0

The full content of Annals is available to subscribers

Subscribe/Learn More  >
Reviews |

Are Metabolically Healthy Overweight and Obesity Benign Conditions?: A Systematic Review and Meta-analysis

Caroline K. Kramer, MD, PhD; Bernard Zinman, CM, MD; and Ravi Retnakaran, MD
[+] Article and Author Information

From Leadership Sinai Centre for Diabetes, Mount Sinai Hospital; University of Toronto; and Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, Ontario, Canada.

Grant Support: By intramural funds from the Leadership Sinai Centre for Diabetes. Dr. Kramer holds a Canadian Diabetes Association Postdoctoral Fellowship Award. Dr. Zinman holds the Sam and Judy Pencer Family Chair in Diabetes Research at Mount Sinai Hospital and University of Toronto. Dr. Retnakaran holds an Ontario Ministry of Research and Innovation Early Researcher Award.

Potential Conflicts of Interest: None disclosed. Forms can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M13-1059.

Requests for Single Reprints: Ravi Retnakaran, MD, Leadership Sinai Centre for Diabetes, Mount Sinai Hospital, 60 Murray Street, Suite L5-025, Mailbox-21, Toronto, Ontario M5T 3L9, Canada; e-mail: rretnakaran@mtsinai.on.ca.

Current Author Addresses: Dr. Kramer: Leadership Sinai Centre for Diabetes, Mount Sinai Hospital, 60 Murray Street, Suite L5-009, Mailbox-21, Toronto, Ontario M5T 3L9, Canada.

Dr. Zinman: Leadership Sinai Centre for Diabetes, Mount Sinai Hospital, 60 Murray Street, Suite L5-024, Mailbox-17, Toronto, Ontario M5T 3L9, Canada.

Dr. Retnakaran: Leadership Sinai Centre for Diabetes, Mount Sinai Hospital, 60 Murray Street, Suite L5-025, Mailbox-21, Toronto, Ontario M5T 3L9, Canada.

Author Contributions: Conception and design: C.K. Kramer, B. Zinman, R. Retnakaran.

Analysis and interpretation of the data: C.K. Kramer, B. Zinman, R. Retnakaran.

Drafting of the article: C.K. Kramer.

Critical revision of the article for important intellectual content: C.K. Kramer, B. Zinman, R. Retnakaran.

Final approval of the article: C.K. Kramer, B. Zinman, R. Retnakaran.

Statistical expertise: C.K. Kramer.

Obtaining of funding: B. Zinman.

Administrative, technical, or logistic support: B. Zinman.

Collection and assembly of data: C.K. Kramer, R. Retnakaran.


Ann Intern Med. 2013;159(11):758-769. doi:10.7326/0003-4819-159-11-201312030-00008
Text Size: A A A

Background: Recent interest has focused on a unique subgroup of overweight and obese individuals who have normal metabolic features despite increased adiposity. Normal-weight individuals with adverse metabolic status have also been described. However, it remains unclear whether metabolic phenotype modifies the morbidity and mortality associated with higher body mass index (BMI).

Purpose: To determine the effect of metabolic status on all-cause mortality and cardiovascular events in normal-weight, overweight, and obese persons.

Data Sources: Studies were identified from electronic databases.

Study Selection: Included studies evaluated all-cause mortality or cardiovascular events (or both) and clinical characteristics of 6 patient groups defined by BMI category (normal weight/overweight/obesity) and metabolic status (healthy/unhealthy), as defined by the presence or absence of components of the metabolic syndrome by Adult Treatment Panel III or International Diabetes Federation criteria.

Data Extraction: Two independent reviewers extracted the data. Metabolically healthy people of normal weight made up the reference group.

Data Synthesis: Eight studies (n = 61 386; 3988 events) evaluated participants for all-cause mortality and/or cardiovascular events. Metabolically healthy obese individuals (relative risk [RR], 1.24; 95% CI, 1.02 to 1.55) had increased risk for events compared with metabolically healthy normal-weight individuals when only studies with 10 or more years of follow-up were considered. All metabolically unhealthy groups had a similarly elevated risk: normal weight (RR, 3.14; CI, 2.36 to 3.93), overweight (RR, 2.70; CI, 2.08 to 3.30), and obese (RR, 2.65; CI, 2.18 to 3.12).

Limitation: Duration of exposure to the metabolic–BMI phenotypes was not described in the studies and could partially affect the estimates.

