Comparison of bioimpedance spectroscopy and the Watson formula for measuring body volume in patients on peritoneal dialysis.

Introduction: Knowing total body volume (V) is crucial in patients on peritoneal dialysis (PD). It is usually calculated by the Watson anthropometric formula, although the use of bioimpedance spectroscopy (BIS) is becoming increasingly widespread. Measuring V with both methods can at times produce quite different results.

Objective: We aimed to identify differences between the 2 forms of measuring volume in a PD unit and determine which clinical factors are associated with these differences.

Methods: Ours is an observational study of 74 patients on PD. We measured V using BIS (Vbis) and the Watson formula (Vw); 271 measurements were made with each method. We calculated the difference between Vbis and Vw in each patient and classified them into 2 groups: Difference between volumes ≥10% or <10% Vbis. We assessed the presence of several clinical parameters in our patients. We assessed whether there were any differences between Vbis and Vw (Student t-test). We determined whether there was any association between the difference in volumes and the presence of the clinical parameters analysed (chi square test).

Results: V was 2.15 l higher measured by the Watson formula than with BIS (P<.01). In 58.67% of the measurements, the difference between Vbis and Vw was ≥10%. Significant differences were found when comparing the presence of difference between volumes and the presence or not of diabetes mellitus (DM) (p=0,03), hypertension (HTN) (p=0,036), hypoalbuminemia (p<0,01), hypoprealbuminemia (p<0,01), low phase angle at 50 Hz (p<0,01), high C reactive protein (p<0,01), obesity (p=0,027), E/I ratio (ratio between extracellular and intracellular water) ≥1 (p<0.01) and residual diuresis (p=0.029).

Conclusions: There are significant differences in the V of PD Unit patients when obtained by Watson formula or by BIS. A difference between the measurements is associated with the presence of DM, HTN, hypoalbuminaemia, obesity, malnutrition, inflammation, E/I ratio ≥1 and the absence of residual diuresis.