**3.1 Tanita (TBF-300)**

Tanita (TBF-300) is a device for measuring body impedance between the two legs. It is simple to use and does not require electrodes. Manufactured in Japan, the devise presents as a weight scale with a built-in body composition analyzer which calculates TBW, total body fat and fat-free body mass [61]. The National Institute of Health has not validated the results. For these measurements, the person stands barefoot on a four-point platform. Impedance measurements are made using a high frequency (50kHz) and low intensity (500 µA) current between the feet. (formula are not known)

Bioimpedance Measurement in the Kidney Disease Patient 175

intracellular water is not constant. Comparing total and extracellular water measurements obtained in the healthy population and in the dialysis population defines the notion of overhydration. Time-course curves of the different compartments provide a picture of the

In Figure 9, the notion of hydration status in the dialysis patient is combined with systolic blood pressure. Four sectors are defined. I: over-hydration and hypertension, II: normohydration and hypertension, III: normo-hydration and normo-tension IV: over-hydration

II I

IV

We first used the Tanita TBF-300 analyzer to obtain bioimpedance measures in 238 hemodialysis sessions. The mean total body water (TBW) was 41.9±7.9 kg before sessions and 39.6±7.3 kg after sessions. Mean body weight was 76.6±18.2 kg before and 74.2±17.1 kg after. Statistically, body weight remained stable during this study and was thus valid for assessing reliability. This enabled the definition of patient populations used to hypothesize the pathophysiology of hyperhydration, dehydration and normohydration state based on a body-weight index (Fig.10). Change in weight (W) and TBW (TBW) was noted and compared with ultrafiltration. In theory, these three variables should be similar, because during dialysis sessions, the ultrafiltration applied corresponds to a subtraction of water, and consequently to the change in w or TBW before and after the water depletion. But net ultrafiltration and W correlated poorly (c=0.62) with TBW. This led to the definition of a dry-weight index, It was hypothesized that this Index would be 0 when patients reach

OH

evolving state of hydration.

140

Fig. 9. Hydration reference

**3.3.1 Hemodialysis patients** 

their dry weight: **I = W– TBW** 

**3.3 Experimental works and results** 

and normo- or hypo-tension. Trends can be plotted.

Blood Pressure

0

**3.3.1.1 Elaborating a dry-weight index using Tanita** 

III

#### **3.2 Body composition monitor**

The body composition monitor (BCM) was developed by Fresenius® Medical Care. BCM is a rapid non-invasive 4-electrodes method for monitoring body composition by measuring the electrical impedance response to a multifrequency signal (50 frequencies from 5 kHz to 1 MHz). The Cole-Cole model is applied, with Hanaï mixture theory, to distinguish extracellular water (ECW), intracellular water (ICW) and total body water. Integrated software can also display over-hydration status. The device has been validated in healthy people, in comparison with reference methods (isotopic dilution).

The advantage of this device is to provide a quantitative diagnosis of over-hydration. The model is based on tissue hydration constants observed in the healthy individual which can then be applied to quantify excess water in kidney disease patients. The validation study, based on gold standard methods (bromide dilution for the extracellular compartment, deuterium dilution for TBW, potassium dilution for intracellular compartment and DEXA for fat mass), included 500 healthy subjects. The measures were then compared with the hemodialysis patient, before and after dialysis. This technique assumes that intracellular water, fat-free mass, and fat mass remain unchanged during the course of the dialysis session despite the ultrafiltration [62].

Fig. 8. Physiological model used by the body composition monitor (BCM) to determine the hydration constant of tissues and thus overhydration (OH). ATM: adipocyte tissue mass; LTM: lean tissue mass

The novel aspect of this model is the introduction of a third compartment corresponding to the excess fluid (Fig. 8) or over-hydration (OH) in comparison with the healthy individual. This over-hydration is considered to concern the extracellular compartment exclusively. Lean body mass contains exactly 73.2% water and includes the intra- and extra-cellular compartments, the intracellular compartment predominating. Lean body mass includes protein (muscles) and mineral (bone) matter. The sum of the intracellular compartment plus the protein mass defines the cell mass. Adipose tissue mass, or fat mass, contains 26.8% water, the extracellular compartment predominates. The notion of a third compartment avoids the problems related to the fact that the ratio between the extracellular and