Conclusion: Compared with metabolically healthy normal-weight individuals, obese persons are at increased risk for adverse long-term outcomes even in the absence of metabolic abnormalities, suggesting that there is no healthy pattern of increased weight.

Primary Funding Source: Intramural funds from the Leadership Sinai Centre for Diabetes.

Figures

Grahic Jump Location
Appendix Figure 1.

Summary of evidence search and selection.

BMI = body mass index.

Grahic Jump Location
Grahic Jump Location
Appendix Figure 2.

Prevalence of metabolically healthy and unhealthy individuals in normal-weight, overweight, and obese groups.

Grahic Jump Location
Grahic Jump Location
Figure 1.

Meta-analyses of metabolically healthy body mass index categories for the risk for all-cause mortality and cardiovascular events compared with metabolically healthy normal-weight persons (reference).

A. Metabolically healthy overweight group. B. Metabolically healthy overweight group, including only studies with at least 10 y of follow-up. C. Metabolically healthy obese group. D. Metabolically healthy obese group, including only studies with at least 10 y of follow-up. CV = cardiovascular.

Grahic Jump Location
Grahic Jump Location
Figure 2.

Meta-analyses of metabolically unhealthy body mass index categories for the risk for all-cause mortality and cardiovascular events compared with metabolically healthy normal-weight persons (reference).

A. Metabolically unhealthy normal-weight group. B. Metabolically unhealthy overweight group. C. Metabolically unhealthy obese group. CV = cardiovascular.

Grahic Jump Location
Grahic Jump Location
Appendix Figure 3.

Meta-analyses of unhealthy normal-weight phenotype for the risk for all-cause mortality and cardiovascular events compared with metabolically unhealthy obese (A) and metabolically unhealthy overweight (B) persons.

CV = cardiovascular.

Grahic Jump Location
Grahic Jump Location
Figure 3.

Meta-analyses of various clinical characteristics, by metabolic–body mass index categories.

Data shown as weighted mean difference compared with metabolically healthy normal-weight persons (reference). To convert cholesterol, triglyceride, and glucose values to traditional units (mg/dL), divide by 0.0259, 0.0113, and 0.0555, respectively. HOMA-IR = Homeostasis Model Assessment of Insulin Resistance.* P < 0.05.

Grahic Jump Location

Tables

References

Letters

NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Comments

Submit a Comment
Comment
Posted on December 6, 2013
Carmine Finelli, MD, 2Giovanni Tarantino, MD
Stella Maris Mediterraneum Foundation, University of naples
Conflict of Interest: None Declared
Kramer et al in their meta-analysis (1) purposed “to determine the effect of metabolic status on all-cause mortality and cardiovascular events in normal-weight, overweight, and obese persons”. The limitation of this study is the “duration of exposure to the metabolic-BMI phenotypes was not described in the studies and could partially affect the estimates”. However, they concluded that “compared with metabolically healthy normal – weight individuals, obese persons are at increased risk for adverse long-term outcomes even in the absence of metabolic abnormalities, suggesting that there is no healthy pattern of increased weight” (1).
Nevertheless, it is well-known that the risk of death from all causes increases throughout the range of moderate and severe increases in body weight for both men and women in all age groups. Interventions at home, in the office, and in the community are required to empower adults to increase physical activity and to modify eating habits. Obviously, before these intervention strategies are set up, it is mandatory to establish exactly which individuals should be treated and when.
It has recently (2) been postulated that “metabolically benign obesity” exists, and is not accompanied by insulin resistance and early atherosclerosis. Could it be the case for some overweight/obese subjects? In such direction, our group has sought to identify which parameters could distinguish between healthy and non healthy overweight/obese individuals.
We propose two imaging parameters (ultrasound), i.e., hepatic steatosis and spleen longitudinal diameter (3). The use of simple parameters in assessing the need for medical intervention with respect to healthy and non healthy overweight/obese individuals could not only reduce the cost of medical care but also provide more reliable identification of patients in need of weight loss. In other words, the presence and severity of hepatic steatosis, mainly adjusted for gamma-GT values (4), could be the discriminating point in intervention strategies.
There is not the “metabolically benign obesity” — there is just a complicated situation, indeed.
Obesity generally predisposes to developing a number of health problems. In some cases patients who are obese have a longevity advantage over patients who are not. Again, could they represent the metabolically healthy ones?
. Coming back to body weight gain, our group has recently emphasized (5) the consistent relationship of visceral fat, strictly associated with hepatic steatosis, to disease and mortality risk in humans, the distinct metabolic capacity of visceral fat, the importance of accounting for body fat distribution in disease risk, and visceral fat depletion as a potential treatment strategy to prevent or delay age-related diseases and to improve longevity.

