyslipidemia

• Seven studies focused on dyslipidemia.

• WHtR again showed the highest pooled AUC (0.67 for males; 0.68 for females), but the differences were not statistically significant.

Discussion

• This meta-analysis encompassing over 88,000 individuals affirms the advantages of WHtR for assessing cardiovascular risk compared to BMI.

• While BMI fails to account for fat distribution, WHtR reflects central fat deposition which is more closely associated with cardiovascular risk.

• Previous assumptions suggested that combining BMI with WC could enhance predictive capabilities; however, this analysis found that the combination did not improve discrimination.

• Variations noted in discriminative capabilities between sexes with females demonstrating higher AUC on average compared to males.

Implications for Practice

• The study suggests adopting WHtR as a preferred measure of obesity due to its practicality (only requiring a tape measure) and its proven efficacy in predictive capabilities for cardiovascular risks.

Conclusion and Future Directions

• While the results underscore WHtR's effectiveness, the small differences detected in discriminative power may lack clinical significance.

• Further studies are warranted to confirm whether these findings extend to the prediction of hard cardiovascular disease endpoints and across different ethnic populations.

Acknowledgments

• Acknowledgment for contributions from Gary T.C. Ko, Cuong Q. Tran, and Harald J. Schneider.

• Funding support for authors from respective scholarships and fellowships.

References

[List of references as provided in the original transcript included here]