The body composition monitor (BCM) was developed by Fresenius® Medical Care. BCM is a rapid non-invasive 4-electrodes method for monitoring body composition by measuring the electrical impedance response to a multifrequency signal (50 frequencies from 5 kHz to 1 MHz). The Cole-Cole model is applied, with Hanaï mixture theory, to distinguish extracellular water (ECW), intracellular water (ICW) and total body water. Integrated software can also display over-hydration status. The device has been validated in healthy

The advantage of this device is to provide a quantitative diagnosis of over-hydration. The model is based on tissue hydration constants observed in the healthy individual which can then be applied to quantify excess water in kidney disease patients. The validation study, based on gold standard methods (bromide dilution for the extracellular compartment, deuterium dilution for TBW, potassium dilution for intracellular compartment and DEXA for fat mass), included 500 healthy subjects. The measures were then compared with the hemodialysis patient, before and after dialysis. This technique assumes that intracellular water, fat-free mass, and fat mass remain unchanged during the course of the dialysis

ECW

Fig. 8. Physiological model used by the body composition monitor (BCM) to determine the hydration constant of tissues and thus overhydration (OH). ATM: adipocyte tissue mass;

The novel aspect of this model is the introduction of a third compartment corresponding to the excess fluid (Fig. 8) or over-hydration (OH) in comparison with the healthy individual. This over-hydration is considered to concern the extracellular compartment exclusively. Lean body mass contains exactly 73.2% water and includes the intra- and extra-cellular compartments, the intracellular compartment predominating. Lean body mass includes protein (muscles) and mineral (bone) matter. The sum of the intracellular compartment plus the protein mass defines the cell mass. Adipose tissue mass, or fat mass, contains 26.8% water, the extracellular compartment predominates. The notion of a third compartment avoids the problems related to the fact that the ratio between the extracellular and

ECW

ICW

ICW

ECW

people, in comparison with reference methods (isotopic dilution).

**3.2 Body composition monitor** 

session despite the ultrafiltration [62].

OH

LTM

ATM

LTM: lean tissue mass

intracellular water is not constant. Comparing total and extracellular water measurements obtained in the healthy population and in the dialysis population defines the notion of overhydration. Time-course curves of the different compartments provide a picture of the evolving state of hydration.

In Figure 9, the notion of hydration status in the dialysis patient is combined with systolic blood pressure. Four sectors are defined. I: over-hydration and hypertension, II: normohydration and hypertension, III: normo-hydration and normo-tension IV: over-hydration and normo- or hypo-tension. Trends can be plotted.

Fig. 9. Hydration reference

#### **3.3 Experimental works and results 3.3.1 Hemodialysis patients**

#### **3.3.1.1 Elaborating a dry-weight index using Tanita**

We first used the Tanita TBF-300 analyzer to obtain bioimpedance measures in 238 hemodialysis sessions. The mean total body water (TBW) was 41.9±7.9 kg before sessions and 39.6±7.3 kg after sessions. Mean body weight was 76.6±18.2 kg before and 74.2±17.1 kg after. Statistically, body weight remained stable during this study and was thus valid for assessing reliability. This enabled the definition of patient populations used to hypothesize the pathophysiology of hyperhydration, dehydration and normohydration state based on a body-weight index (Fig.10). Change in weight (W) and TBW (TBW) was noted and compared with ultrafiltration. In theory, these three variables should be similar, because during dialysis sessions, the ultrafiltration applied corresponds to a subtraction of water, and consequently to the change in w or TBW before and after the water depletion. But net ultrafiltration and W correlated poorly (c=0.62) with TBW. This led to the definition of a dry-weight index, It was hypothesized that this Index would be 0 when patients reach their dry weight: **I = W– TBW** 

Bioimpedance Measurement in the Kidney Disease Patient 177

Body composition was monitored for a 1.5-month period under standardized conditions in 10 hemodialysis patients aged over 80 years with the BCM. The data collected (Table 1) showed that these elderly patients exhibited a different water distribution pattern during ultrafiltration and also different refilling behavior. Thus ultrafiltration did not fully reflect changes in over-hydration. The only valid and clearly coherent measures were those recorded before the dialysis session. While the Cole-Cole model remained valid, implanted devices or specific clinical situations such as heart failure affected the impedance results. Pacemaker behavior (patient n° 8) was unaffected by the very small current used, but the presence of conductive metal (orthopedic prosthesis in patient 10) modified the bioimpedance measures. In such patients, change in weight and ultrafiltration followed a similar pattern, but TBW recordings were obviously incorrect in light of the weight loss