1Carmine Finelli, MD, PhD and 2Giovanni Tarantino, MD
1Center of Obesity and Eating Disorders, Stella Maris Mediterraneum Foundation, C/da S. Lucia 80035, Chiaromonte, Potenza (Italy)
2Department of Clinical Medicine and Surgery, Federico II University Medical School of Naples, Italy and INT "Fondazione Pascale" - Cancer Research Center of Mercogliano 83013
Mercogliano (AV), Italy
Address to correspondence
tarantin@unina.it

References
1. Kramer CK, Zinman B, Retnakaran R. Are Metabolically Healthy Overweight and Obesity Benign Conditions?:A Systematic Review and Meta-analysis Ann Intern Med. 2013; 159 (11): 758-769.
2. de Gusmão Correia ML. Is 'metabolically healthy' obesity a benign condition? J Hypertens. 2013 Jan;31(1):39-41.
3. Tarantino G, Colicchio P, Conca P, Finelli C, Di Minno MN, Tarantino M, Capone D, Pasanisi F. Young adult obese subjects with and without insulin resistance: what is the role of chronic inflammation and how to weigh it non-invasively? J Inflamm (Lond). 2009 Mar 16; 6:6.
4. Tarantino G, Finelli C, Colao A, Capone D, Tarantino M, Grimaldi E, Chianese D, Gioia S, Pasanisi F, Contaldo F, Scopacasa F, Savastano S.Are hepatic steatosis and carotid intima media thickness associated in obese patients with normal or slightly elevated gamma-glutamyl-transferase?J Transl Med. 2012 Mar 16.
5. Finelli C, Sommella L, Gioia S, La Sala N, Tarantino G. Should visceral fat be reduced to increase longevity? Ageing Res Rev. 2013 Jun 11.

Metabolically Healthy but Obese Individuals do Exist
Posted on December 9, 2013
Jean-Philippe Chaput, Ph.D., Arya M. Sharma, M.D., Ph.D.
University of Alberta
Conflict of Interest: None Declared
To the Editor:
In their systematic review and meta-analysis examining the effect of metabolic status on all-cause mortality and cardiovascular events in normal-weight, overweight, and obese individuals, Kramer and colleagues (1) concluded that “there is no healthy pattern of increased weight”. It is, however, worth noting that all of the studies included in this analysis define unhealthy obesity as having at least two or more components of the metabolic syndrome. Thus, by definition, individuals with just one risk factor (such as hypertension, impaired fasting glucose or hypercholesterolemia) would be considered “metabolically healthy”. It is therefore not surprising that the authors find elevated cardiovascular risk in this group. In our opinion, this meta-analysis simply adds to the confusion on the topic by defining “healthy obesity” with criteria that we would consider anything but healthy.

Previous studies, that have used a far more stringent definition of “healthy obesity”, defined as the absence of any medical, mental or functional risk factors or limitations associated with excess weight (Edmonton Obesity Staging System Stage 0) (2-4), show virtually no increased mortality risk in overweight and obese individuals even over a 200-month follow-up period. In contrast, obese individuals with even just one metabolic or other risk factor are considered to have Stage 1 or 2 and have clearly elevated mortality risk (3,4).

Mislabelling “unhealthy obese individuals” as supposedly “healthy” further promotes weight bias and reinforces the widespread misconception that health can be measured simply by stepping on a scale. Rather, we should not assume that everyone with excess body fat is at high-risk and in immediate need of treatment. Instead, we should look at individual risk factors (smoking, high blood pressure, dysglycemia, dyslipidemia etc.) to determine if the given individual needs medical attention. After all, one size does not fit all!

