Bioelectrical impedance vector analysis in critically ill patients: a prospective, clinician-blinded investigation

Abstract

Introduction: Assessment of fluid status in critically ill patients is challenging. We aimed to assess the feasibility and validity of bioelectrical impedance vector analysis (BIVA) as a measure of hydration in critically ill patients.

Methods: We performed twice-daily BIVA measurements and fluid balance calculations and recorded physiological variables in mechanically ventilated patients within 24 h of intensive care unit (ICU) admission for up to 5 days. Treating clinicians were blinded to BIVA results.

Results: We performed 344 BIVA measurements in 61 patients. According to BIVA, 14 patients (23%) were dehydrated, 22 (36%) were normally hydrated and 25 (41%) were overhydrated upon ICU admission. Patients with normal BIVA hydration were less sick, had fewer comorbidities and had less deranged physiology than patients found to be dehydrated or overhydrated with BIVA. Cumulative fluid balance increased in patients found to be dehydrated with BIVA by a mean of 3.4±2.2 L, whereas in patients found to be overhydrated with BIVA, it decreased by a mean of 4.5±6.9 L. In patients found to be normally hydrated with BIVA, fluid balance remained unchanged. BIVA-defined hydration increased with 1 L (median change 1.5%, P =0.09) or 2 L (median change 0.7%, P =0.09) of calculated fluid gains. BIVA-defined hydration decreased (median change -0.8%, P =0.02) with a negative cumulative fluid balance of >2 L. BIVA-defined hydration between first and last measurement correlated with the corresponding change in fluid balance (ρ =0.25, P =0.05).

Conclusions: BIVA is feasible in critically ill patients. Its validity is supported by the observed characteristics of patients with different degrees of BIVA hydration upon admission and by different fluid management of such patients by blinded clinicians. The sensitivity of repeated BIVA hydration measurements to detect fluid accumulation or fluid balance changes <2 L was low, however. These contradictory findings provide the rational basis for studies of BIVA-assisted fluid management in ICU patients.

Fig. 1

Examples of values obtained with the Renal EFG BIVA™ Technology. BIVA bioelectrical impedance vector analysis, H height, PA phase angle, Xc reactance, BCM body cell mass

Fig. 2

Selection of study patients. BIVA bioelectrical impedance vector analysis, ICU intensive care unit

Fig. 3

Cumulative fluid balance during the first 4 days in the intensive care unit (ICU) in patients with dehydration, normal hydration and overhydration on admission. Values are mean ± SE. BIVA bioelectrical impedance vector analysis

Fig. 4

Mean daily hydration determined by bioelectrical impedance vector analysis (BIVA) in patients with dehydration (a), normal hydration (b) and overhydration (c) upon admission. P values represent repeated-measures analysis of variance for change in hydration level over time

Fig. 5

Changes in hydration determined by bioelectrical impedance vector analysis (BIVA) (a), central venous pressure (CVP) (b) and lactate (c) versus changes in cumulative fluid balance and changes in BIVA hydration versus changes in lactate (d) between the time points of the first and last BIVA recordings