Table 4 A summary of the findings of the four included studies
Author and Year | BIA Equipment/Protocol | BIA data (R [Ohm/m]; Xc [Ohm/m] and Pa [degrees]) | Measure of Hydration Status | Descriptive Findings |
|---|---|---|---|---|
Jones, et al. [30] (2015) | Bioelectrical impedance vector analysis (Renal EFG BIVA™ Technology; EFG Diagnostic, Belfast, UK) Impedance of 50 kHz Pairs of electrodes were placed on the dorsum of the wrist and the ipsilateral ankle The measurements were taken twice daily (in the morning and afternoon) for the first five days of each patient’s stay in the ICU or until ICU discharge, where the patients were positioned horizontally and placed in the supine position for ≥ 2 min | Dehydration (n = 14): R = 321.0 Xc = 43.6 Pa = 9.1 Euhydration (n = 22): R = 314.0 Xc = 29.6 Pa = 95.5 Overhydration (n = 25): R = 224.0 Xc = 16.5 Pa = 3.8 | A quantitative estimation was made of TBW volume as a percentage of fat-free body mass Dehydration (TBW of ≤ 72% of fat-free body mass) Euhydration (TBW of 73–74% of fat-free body mass) Overhydration (TBW of ≥ 75% of fat-free body mass) | BIA was used to categorise the patients according to hydration status using TBW volume thresholds of ≤ 72% (signifying dehydration), 73–74% (reflecting normal hydration), and ≥ 75% (indicating overhydration). While no direct measures of diagnostic accuracy were reported given the lack of a comparator group, the results showed that BIA reliably identified dehydration based on positive physiological responses to fluid challenges and maintenance fluid therapy. Dehydration, identified using BIA, was associated with a non-significant increase in the requirement for renal replacement therapy (p = 0.800), length of ICU stay (p = 0.870), length of hospital stay (p = 0.220), admission to intensive care (p = 0.890), and the rate of hospital mortality (p = 0.550) compared to normally hydrated subjects (all p > 0.05) |
Kafri, et al. [31] (2013) | Multi-frequency BIA (Maltron BioScan® 920–2; Maltron International, Essex, UK) Impedance of 5 kHz, 50 kHz, and 100 kHz Pairs of electrodes were placed on the talus and the third and fifth digits of the foot and on the third and fifth knuckles of the hand and the wrist The participants fasted for at least two hours and were asked to remove any jewellery and to micturate if they wished before the BIA measurements were taken Two consecutive measurements were taken over a couple of seconds within 20 min of the blood samples being taken while the subjects were in the supine position. The recordings were repeated a few minutes later. An average of the two consecutive measurements was calculated. The first data set was used in the event of variation of ≥ 3% Height, weight, gender, age, and ethnicity were entered into the MF-BIA device BIA outputs (the mean of the two readings) were used to calculate TBW (L) and ECW (L) using published equations for older people TBW, ECW, and ICW were calculated as body weight percentages using equations specifically developed for older adults (rather than those already programmed in the device) TBW was estimated using the Vaché equation: TBW = (2.896) + (0.366*height2/R100) + (0.137*weight) + (2.485*G) R100 = impedance at 100 Hz, and G = gender, with a value of 1 for men and 0 for women ECW was estimated using the Visser equation: ECW (Women) = (1.7) + (0.2*height2)/(R5) + (0.057*weight) ECW (men) = (4.8) + (0.225* height2)/(R5) | Not reported | A quantitative estimation was made of (1) TBW volume as a percentage of body weight, (2) ICW as a percentage of TBW, and (3) ECW as a percentage of TBW using MF-BIA (using published equations for older adults) | The accuracy of BIA varied with the threshold of TBW volume congruent with dehydration. The highest sensitivity (100%) was observed for TBW volume of 55% with low corresponding specificity (i.e. of only 14%). The highest specificity (91%) was observed for a TBW volume of 45%; however, the corresponding sensitivity was only 17%. The positive and negative predictive values were 25–33% and 79–100%, respectively. Optimal accuracy with a modest sensitivity and specificity (62–67%) was observed for a TBW volume threshold of 52% |
Powers, et al. [22] (2009) | Single-frequency BIA (Real-Time RJL Systems® Analyser, Clinton Township, Michigan, USA) Impedance of 50 kHz Pairs of electrodes were placed on the dorsal surfaces of the right hand and foot proximal to the metacarpal, phalangeal, and metatarsal phalangeal joints One additional pair of electrodes was applied at the pisiform bone of the right wrist and between the medial and lateral malleoli of the right ankle The participants were placed in the supine position, with their arms and legs abducted at an angle of 30–45°. Overnight fasting was required | Not reported | R, Xc, Ht, wt, gender, age, and amount of exercise were entered into a software programme called Cyrus TBW and ECW were estimated using the BIA device | The use of BIA was associated with small inter-individual variability in relation to the accurate measurement of TBW volume percentage (4.1%), compared to the reference tests, which suggests that the method feasibly identified dehydration and changes in hydration status. ECW volume, measured using BIA, was not significantly different to that measured using NaBr (p = 0.430) |
Ritz [32] (2001) | Multi-frequency BIA (Analycor-3®; Spengler, Cachan, France) Impedance of 5 kHz, 50 kHz, and 100 kHz, with current of 400 µA Pairs of electrodes were placed on the distal end of the third metacarpal bone and the distal end of the second metatarsal bone. One additional pair of electrodes was applied between the styloid processes of the radius and ulna and between the two malleoli of the ankle The measurements were taken on both sides of the body. Overnight fasting (approximately 12 h) was required. The measurements were taken after resting for at least 30 min and up to five hours post the administration of the H218O and Br doses | Not reported | A quantitative estimation was made of TBW and ECW as a percentage of body weight TBW was estimated at 50 kHz and 100 kHz using the following equations: TBW (l) 2.896 0.366 Ht2 = + ⁄ I100 + 0.137 wt + 2.485G TBW (l) 3.026 0.358 Ht2 = + ⁄ I50 + 0.149 wt + 2.924G ECW was estimated at 5 kHz using the following equations: ECW (Segal,1) = -6.1 + 0.284 Ht2/I5 + 0.112 wt ECW (Visser, men,1) = 4.8 + 0.225 Ht2/I5 ECW (Visser, women,1) = 1.7 + 0.2 Ht2/I5 + 0.057 wt In all the equations, Ht was measured in centimetres, and wt in kilogrammes. I signifies ‘impedance’, and G ‘gender’ (with values of 0 and 1 for women and men, respectively) | Sufficient comparability was observed between BIA and the reference tests in measuring TBW volume; notably, the method was able to discriminate between dehydration and normal hydration based on a TBW volume of 0.25–0.39 L. In this regard, average TBW volume in normally hydrated subjects was 0.69–0.83 L, considerably higher than that in dehydrated subjects |