References

1. Kramer CK, Zinman B, Retnakaran R. Are metabolically healthy overweight and obesity benign conditions? A systematic review and meta-analysis. Ann Intern Med. 2013;159:758-69.
2. Sharma AM, Kushner RF. A proposed clinical staging system for obesity. Int J Obes (Lond). 2009;33:289-95.
3. Padwal RS, Pajewski NM, Allison DB, Sharma AM. Using the Edmonton obesity staging system to predict mortality in a population-representative cohort of people with overweight and obesity. CMAJ 2011;183:E1059-66.
4. Kuk JL, Ardern CI, Church TS, Sharma AM, Padwal R, Sui X, et al. Edmonton Obesity Staging System: association with weight history and mortality risk. Appl Physiol Nutr Metab. 2011;36:570-6.
Cardiovascular Risk and Inflammation in Metabolically Healthy Obesity
Posted on December 18, 2013
Nathalie Esser, MD ; André J. Scheen, MD, PhD ; Nicolas Paquot, MD, PhD
Department of Diabetes, Nutrition and Metabolic Disorders, University Hospital of Liege, Belgium
Conflict of Interest: None Declared
Kramer and colleagues (1) reported that obese individuals are at risk for cardiovascular events and total mortality over the long term regardless of the metabolic status. The discrimination of participants according to a metabolic score is an oversimplified dichotomous classification, which does not reflect the huge heterogeneity of obesity. Metabolically healthy individuals, as classically defined, may still have metabolic disorders. In Kramer’s meta-analysis, metabolically healthy obese persons had an intermediate risk for total mortality and cardiovascular events, higher than metabolically healthy lean/overweight individuals, but lower than metabolically unhealthy overweight/obese individuals. However, the metabolically healthy obese group was only at increased risk in the studies with more than 10 years of follow-up. These results might be influenced by variation within some metabolic syndrome components over time. Indeed, metabolically healthy obesity is a transient state for one-third of patients who may progress to metabolic risk and type 2 diabetes (2). Kramer and colleagues speculate that metabolically healthy obese persons probably have subclinical levels of risk factors that worsen overtime (1).
A pro-inflammatory state is recognized as an important component of metabolic syndrome, which is mainly associated with abdominal obesity. We recently showed that, compared to metabolically healthy obese individuals, metabolically unhealthy obese individuals have a less favorable inflammatory profile in their visceral adipose tissue, which results from the infiltration by pro-inflammatory adipose tissue macrophages with an increased NLR family pyrin domain containing-3 inflammasome activity and interleukin-1β production. Furthermore, metabolically unhealthy obesity was associated with a decreased number of anti-inflammatory T regulatory lymphocytes in visceral adipose tissue (3). Interestingly, Wildman and colleagues (4) reported that overweight/obese women without clustering of cardiometabolic risk factors still possess abnormal levels of inflammatory markers. Similarly in our study, metabolically healthy obese persons had an intermediate inflammatory pattern in their visceral adipose tissue concerning interleukin-1β production and gene expression, lower than metabolically unhealthy obese individuals but higher than metabolically healthy lean individuals (3), a finding in agreement with the data presented by Kramer and colleagues regarding the mortality and cardiovascular events (1).
Chronic silent inflammation may contribute to the elevated risk of cardiovascular events and overall mortality (1, 5) and may be a potential mechanism linking abdominal obesity and cardiovascular risk. Differences in the visceral adipose tissue inflammatory pattern may explain why metabolically healthy obese persons still have an elevated cardiovascular risk, intermediate between that of metabolically healthy lean/overweight individuals and that of metabolically unhealthy overweight/obese persons.

REFERENCES
1. Kramer CK, Zinman B, Retnakaran R. Are metabolically healthy overweight and obesity benign conditions?: A Systematic Review and Meta-analysis. Ann Intern Med. 2013;159(11):758-69.
2. Appleton SL, Seaborn CJ, Visvanathan R, Hill CL, Gill TK, Taylor AW, et al. Diabetes and cardiovascular disease outcomes in the metabolically healthy obese phenotype: a cohort study. Diabetes Care. 2013;36(8):2388-94.
3. Esser N, L'Homme L, De Roover A, Kohnen L, Scheen AJ, Moutschen M, et al. Obesity phenotype is related to NLRP3 inflammasome activity and immunological profile of visceral adipose tissue. Diabetologia. 2013;56(11):2487-97.
4. Wildman RP, Kaplan R, Manson JE, Rajkovic A, Connelly SA, Mackey RH, et al. Body size phenotypes and inflammation in the Women's Health Initiative Observational Study. Obesity (Silver Spring). 2011;19(7):1482-91.
5. Hinnouho GM, Czernichow S, Dugravot A, Batty GD, Kivimaki M, Singh-Manoux A. Metabolically healthy obesity and risk of mortality: does the definition of metabolic health matter? Diabetes Care. 2013;36(8):2294-300.
Comment
Posted on December 19, 2013
Juhee Cho, PhD, Yoosoo Chang, MD, Seungho Ryu, MD, PhD
Johns Hopkins School of Public Health
Conflict of Interest: None Declared
We read with interest the meta-analysis of Kramer et al. on the health consequences of metabolically healthy overweight and obesity (1), but we were very surprised that the authors used crude, unadjusted data from individual studies for their pooled estimates. In the Discussion, the authors acknowledge this limitation and indicate that smoking and physical activity could confound the observed associations. While we agree with this statement, the authors do not mention that age and sex are also important confounders, as both body mass index and the rate of mortality and cardiovascular disease are deeply influenced by age and sex. The effects of confounding by all these factors can be substantial, and we believe that evaluation of the association between body mass index and mortality or cardiovascular disease is only meaningful after adjustment at least for age, sex, and smoking. Furthermore, meta-regression analyses cannot overcome the limitations introduced by lack of within-study adjustment (2).

