Association of Height with Mortality in Patients Undergoing Maintenance Hemodialysis

Abstract

Background and objectives: Body mass index (BMI), determined as kilograms in body weight divided by the square of the height in meters (m(2)), is inversely associated with mortality in patients undergoing maintenance hemodialysis (MHD). It is commonly inferred that differences in the weight component of the BMI equation are responsible for this negative correlation. However, there are almost no data on the relationship between height and mortality in these patients. This study was conducted to examine the association between height and mortality in MHD patients and to evaluate the contribution of height to the BMI-mortality relationship.

Design, setting, participants, & measurements: A prospective study conducted from July 1, 2001, through June 30, 2006, enrolled a nationally representative cohort of 117,644 MHD patients receiving treatment in DaVita, Inc. outpatient dialysis facilities with (1) known height and weight, (2) age ≥18 years, (3) dialysis vintage ≥90 days, and (4) nonoutlying BMI values (≥12 to ≤60 kg/m(2)). The end date of follow-up was June 30, 2007, and median follow-up was 852 days (interquartile range, 504-1367 days). Mortality hazard ratios were computed within sex-standardized deciles of height and weight, and outcomes included all-cause mortality and cardiovascular, gastrointestinal, cancer, and infection mortality. Hazard models were unadjusted, adjusted for case-mix variables, or adjusted for case-mix variables plus laboratory variables.

Results: Mean age was 61±15 years; 45% of patients were women and 57% had diabetes. In adjusted models, height, also adjusted for weight, was directly associated with all-cause mortality and cardiovascular, infection, and cancer mortality. Compared with the median height decile, mortality risk in the highest height decile was 1.18 (95% confidence interval, 1.14 to 1.23) in fully adjusted analyses (P<0.001). Receiver-operating characteristic curves indicated that in adjusted analyses the contribution of height to the relationship between BMI and mortality was almost identical to that of weight.

Conclusions: In MHD patients, height is positively associated with mortality risk and contributes similarly to weight with regard to the negative BMI-mortality relationship.

Keywords: body mass index; chronic renal disease; clinical epidemiology.

Figure 1.

Algorithm summarizing the inclusion and exclusion criteria used to constitute the cohort for analysis. BMI, body mass index.

Figure 2.

Case-mix plus laboratory-adjusted mortality hazard ratios by sex-standardized decile of height and weight. Error bars represent 95% confidence intervals of hazard ratios.

Figure 3.

Case-mix plus laboratory-adjusted mortality hazard ratios by sex-standardized decile of height and weight. (A) cardiovascular (CV), (B) infection, (C) gastrointestinal (GI), and (D) cancer mortality outcomes. Error bars represent 95% confidence intervals of hazard ratios.

Figure 4.

All-cause mortality risks of taller patients (>70th sex-standardized percentile of height) relative to shorter patients (≤70th sex-standardized percentile of height), adjusted for case-mix plus laboratory variables, in subgroup analyses. A hazard ratio >1.0 indicates higher mortality risk in these taller patients compared with shorter patients.

Figure 5.

Receiver-operating characteristic (ROC) curve assessing the performance of height (unadjusted for weight) and weight (unadjusted for height) in predicting mortality in a case-mix model (adjusted for age, sex, and presence of diabetes).