Influence of Body Mass Index on the Association of Weight Changes with Mortality in Hemodialysis Patients

• Published in: Clinical journal of the American Society of Nephrology Vol. 8; no. 10; pp. 1725 - 1733
• Main Authors: Cabezas-Rodriguez, Iván, Carrero, Juan Jesús, Zoccali, Carmine, Qureshi, Abdul Rashid, Ketteler, Markus, Floege, Jürgen, London, Gérard, Locatelli, Francesco, Gorriz, José Luis, Rutkowski, Boleslaw, Memmos, Dimitrios, Ferreira, Anibal, Covic, Adrian, Teplan, Vladimir, Bos, Willem-Jan, Kramar, Reinhard, Pavlovic, Drasko, Goldsmith, David, Nagy, Judit, Benedik, Miha, Verbeelen, Dierik, Tielemans, Christian, Wüthrich, Rudolf P, Martin, Pierre-Yves, Martínez-Salgado, Carlos, Fernández-Martín, José Luis, Cannata-Andia, Jorge B
• Format: Journal Article
• Language: English
• Published: United States American Society of Nephrology 07.10.2013
• Subjects: Adult; Aged; Body Mass Index; Body Weight; Female; Humans; Male; Medicin och hälsovetenskap; Middle Aged; Original; Proportional Hazards Models; Prospective Studies; Renal Dialysis - mortality
• ISSN: 1555-9041; 1555-905X; 1555-905X
• DOI: 10.2215/CJN.10951012

A high body mass index (BMI) is associated with lower mortality in patients undergoing hemodialysis. Short-term weight gains and losses are also related to lower and higher mortality risk, respectively. The implications of weight gain or loss may, however, differ between obese individuals and their nonobese counterparts. The Current Management of Secondary Hyperparathyroidism: A Multicenter Observational Study (COSMOS) is an observational study including 6797 European hemodialysis patients recruited between February 2005 and July 2007, with prospective data collection every 6 months for 3 years. Time-dependent Cox proportional hazard regressions assessed the effect of BMI and weight changes on mortality. Analyses were performed after patient stratification according to their starting BMI. Among 6296 patients with complete data, 1643 died. At study entry, 42% of patients had a normal weight (BMI, 20-25 kg/m(2)), 11% were underweight, 31% were overweight, and 16% were obese (BMI ≥ 30 kg/m(2)). Weight loss or gain (<1% or >1% of body weight) was strongly associated with higher rates of mortality or survival, respectively. After stratification by BMI categories, this was true in nonobese categories and especially in underweight patients. In obese patients, however, the association between weight loss and mortality was attenuated (hazard ratio, 1.28 [95% confidence interval (CI), 0.74 to 2.14]), and no survival benefit of gaining weight was seen (hazard ratio, 0.98 [95% CI, 0.59 to 1.62]). Assuming that these weight changes were unintentional, our study brings attention to rapid weight variations as a clinical sign of health monitoring in hemodialysis patients. In addition, a patient's BMI modifies the strength of the association between weight changes with mortality.