In addition to the use of unadjusted measures of association, we were concerned by the analyses restricted to studies with more than 10 years of follow-up, as these subgroup analyses may also be affected by ecological bias (2). Overall, we believe that the conclusions of the paper, based on crude data and across-study subgroup comparisons, do not provide solid evidence on the effect of body mass index categories by metabolic health status on mortality and cardiovascular endpoints, which is still an open question.

Juhee Cho, PhD, Yoosoo Chang, MD, Seungho Ryu, MD, PhD


1. Kramer CK, Zinman B, Retnakaran R. Are metabolically overweight and obese benign conditions? A systematic review and meta-analysis. Ann Intern Med 2013:159:758-69.

2. Berlin JA, Santanna J, Schmid CH, et al. Individual patient- versus group-level data meta-regressions for the investigation of treatment effect modifiers: ecological bias rears its ugly head. Stat Med. 2002;21:371-87.


Juhee Cho, PhD
Department of Health Sciences and Technology, Samsung Advanced Institute for Health Sciences and Technology, Sungkyunkwan University, Seoul, South Korea and Departments of Epidemiology and Health, Behavior, and Society, Johns Hopkins University Bloomberg School of Public Health, Baltimore, MD.

Yoosoo Chang, MD, and Seungho Ryu, MD, PhD
Center for Cohort Studies, Total Healthcare Center, Kangbuk Samsung Hospital, Sungkyunkwan University, School of Medicine, Seoul, Korea, and Department of Occupational and Environmental Medicine, Kangbuk Samsung Hospital, Sungkyunkwan University, School of Medicine, Seoul, Korea.

Metabolically healthy obesity – what exactly are we talking about?
Posted on December 26, 2013
Dorit Samocha-Bonet, Antony D Karelis and Remi Rabasa-Lhoret
(1) Diabetes & Obesity Research Program, Garvan Institute of Medical Research and Faculty of Medicine, University of New South Wales, Sydney, Australia (2) Département of Kinanthropology, Un
Conflict of Interest: None Declared
Kramer and colleagues in their meta-analysis concluded that “there is no healthy pattern of increased weight” (1). Indeed, their data suggest that metabolically healthy obese individuals have increased cardiovascular and all-cause mortality risk compared with their normal weight counterparts. Notably however, the magnitude of the RR reported is approximately half of that reported for the metabolically unhealthy obese group (1.24 vs. 2.65) (1), suggesting that metabolically healthy obese individuals are relatively protected from some adverse outcomes of obesity.
Cross-sectional studies suggest that metabolic health in obesity is associated with “healthy” adipose tissue, capable of storing fat away from insulin sensitive tissues, primarily liver (2, 3). Longitudinal studies, with long term follow up periods, are vital to determine whether a metabolically healthy obesity snapshot is stable and whether it ultimately translates to decreased cardiovascular disease, type 2 diabetes, cancer and all-cause mortality risk at older age. The main obstacle in advancing our understanding of the phenotype and its metabolic risks is the inconsistent definition of metabolic health across studies, showcased in the meta-analysis with definitions based on either absence of the metabolic syndrome (defined by 2 sets of criteria including ATP III, IDF or versions thereof) or <2 abnormalities (1). Definition of metabolic health based on different criteria results in wide range of prevalence reported for the phenotype, ranging from less than 10 to almost 50 percent of the obese population (2, 3). We and others have reported significant differences in metabolic characteristics (4) and all-cause mortality (5) between metabolically healthy obese individuals defined based on different sets of criteria. Furthermore, while inclusion of metabolically unhealthy normal weight group in the meta-analysis is important and emphasized the increased risk of metabolic abnormality at any BMI, some well-conducted longitudinal studies were excluded, leaving the meta-analysis with only 4 studies with >10 years of follow up.
In summary, standardization of the definition of metabolic health is required before the metabolically healthy obese phenotype is dismissed. Without being totally “healthy” some obese subjects still appear healthier than other.

