3.2 Effects of BMI on mortality in COPD
Compared with normal weight, the RRs of underweight, overweight, and obese individuals were 1.40 (95% CI, 1.20–1.63; P <0.0001), 0.80 (95% CI, 0.67–0.96; P <0.0001), and 0.77 (95% CI, 0.62–0.95; P = 0.0162), respectively (Figs. 2–4). The heterogeneity test found I2 values of 94.84%, 93.27%, and 86.71%, for underweight, overweight, and obesity, respectively, and all the P-values obtained in the Q statistic test were <0.0001.
3.3 Dose–response analysis
We found a significant nonlinear relationship (P = 0.0078) between BMI categories and mortality in patients with COPD using random effects models. Figure 5 shows that the COPD patients whose BMI was less than the median of the normal weight category (<21.75 kg/m2) had a higher risk of death. In addition, an increase in the BMI resulted in a decrease in the risk of death. The risk of death was lowest (RR = 0.69; 95% CI, 0.53–0.89) when BMI was 30 kg/m2. The increase in BMI was no longer a protective factor for the risk of death when BMI reached >32 kg/m2, which was defined as obesity. The dose–response curve was “U” shaped.
3.4 Subgroup analysis
To explore the sources of heterogeneity, we performed a subgroup analysis by stratifying the BMI categories by age (<65 years and >65 years), follow-up period (>5 years and <5 years), and study design (perspective and retrospective study). As shown in Fig. 6, compared with normal weight, the effect of underweight was similar in each subgroup. For COPD patients who were overweight or obese, the association between higher body weight and lower mortality was not significant when the follow-up period was >5 years or the design of the studies was prospective. The follow-up period of the prospective studies included in our meta-analysis was >5 years. Therefore, this association may occur only with a relatively short (5-year) follow-up and for a high BMI. We also found that obesity was associated with a lower risk of all-cause mortality in COPD patients >65 years. However, no relationship between obesity and mortality was found for COPD patients <65 years. In addition, the subgroup analysis to some extent decreased the heterogeneity between the studies.
3.5 Publication bias
Egger's test indicated the absence of publication bias for underweight, overweight, and obesity compared with normal weight. The P-values in this test were 0.38, 0.20, and 0.73, respectively.
3.6 Sensitivity analysis
To determine sensitivity, we excluded one study at a time. In the comparison between normal weight, underweight, and overweight, the RRs and 95% CIs did not change substantially after removing one study at a time. However, after removing several single studies out one by one, the relationship between normal weight and obesity become not significant. This result was in agreement with the dose–response analysis results.
In this study, we updated the evidence and conducted a dose–response meta-analysis to elucidate the relationship between BMI and mortality in COPD patients. Our analysis indicated that underweight significantly increased the risk of all-cause mortality by 40%. By contrast, overweight and obesity were significantly associated with a decrease by 20% and 23% in the risk of all-cause mortality in patients with COPD, respectively. However, the relationship between obesity and higher risk of mortality was not stable so we removed several studies[15,25] one by one. Furthermore, the dose–response analysis indicated that the relationship between BMI and mortality in COPD patients was nonlinear. Compared with a BMI of 21.75 kg/m2, BMI <21.75 kg/m2 increased the risk of death in these patients whereas BMI >21.75 kg/m2 significantly decreased this risk. As the BMI was increasing, the RR was larger and larger than BMI with 21.75 kg/m2. When BMI reached 30 kg/m2, the RR was largest. The RR decreased for values >30 kg/m2, and when the BMI was >32 kg/m2, there was no significant difference in the risk of mortality compared with patients with a BMI of 21.75 kg/m2.
BMI is relatively easy to measure and has become a part of the BODE (BMI, Obstruction, Dyspnea, and Exercise capacity) index which is the most widely used tool to predict mortality in COPD patients. The BODE index is calculated as the sum of the BMI, obstruction, dyspnea, and exercise capacity scores and ranges between 0 and 10. A higher BODE index indicates a greater risk of death. The BMI in BODE is scored as 0 and 1 and its cutoff value is 21 kg/m2. However, the result of the dose–response analysis indicated that the relationship between BMI and mortality was non-linear and the effect of BMI on mortality was different in each category. Therefore, a more consistent scoring system for BMI should involve a division into three categories: a score of 2 for a BMI <21 kg/m2, 0 for a BMI of 21 to 30 kg/m2, and 1 for a BMI >30 kg/m2.
Our results suggest that underweight is associated with a higher risk of all-cause mortality in COPD patients. The reasons for this are still unclear. However, several hypotheses have been put forward to interpret this phenomenon, including respiratory muscle weakness, impaired gas exchange, impaired immune response, and loss of metabolically and functionally active fat-free mass (FFM). Underweight patients have an increased frequency of exacerbation, which leads to a faster decline in FEV1, impaired quality of life, and high mortality. On the contrary, obese patients with COPD may receive medical attention earlier than normal weight patients possibly because obesity is also associated with dyspnea.
With regard to obesity, our sensitivity analysis indicated that the association between body weight and all-cause mortality was not significant. Furthermore, Jordan and Mann observed an increase in mortality because of respiratory disease in extremely obese COPD patients. Therefore, there is limited evidence to demonstrate a close relationship between obesity and mortality. For this reason, further studies are needed to elucidate this relationship.
This study has several limitations. First, although several included studies were adjusted for potential covariates, we could not rule out the influence of other confounding factors. Second, different studies used different BMI category in particular for the definition of underweight. These differences may underestimate the risk of death in underweight patients with COPD. For this reason, we conducted a dose–response analysis to estimate the relationship between BMI and mortality accurately and to eliminate the inconsistency in the definition of the BMI categories. Third, this meta-analysis indicated a significant heterogeneity between the studies. This heterogeneity might be attributed to differences in sample size, duration of follow-up, disease severity, cutoff value for underweight, sex ratio, and the mean age of the population. Therefore, we conducted a subgroup analysis to explore the source of heterogeneity.
