The metabolically healthy obese phenotype as a paradigm to clarify the relationship between obesity and disease

Open Access
- Author:
- Durward, Carrie Margaret
- Graduate Program:
- Nutrition
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- August 08, 2012
- Committee Members:
- Sharon Nickols Richardson, Dissertation Advisor/Co-Advisor
Penny Margaret Kris Etherton, Committee Member
Terryl Johnson Hartman, Committee Member
Connie Jo Rogers, Committee Member
Eric Loken, Committee Member - Keywords:
- health
Metabolic health
obese
obesity
overweight
weight loss
survival analysis
total mortality
diet - Abstract:
- Although obesity is associated with increased risk of chronic disease (cardiovascular disease, diabetes, some cancers) and mortality, a subset of the obese population presents without clinical precursors for these diseases. This phenotype is termed Metabolically Healthy Obese (MHOB). Research in this area has the potential to clarify the relationship between excess adiposity and health outcomes. However, there are still many unanswered questions about this phenotype, including: 1) how to define metabolic health, 2) whether these individuals are protected from disease risk or if it is just delayed, 3) if MHOB should be encouraged to lose weight, and 4) if differences in dietary intake could provide an explanation for the cause of their relatively healthy cardiometabolic profile. The primary objective of this work was to contribute to answering these essential ques-tions. Previous research to determine if MHOB participants are protected from the in-creased disease and mortality risk usually associated with obesity is equivocal. Chapter three examined all-cause mortality risk associated with different body mass index (BMI) and metabolic health phenotypes using the National Health and Nutrition Examination Survey (NHANES)-III and the linked mortality file. Exclusion criteria included pregnan-cy, breastfeeding, incomplete data for variables of interest, BMI <18.5, and age ≥60 years (n=4373, mean±SD, age 37.1±10.9 years, BMI 27.3±5.8 kg/m2, 49.4% female). Metabolic health was defined using several definitions to examine the effect on prevalence and results. The definitions included: 1) homeostatic model assessment insulin resistance (HOMA-IR) <2.5; 2) ≤2 Adult Treatment Panel (ATP)-III Metabolic Syndrome Criteria, and 3) a combined definition using metabolic criteria and insulin resistance. Cox propor-tional hazard regression models were used to estimated hazard ratios and 95% confidence intervals, adjusted for potential confounding variables. The three definitions resulted in different samples. There was little overlap in classification, with only 40 of 1160 obese participants classified as metabolically healthy by all three definitions. MHOB partici-pants were not at significantly increased risk of all-cause mortality by either the HOMA-IR or ATP-III definitions. In contrast, when metabolic health was defined using the combined definition, MHOB were at 165% elevated risk of mortality when compared to the healthy lean reference group (hazard ratio 2.65, 95% confidence interval 1.34-5.24). In conclusion, MHOB individuals risk of all-cause mortality varies depending on which definition of metabolic health is used. Current public health recommendations encourage all people with obesity to lose weight for the purpose of improving health. However, some researchers and clinicians question whether weight loss is necessary for MHOB individuals, and in one study weight loss actually caused increased insulin resistance. To examine the effect of weight loss in MHOB, the fourth chapter is a secondary analysis of a combined diet- and exer-cise-induced weight-loss intervention in overweight/obese premenopausal women with varying levels of metabolic health. Participants (n=80 completers) were asked to reduce calories and add 30-40 minutes of walking to their daily routine. This six-month interven-tion also included weekly nutrition education classes. As in the previous study, multiple definitions (the same three listed above) of metabolic health were used to allow for better comparability with the literature, and to identify the effects of multiple definitions on the results of this analysis. In comparison to the previous work, in this sample, the different definitions resulted in samples with a larger degree of overlap; 24% of women were identified as healthy by all three definitions and 64% of obese women were classified as metabolically healthy by two or more definitions. However, there was also a high rate of misclassification between definitions, 66% of participants classified as both metabolically healthy and unhealthy. The lifestyle intervention resulted in significant weight and fat loss in all participants, regardless of metabolic health status or definition used. There were no negative effects of weight loss on any of the health markers measured. The MHOB group improved both anthropometric markers of adiposity and some health markers with weight loss. These results indicate that MHOB individuals will benefit from diet- and exercise-induced weight loss. In the search for what makes these individuals different, few studies have examined dietary intake in MHOB participants. Previous studies have only examined macronutrient content, finding few differences between metabolic health and BMI categories. Chapter five examines the relationship between dietary intake (nutrients and food group servings) and metabolic health and BMI status in NHANES 1999-2004 data. Metabolic health was defined as the absence of both insulin resistance (HOMA-IR <2.5) and metabolic syn-drome by the ATP-III criteria. Univariate linear regression models as well as multivariate models adjusted for potential confounders were used to identify significant differences in food group and nutrient intakes between metabolic health groups. The most robust finding was lower intake of total and red meat servings in MHOB compared to metaboli-cally unhealthy obese (MUOB) participants in both the univariate and multivariate models. These results suggest that red meat intake is related to metabolic health status in obese participants. Overall, these results contribute to the understanding of metabolic health in obese participants. Primary conclusions include the fact that the risk of mortality in this population is dependent on the definition of metabolic health used, and that exercise- and diet- induced weight loss produced positive anthropometric and health changes in a cohort of premenopausal women. The results presented here indicate that MHOB individuals have an intermediate metabolic health profile, healthier than MUOB participants, but more abnormal than healthy participants with lower BMIs (MHLN in chapter five and MHOW in chapter two). Based on the inconclusive nature of the literature regarding whether these individuals are protected from risk, the intermediate profile of their metabolic health profile, and the positive effects of the lifestyle intervention, MHOB individuals should be encouraged to lose weight through diet and exercise.