1. Kramer CK, Zinman B, Retnakaran R. Are Metabolically Healthy Overweight and Obesity Benign Conditions?A Systematic Review and Meta-analysis. Ann Intern Med. 2013;159(11):758-69.
2. Samocha-Bonet D, Chisholm DJ, Tonks K, Campbell LV, Greenfield JR. Insulin-sensitive obesity in humans - a 'favorable fat' phenotype? Trends Endocrinol Metab. 2012;23(3):116-24.
3. Primeau V, Coderre L, Karelis AD, Brochu M, Lavoie ME, Messier V, et al. Characterizing the profile of obese patients who are metabolically healthy. Int J Obes. 2011;35(7):971-81.
4. Messier V, Karelis AD, Prud'homme D, Primeau V, Brochu M, Rabasa-Lhoret R. Identifying metabolically healthy but obese individuals in sedentary postmenopausal women. Obesity. 2010;18(5):911-7.
5. Hinnouho GM, Czernichow S, Dugravot A, Batty GD, Kivimaki M, Singh-Manoux A. Metabolically healthy obesity and risk of mortality: does the definition of metabolic health matter? Diabetes Care. 2013;36(8):2294-300.
Results are not adjusted for age, sex or smoking and thus differ from published results in the studies being reviewed
Posted on December 24, 2013
Katherine M. Flegal
National Center for Health Statistics, Centers for Disease Control and Prevention
Conflict of Interest: None Declared
Kramer et al (1) performed a systematic review and meta-analysis that compared relative risks of mortality or CVD events by metabolic status and BMI categories. In their meta-analysis, Kramer et al took data from observational studies, calculated new unadjusted relative risks and summarized the unadjusted risks. The studies that they summarized had all published relative risks that were adjusted for confounding factors such as age and sex. Kramer et al. did not use those published relative risks but instead calculated new unadjusted relative risks for each study from counts of sample sizes and events. As a result, the unadjusted relative risks for individual studies in the article by Kramer et al do not match the published relative risks in the original studies.
The differences are sometimes large. For example, Kramer et al. included results from a study by Kuk and Ardern (2) that had used data from the Third National Health and Nutrition Examination Survey, a complex sample survey. The original publication presented hazard ratios that were adjusted for age, sex, income, ethnicity, smoking status, and alcohol consumption and that incorporated sample weights. Kuk and Ardern found relative risks of 1.24 for metabolically unhealthy normal weight participants and 0.45 for metabolically healthy overweight participants, relative to metabolically healthy normal weight participants. The corresponding unadjusted relative risks for that study provided by Kramer et al were 2.52 and 0.96, but those results are not adjusted for the complex sample design or for any confounding factors, not even for age.
Kramer et al list the use of pooled unadjusted estimates as one limitation of their study and note that their procedure did not account for physical activity or smoking, but do not mention the potentially more serious limitation of failing to adjust for age and sex. The procedure followed by Kramer et al is more appropriate for summaries of randomized trials than for observational studies. The lack of adjustment for confounding by age, sex, smoking or other factors makes the summary unadjusted results from Kramer et al difficult to interpret. A summary of adjusted risks, such as that previously provided by Fan et al (3) on the same topic, could be useful.


1. Kramer CK, Zinman B, Retnakaran R. Are Metabolically Healthy Overweight and Obesity Benign Conditions?: A Systematic Review and Meta-analysis. Annals of internal medicine. 2013;159(11):758-69.
2. Kuk JL, Ardern CI. Are metabolically normal but obese individuals at lower risk for all-cause mortality? Diabetes Care. 2009;32(12):2297-9.
3. Fan J, Song Y, Chen Y, Hui R, Zhang W. Combined effect of obesity and cardio-metabolic abnormality on the risk of cardiovascular disease: a meta-analysis of prospective cohort studies. International journal of cardiology. 2013;168(5):4761-8.