In summary, underweight increased the risk of mortality in patients with COPD whereas overweight decreased this risk. Furthermore, we found a nonlinear dose–response relationship between BMI and mortality in COPD patients. The risk of mortality was lowest for a BMI of 30 kg/m2. In this context, nutritional support may improve the prognosis of COPD patients in clinical practice.
1. The Asia Pacific COPD Roundtable Group. Global Initiative for Chronic Obstructive Lung Disease strategy for the diagnosis, management and prevention of chronic obstructive pulmonary disease
: An Asia–Pacific perspective. Respirology
2. Flegal KM, Kit BK, Orpana H, et al Association of all-cause mortality
with overweight and obesity using standard body mass index
categories: a systematic review and meta-analysis. JAMA
3. Cao C, Wang R, Wang JM, et al Body mass index
in chronic obstructive pulmonary disease
: a meta-analysis. PloS One
4. Liu XM, Liu YJ, Zhan J, et al Overweight, obesity and risk of all-cause and cardiovascular mortality
in patients with type 2 diabetes mellitus: a dose–response meta-analysis
of prospective cohort studies. Eur J Epidemiol
5. Bagheri M, Speakman JR, Shabbidar S, et al A dose–response meta-analysis
of the impact of body mass index
on stroke and all-cause mortality
in stroke patients: a paradox within a paradox. Obes Rev
6. Kovesdy CP, Anderson JE, Kalantar-Zadeh K. Paradoxical association between body mass index
in men with CKD not yet on dialysis. Am J Kidney Dis
7. Galesanu RG, Bernard S, Marquis K, et al Obesity in chronic obstructive pulmonary disease
: is fatter really better? Can Resp J
8. Wells G, Shea B, O’Connel D, et al The Newcastle–Ottawa Scale (NOS) for Assessing the Quality of Nonrandomised Studies in Meta-analyses. Ottawa, ON: Ottawa Hospital Research Institute; 2009.
9. Bianchini F, Kaaks R, Vainio H. Weight control and physical activity in cancer prevention. Obes Rev
10. Orsini N, Bellocco R, Greenland S. Generalized least squares for trend estimation of summarized dose–response data. Stata J
11. Orsini N, Li R, Wolk A, et al Meta-analysis for linear and nonlinear dose-response relations: examples, an evaluation of approximations, and software. Am J Epidemiol
12. Higgins JPT, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med
13. Egger M, Davey Smith G, Schneider M, et al Bias in meta-analysis detected by a simple, graphical test. BMJ
14. Duval S, Tweedie R. Trim and fill: a simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics
15. Chailleux E, Laaban J-P, Veale D. Prognostic value of nutritional depletion in patients with COPD treated by long-term oxygen therapy: data from the ANTADIR observatory. Chest
16. Chang CL, Sullivan GD, Karalus NC, et al Audit of acute admissions of chronic obstructive pulmonary disease
: inpatient management and outcome. Int Med J
17. Collins PF, Stratton RJ, Kurukulaaratchy R, et al Deprivation is an independent predictor of 1-year mortality
in outpatients with chronic obstructive pulmonary disease
18. Gunen H, Hacievliyagil SS, Kosar F, et al Factors affecting survival of hospitalised patients with COPD. Eur Resp J
19. Hallin R, Gudmundsson G, Ulrik CS, et al Nutritional status and long-term mortality
in hospitalised patients with chronic obstructive pulmonary disease
(COPD). Resp Med
20. Jordan JG Jr, Mann JR. Obesity and mortality
in persons with obstructive lung disease using data from the NHANES III. South Med J
21. Lainscak M, Haehling Sv, Doehner W, et al Body mass index
and prognosis in patients hospitalized with acute exacerbation of chronic obstructive pulmonary disease
. J Cachexia Sarcopenia Muscle
22. Landbo C, Prescott E, Lange P, et al Prognostic value of nutritional status in chronic obstructive pulmonary disease
. Am J Resp Crit Care Med
23. Marti S, Munoz X, Rios J, et al Body weight and comorbidity predict mortality
in COPD patients treated with oxygen therapy. Eur Resp J
24. Schembri S, Anderson W, Morant S, et al A predictive model of hospitalisation and death from chronic obstructive pulmonary disease
. Resp Med
25. Yamauchi Y. Paradoxical association between body mass index
and in-hospital mortality
in elderly patients with chronic obstructive pulmonary disease
in Japan. Int J Chronic Obstr Pulm Dis
26. Prescott E, Almdal T, Mikkelsen KL, et al Prognostic value of weight change in chronic obstructive pulmonary disease
: results from the Copenhagen City Heart Study. Eur Resp J
27. Ringbaek TJ, Viskum K, Lange P. BMI and oral glucocorticoids as predictors of prognosis in COPD patients on long-term oxygen therapy. Chronic Resp Dis
28. Celli BR, Cote CG, Marin JM, et al The body-mass index, airflow obstruction, dyspnea, and exercise capacity index in chronic obstructive pulmonary disease
. N Engl J Med
29. Schols AM, Broekhuizen R, Weling-Scheepers CA, et al Body composition and mortality
in chronic obstructive pulmonary disease
. Am J Clin Nutr
30. Sin DD, Jones RL, Man SFP. Obesity is a risk factor for dyspnea but not for airflow obstruction. Arch Int Med
Keywords:Copyright © 2016 The Authors. Published by Wolters Kluwer Health, Inc. All rights reserved.
body mass index; chronic obstructive pulmonary disease; dose–response meta-analysis; mortality