- Top of page
OBJECTIVES: To evaluate the correlation between body mass index (BMI), body composition, and all-cause mortality in an elderly Asian population.
DESIGN: A prospective observational cohort study with 3.5-year follow-up.
SETTING: The Korean Longitudinal Study on Health and Aging Project for elderly residents in Seongnam City, Korea.
PARTICIPANTS: Eight hundred seventy-seven subjects aged 65 and older for whom baseline body composition data was available.
MEASUREMENTS: BMI, waist circumference, and body composition of each subject was evaluated. Body composition was examined using bioelectrical impedance analyses of measures, including lean mass (kg), fat mass (kg), and fat proportion (%). In addition, lean mass index (LMI, kg/m2) was calculated by dividing lean mass by the square of height. Participants were divided into three groups: Group 1 (<25th percentile), Group 2 (25–75th percentiles), and Group 3 (≥75th percentile) for BMI, waist circumference, body composition, and LMI.
RESULTS: In the fully adjusted Cox proportional hazard model, BMI, waist circumference, and fat composition were not correlated with mortality, but higher lean mass and LMI were considered predictors of lower mortality when comparing Group 3 and Group 1 (in lean mass, relative risk reduction of 84%, 95% confidence interval (CI)=45–96%, P=.004; in LMI, relative risk reduction of 69%, 95% CI=12–89%, P=.03).
CONCLUSION: The present study indicates that the recommendation of low BMI as a means of obtaining a survival advantage in the elderly is not supported. Instead, higher lean mass and higher LMI are associated with better survival in the elderly Asian population.
Obesity is an important risk factor for diabetes mellitus, cardiovascular disease, chronic kidney disease (CKD), and mortality.1–4 Accordingly, the World Health Organization (WHO) recommends the screening and treatment of obesity to prevent further morbidity. Despite widespread knowledge of the health consequences of obesity, its prevalence continues to increase worldwide. Data from the National Health and Nutrition Examination Survey indicate that half of the American population is overweight or obese.5 In Korea, the prevalence of obesity has increased rapidly since the 1990s, with 30.6% of the population considered overweight in 2001.6
Body mass index (BMI) is commonly used as an indicator of obesity for practical reasons. The WHO grades obesity according to BMI: underweight (<18.5 kg/m2), normal range (18.5–24.9 kg/m2), overweight (25.0–29.9 kg/m2), and obese (≥30.0 kg/m2),7 although variables such as age, sex, and race affect BMI.8,9 For example, older adults have a particular body size, shape, and composition, and they tend to lose fat-free mass and increase fat mass.10 Furthermore, Asians have a higher proportion of body fat for a given BMI than Caucasians.11 Therefore, it is almost certain that the relationship between BMI and other variables is different for elderly Asians than for younger Asians or Caucasians (who have been the subjects of most previous research).
Epidemiological studies have demonstrated that high BMI, consistent with being overweight or obese, is associated with a greater risk of mortality,2 but this relationship is not evident in older adults or in patients with chronic disease.12,13 Rather, a higher BMI appears to be correlated with a lower risk of mortality in what is known as the “obesity paradox.” A possible explanation for this reverse correlation is the insensitivity of BMI; nutrition and body composition might be factors of greater importance in relation to mortality in these populations. In the present study, the correlations between BMI, body composition, and mortality were evaluated in elderly Asians, who have a body composition that differs from that of people of other races.
- Top of page
Several studies have found a positive correlation between BMI and all-cause mortality in younger adults, not focused on the elderly population. In the present study, lean mass parameters (lean mass or LMI), rather than BMI, were significant risk factors for all-cause mortality in elderly Asians over a 3.5-year period. Other parameters of body composition (e.g., fat mass and fat proportion) were not associated with all-cause mortality. BMI was significantly correlated with various body composition factors. These correlations occurred at different degrees, and the relationship was weaker for lean mass than for fat mass.
Obesity is a known risk factor for mortality in the general population (in terms of age). In a prospective study from the Cancer Prevention Study cohort, high BMI was identified as increasing the risk of mortality over a 12-year period.4 Subject age could alter the relationship between BMI and mortality, because the mortality risk associated with high BMI in older adults was lower than in younger adults. In Korea, the association between BMI and mortality in the general population has been investigated in a large cohort study.2 Among 1,213,829 Koreans, deaths from any cause were found to have a J-shape association with BMI, but age modified this association, because BMI of 25.0 kg/m2 or greater was not associated with greater risk of all-cause mortality in the population aged 65 and older. Because conventional BMI criteria were not suitable for Asians, new BMI criteria were proposed (underweight, <18.5 kg/m2; normal range, 18.5–22.9 kg/m2; overweight, 23.0–24.9 kg/m2; obese, ≥25 kg/m2),19 but when analyzing these criteria in the current study, a positive correlation was not found between BMI and mortality (data not shown).