Metabolic Health and BMI Categories
Posted on December 30, 2013
Gerson T. Lesser, MD
Department of Medicine, NYU School of Medicine
Conflict of Interest: None Declared
Kramer and colleagues (1) and editorialists (2) suggest that "healthy obesity" is a myth. However, there are problems with the data and, more importantly, with the concepts. Kramer's two very large studies (3, 4) represent ~90% of subjects followed for 10 years; compared to metabolically healthy, normal weight subjects, one has a mortality rate just short of significance and one an RR of 1.00, hardly "strong evidence".

The fact that "obesity" has come to be defined as a higher BMI value, rather than as demonstrated excess body fat (WF/W) is misleading, and confuses the argument. The assumption that adiposity (increased WF/W above some given value) is a factor for health risk and probably for earlier mortality appears to underlie the present report. Unfortunately, the anthropometrically calculated BMI does not separate or define the components of its two-compartment model: lean body mass (LBM) and total body fat (WF). The authors' text tends to equate elevated BMI with "increased adiposity"; this incorrectly assumes LBM to be constant at any given height, so that any BMI difference is taken to represent a difference of WF/W. Employing a direct measurement of body fat, the LBM of individuals of a homogeneous cohort of similar height and gender varies at least +/- 20% from the group mean; and an individual's deviation from average weight at this height can be all WF, all LBM or any combination of the two (5). Consequently, for healthy individuals of identical height, weight (and identical BMI) such a +/-  20% range of LBM may be associated with a remarkable range of fat content (WF/W), roughly from 10 to 40% (calculated from ref 5).

This leads to great uncertainty in categorizing individuals for health risk within the commonly defined BMI groups. There can be an important number of those with high WF/W even within lower BMI groupings. Although the "overweight" or "obese" BMI groups likely include greater proportions of those with high WF/W, they also include some individuals of larger LBM but normal or low WF/W. Are some high-fat content individuals perhaps overrepresented as "metabolically unhealthy" within "normal" and "overweight" BMI categories and some with high LBM overrepresented as "metabolically healthy" within the "obese" BMI? Does LBM size also affect health risk? Would that alter the approach of Kramer and other investigators?

The 10 year interval seems minimal. Subjects (3,4) were largely in their mid-50s; death rates before 65 years are very low.

References:

1. Kramer CK, Zinman B, Retnakaran R. Are metabolically healthy overweight and obesity benign conditions? Ann Intern Med. 2013;159:758-69.

2. Hill JO, Wyatt HR. Ann Intern Med. 2013;159:789-90.

3. Song Y, Manson JE, Meigs JB, Ridker PM, Buring JE, Liu S. Comparison of usefulness of body mass index versus metabolic risk factors in predicting 10-year risk of cardiovascular events in women. Am J Cardiol. 2007;100:1654-8.

4. Ogorodnikova AD, Kim M, McGinn AP, Muntner P, Khan U, Wildman RP. Incident cardiovascular disease events in metabolically benign obese individuals. Obesity (Silver Spring). 2012;20:651-9.

