Association of mortality and phase angle measured by different bioelectrical impedance analysis (BIA) devices

Purpose: A high phase angle measured by the Nutriguard (R) bioelectrical impedance analysis device is associated with a reduced mortality risk in older people. This retrospective study aims to analyze whether this association persists with the other devices that have been used in our hospital. Methods: This study encompasses all people 65 yrs and older who underwent a phase angle measurement between 1990 and 2011 at the Geneva University Hospitals, with the RJL-101 (R) (RJL Systems), Xitron 4000B (R) (Xitron Technologies), Eugedia (R) (Eugedia-Spengler) and Bio-Z (R) (Spengler). Diseases at the time of phase angle measurement were reported in the form of the Cumulative Illness Rating Scale. Date of death was retrieved until December 2012. Phase angle values were categorized into sex- and device specific quartiles, where quartile 1 represents the lowest quartile and reference value. Cox regressions were performed to evaluate the association between phase angle quartiles and mortality. Results: We considered 1878 people (969 women), of whom 1151 had died. In univariate sex-specific Cox regressions, the death risk decreased progressively as the phase angle quartile measured by the Bio-Z (R) or RJL-101 (R) increased. The HR (95% CI) in quartile 4 was 0.36 (0.26, 0.50) and 0.38 (0.29, 052) in women and men measured with the Bio-Z (R) (both p < 0.001), and 0.23 (0.14, 0.39) and 0.19 (0.10, 0.36) in women and men measured with the RJL-101 (R) (both p < 0.001). The association between phase angle and mortality persisted when adjusted for age, body mass index or co-morbidities. The small number of deaths in people who underwent a measurement by Eugedia (R) (n = 93) or Xitron 4000B (R) (n = 56) did not allow performing multivariate Cox regressions. Conclusions: Phase angle quartiles are associated with mortality in people aged >= 65 years when using the RJL-101 (R) or Bio-Z (R) device. (C) 2017 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.