Body composition varies according to age, sex, and race. Older adults tend to lose fat-free mass and gain fat mass. Therefore, the generalized application of international BMI classifications7 to clinical practice may not detect malnutrition in older adults. Four-year mortality was evaluated in a large Italian population-based sample, aged 65 to 84.20 BMI less than 20 kg/m2 predicted all-cause mortality, highlighting the importance of nutritional management in older adults. Malnutrition–inflammation–cachexia syndrome has been proposed to explain why low BMI increases the risk of mortality in patients undergoing dialysis.21 Similar to the importance of nutritional support in patients undergoing dialysis, older adults need to pay equal attention to becoming underweight and overweight.22
BMI is not a reliable indicator of obesity, especially in older adults, because it does not differentiate lean mass from adipose tissue. BMI also poorly represents central fat mass and nutrition, for which more reliable parameters can be used. It has been found that waist circumference (proxy of central obesity) is a reliable marker of mortality in older adults but that BMI had a paradoxical correlation with mortality.12 Computed tomography (CT) and magnetic resonance imaging (MRI) are the methods of choice for assessing visceral fat,23,24 although in contrast to these expensive techniques, waist circumference and waist to hip ratio are simple and inexpensive methods for making similar assessments.25 Furthermore, the superiority of waist circumference over BMI in predicting CHD has been demonstrated.3 In a study with data from the Korean Acute Myocardial Infarction Registry, the highest level of mortality was observed for patients with the lowest BMI and the highest waist-to-hip ratio.26 In the present study, neither waist circumference nor waist-to-hip ratio was associated with all-cause mortality. This finding was attributed to the body shape of elderly Asians, who are typically leaner than Caucasians. Research shows that fat mass can protect against mortality in patients undergoing hemodialysis, which may help to explain the absence of any clear relationship between mortality and waist circumference or fat mass.27
Muscle mass, as represented by lean mass, is associated with survival, and the protective effect of muscle mass is well known in CKD. Recently, heart failure was reported to be associated with smaller mid-arm muscle circumference in patients undergoing hemodialysis. Muscle mass was measured using 24-hour urinary Cr excretion, and it was found that this influenced cardiovascular and all-cause mortality in patients undergoing hemodialysis.28 The high LMI group in the present study (Group 3) had a greater proportion of younger subjects and of those who exercised regularly and had high levels of serum albumin. These factors could improve survival rates, although those variables were adjusted for in multivariate analyses. Unfavorable factors such as male sex, diabetes mellitus, smoking, and low levels of serum HDL-C were predominant in the high LMI group. This discrepancy may be because patients with metabolic syndrome and those who exercised regularly were included in the high LMI group. Additional large population studies are needed to enable confounding factor stratification.
Although the present observational cohort study of randomly sampled older adults is informative, it is not without limitations. First, the results did not allow for the evaluation of any long-term effects associated with BMI or body composition. Based on the literature, the influence of high BMI on all-cause mortality may easily take 10 years or more to determine. However, similar to the present finding of no association between the short-term effects of BMI and mortality, some long-term studies over 12 years have revealed that elderly individuals with high BMI are not necessarily at greater risk of mortality.6 Second, because all participants in the cohort were elderly Asians, inferences from the data may not be applicable to the general population or to other ethnicities. Third, the sensitivity of BIA used in the present study is not precisely known. Although CT and MRI are more sensitive in measuring visceral adipose tissue, they are also more expensive. BIA is convenient and inexpensive, although the accuracy of this device is limited in older adults.29 Another limitation is that the assessment of lean mass is dependent on fat mass calculation, because weight is held constant, but some studies have demonstrated the reliability and internal validity of BIA.15 Further research is needed to investigate whether the technique used to measure body composition influences associations with mortality. Fourth, the sample size was modest. Thus, although the magnitude of the hazard ratio was large with regard to the potential clinical effect, the confidence intervals were wide. Furthermore, stratifying subjects into the modest sample size may have limited the potential significant results.
Several previous studies of older adults have found obesity to be paradoxically associated with better outcomes, although most studies defined obesity using BMI, which is a poor discriminator of body composition and has been extensively examined only in Caucasians. To the best of the authors' knowledge, the present study is the first to demonstrate the superior predictive power of lean mass parameters (lean mass or LMI), as opposed to BMI, in predicting all-cause mortality in elderly Asians. This suggests that elderly Asians at high risk of mortality could be better identified using body composition measurements, such as lean mass, than BMI measurements.