5. Lesser GT, Deutsch S, Markofsky J. Use of an independent measurement of body fat to evaluate overweight and underweight. Metabolism 1971;20:792-804.
Author's Response
Posted on January 14, 2014
Caroline K. Kramer, MD, PhD, Bernard Zinman MD, Ravi Retnakaran MD
Mount Sinai Hospital/University of Toronto
Conflict of Interest: None Declared
We are thankful for the comments and interest in our systematic review and meta-analysis (1), in which the data of 8 studies were pooled to determine the effect of metabolic status on all-cause mortality and/or cardiovascular events in normal weight, overweight and obese individuals (n = 61,386; 3988 events). One of the reasons to perform a meta-analysis is to confirm the findings observed in smaller individual studies and document the magnitude of effect in a larger study population, in order to enable a more accurate estimate of effect. This approach is particularly important when evaluating a low/moderate-risk population (such as metabolically healthy obesity) because it enhances statistical power to detect differences in outcomes that could not be identified in individual studies. However, we do recognize that the pooling of several studies requires that their data be relatively homogeneous which partly limits the questions that can be answered by a single meta-analysis.
Chaput et al and Samocha-Bonet et al point to the controversies in the literature regarding the definition of metabolic health. In this context, although there currently is not a standard definition of metabolic health, the definition applied in this meta-analysis was that which was most prevalent in our systematic review of 1,443 studies comprising this literature. Of note, although it did not find differences in cardiovascular death, the study of Kuk et al that used a more stringent definition of health showed increased cancer mortality in healthy obese individuals, reinforcing the concept that there may not be a healthy pattern of obesity (2). In addition, the literature referred by Esser et al demonstrating increased inflammation in individuals with healthy obesity is concordant with this concept. Indeed, even when interpreted most conservatively, our meta-analyses clearly demonstrate the complexity of estimating an individual’s risk and that both metabolic status and adiposity should be taken into consideration.
In response to the comment by Lesser et al, it is relevant to reiterate that the statistical method used in our meta-analyses (random-effects model - profile likelihood method) is the most appropriate when risk estimates are close to non-significance as this approach better accounts for the imprecision in the estimate of between-study variance (3). Thus, even when this robust model was used, the healthy obese group had a relative risk (RR) of 1.24 which translates to a significant increase in absolute risk at a population level as discussed in our study. We agree, however, that the definition of obesity based on BMI does not take into consideration the distinction between lean body mass and body fat, which could further characterize the adiposity profile and have implications for outcomes.
We also recognize that pooling unadjusted estimates does not account for other covariates possibly associated with mortality. In this context, we agree with Cho et al and Flegal et al that age and gender are important covariates. However, we believe that this limitation does not invalidate our results for two reasons. First, the distribution of age was similar across the studies included in the meta-analysis (mostly middle-age participants ~45-55 years old), which reduces the confounding effect of age in the meta-analyses estimates; in addition, age did not differ between the BMI-metabolic groups in the majority of the included studies. Second, certain metabolic factor thresholds are gender-based in their definition (i.e. waist, HDL), which again reduces a possible gender confounder effect in our estimates. Thus, considering that the confounding effect of these covariates was partly muted and that the adjusted estimates shown in the studies used heterogeneous models that would make the pooling of adjusted data difficult to interpret, we used unadjusted estimates, while acknowledging their limitations.
With respect to the sub-group analyses that were restricted to studies with >10 years of follow-up (mentioned by Cho et al), this approach allows a longer time for the occurrence of events, which is the most appropriate strategy in evaluating a low/moderate-risk population. Ecological bias is possible in any sub-groups analyses; however, for the reasons noted earlier, we believe that this did not significantly impact our findings. Overall, there are still several unanswered questions with respect to the impact of obesity on health. In providing evidence that obesity should be recognized as harmful regardless of metabolic status, our meta-analysis ideally should lead to further studies of the long-term effects of excess adiposity and their implications for health.


References
1. Kramer CK, Zinman B, Retnakaran R. Are metabolically healthy overweight and obesity benign conditions? A systematic review and meta-analysis. Ann Intern Med. 2013;159:758-69.
2. Kuk JL, Ardern CI, Church TS, Sharma AM, Padwal R, Sui X, et al. Edmonton Obesity Staging System: association with weight history and mortality risk. Appl Physiol Nutr Metab. 2011;36:570-6.
3. Hardy RJ, Thompson SG. A likelihood approach to meta-analysis with random effects. Stat Med. 1996;15:619-29.
Submit a Comment

Summary for Patients

Is Obesity Harmful If Metabolic Factors Are Normal?

The full report is titled “Are Metabolically Healthy Overweight and Obesity Benign Conditions? A Systematic Review and Meta-analysis.” It is in the 3 December 2013 issue of Annals of Internal Medicine (volume 159, pages 758-769). The authors are C.K. Kramer, B. Zinman, and R. Retnakaran.

Read More...

Clinical Slide Sets

Terms of Use

The In the Clinic® slide sets are owned and copyrighted by the American College of Physicians (ACP). All text, graphics, trademarks, and other intellectual property incorporated into the slide sets remain the sole and exclusive property of the ACP. The slide sets may be used only by the person who downloads or purchases them and only for the purpose of presenting them during not-for-profit educational activities. Users may incorporate the entire slide set or selected individual slides into their own teaching presentations but may not alter the content of the slides in any way or remove the ACP copyright notice. Users may make print copies for use as hand-outs for the audience the user is personally addressing but may not otherwise reproduce or distribute the slides by any means or media, including but not limited to sending them as e-mail attachments, posting them on Internet or Intranet sites, publishing them in meeting proceedings, or making them available for sale or distribution in any unauthorized form, without the express written permission of the ACP. Unauthorized use of the In the Clinic slide sets will constitute copyright infringement.

Toolkit

Buy Now

to gain full access to the content and tools.

Want to Subscribe?

Learn more about subscription options

Advertisement
Related Articles
Related Point of Care
Topic Collections
PubMed Articles
Forgot your password?
Enter your username and email address. We'll send you a reminder to the email address on record.
(Required)
(Required)