Identification of fluid overload in elderly patients with chronic kidney disease using bioimpedance techniques

Diagnosis of fluid overload (FO) in early stage is essential to manage fluid balance of patients with chronic kidney disease (CKD) and to prevent cardiovascular disease (CVD). However, the identification of fluid status in patients with CKD is largely dependent on the physician's clinical acumen. The ratio of fluid overload to extracellular volume (FO/ECV) has been used as a reference to assess fluid status. The primary aim of this study was to compare FO/ECV with other bioimpedance methods and clinical assessments in patients with CKD. Whole body ECV, intracellular volume (ICV), total body water (TBW), and calf normalized resistivity (CNR) were measured (Hydra 4200). Thresholds of FO utilizing CNR and ECV/TBW were derived by receiver operator characteristic (ROC) analysis based on data from pooled patients with CKD and healthy subjects (HSs). Clinical assessments of FO in patients with CKD were performed by nephrologists. Patients with CKD (stage 3 and stage 4) (n = 50) and HSs (n = 189) were studied. The thresholds of FO were <= 14.3 (10(-2) Omega m(3)/kg) for females and <= 13.1 (10(-2) Omega m(3)/kg) for males using CNR and >= 0.445 in females and >= 0.434 in males using ECV/TBW. FO was diagnosed in 78%, 62%, and 52% of patients with CKD by CNR, FO/ECV, and ECV/TBW, respectively, whereas only 24% of patients with CKD were diagnosed to be FO by clinical assessment. The proportion of FO in patients with nondialysis CKD was largely underestimated by clinical assessment compared with FO/ECV, CNR, and ECV/TBW. CNR and FO/ECV methods were more sensitive than ECV/TBW in identifying fluid overload in these patients with CKD. NEW & NOTEWORTHY We found that fluid overload (FO) in patients with nondialysis CKD was largely underestimated by clinical assessment compared with bioimpedance methods, which was majorly due to lack of appropriate techniques to assess FO. In addition, although degree of FO by bioimpedance markers positively correlated with the age in healthy subjects (HSs), no difference was observed in the three hydration markers between groups of 50 <= age <70 yr and age >= 70 yr in the patients with CKD.