> **∆ Weight (kg) UF (l) ∆ TBW (l) ∆ ECW (l) ∆ ICW (l) mean** *SD* **mean** *SD* **mean** *SD* **mean** *SD* **mean** *SD*

patient 1 2,03 *0,36* 2,1 *0,59* 2,5 *0,96* 2,07 *0,47* 0,47 *0,47*  patient 2 1,22 *0,2* 1,22 *0,32* 1,26 *0,41* 1,12 *0,19* 0,16 *0,29*  patient 9 1,43 *0,68* 1,76 *0,72* 1,53 *0,46* 1,2 *0,62* 0,3 *0,82*  patient 3 1,58 *0,3* 1,3 *0,23* 0,78 *0,43* 0,55 *0,06* 0,23 *0,46*  patient 5 1,52 *0,51* 1,28 *0,31* 0,46 *1,46* 0,82 *0,48* -0,3 *0,17*  patient 6 2,9 *0,29* 2,8 *0,15* 1 *0,71* 1,85 *0,35* -0,95 *1,06*  patient 10 1,3 *0,99* 1,52 *0,55* 0,77 *0,75* 0,6 *0,36* 0,17 *0,42*  patient 4 1,33 *0,22* 1,18 *0,16* 2,7 *0* 1,7 *0,17* 0,97 *0,15*  patient 8 2,63 *0,43* 2,34 *0,34* 6,83 *2,48* 1,5 *0,53* 5,43 *2,94*  patient 7 1,53 *2,19* 0,72 *0,19* 0,78 *0,57* 1 *0,68* -0,25 *0,17* 

Patients 1, 2 and 9 exhibited good correlation between weight loss, applied

 Patients 3, 5, 6 and 10 exhibited a different type of behavior. The decline in TBW from the beginning to the end of the dialysis session was underestimated as compared with

Inversely, for patients 4 and 8, the decline in TBW during dialysis was overestimated,

In one patient (n° 7), ultrafiltration and TBW correlated well with each other, but not

As expected, dialysis-related water depletion mainly involved a reduction of the extracellular compartment compared with the intracellular compartment. This well-known phenomenon is called refilling. In three patients (n° 5, 6 and 7), the bioimpedance measures predicted fluid overload in the intracellular compartment, with a smaller fall in TBW than in extracellular water. This might be related to a "reverse refilling" phenomenon. For patient

Table 1. Participation of fluid compartments during ultrafiltration (UF)

ultrafiltration, and TBW variation during dialysis sessions.

the variation in weight or applied ultrafiltration.

with weight loss.

Four patient profiles could be identified from the data presented in Table 1.

compared with the very similar variations in weight or ultrafiltration.

**3.3.1.2 Body composition monitoring of the elderly hemodialysis patients** 

(Table 1) and the over-hydration diagnosis (Table 2).

Fig. 10. Pathophysiology of hydration state. **upper panel:** hyperhydration; **middle panel**: dehydration; **lower panel**: normo-hydration. TBW = total body water, W = body weight. Dry-weight index : **I = ΔW– ΔTBW** 

Fig. 10. Pathophysiology of hydration state. **upper panel:** hyperhydration; **middle panel**: dehydration; **lower panel**: normo-hydration. TBW = total body water, W = body weight.

Dry-weight index : **I = ΔW– ΔTBW** 

#### **3.3.1.2 Body composition monitoring of the elderly hemodialysis patients**

Body composition was monitored for a 1.5-month period under standardized conditions in 10 hemodialysis patients aged over 80 years with the BCM. The data collected (Table 1) showed that these elderly patients exhibited a different water distribution pattern during ultrafiltration and also different refilling behavior. Thus ultrafiltration did not fully reflect changes in over-hydration. The only valid and clearly coherent measures were those recorded before the dialysis session. While the Cole-Cole model remained valid, implanted devices or specific clinical situations such as heart failure affected the impedance results. Pacemaker behavior (patient n° 8) was unaffected by the very small current used, but the presence of conductive metal (orthopedic prosthesis in patient 10) modified the bioimpedance measures. In such patients, change in weight and ultrafiltration followed a similar pattern, but TBW recordings were obviously incorrect in light of the weight loss (Table 1) and the over-hydration diagnosis (Table 2).


Table 1. Participation of fluid compartments during ultrafiltration (UF)

Four patient profiles could be identified from the data presented in Table 1.


As expected, dialysis-related water depletion mainly involved a reduction of the extracellular compartment compared with the intracellular compartment. This well-known phenomenon is called refilling. In three patients (n° 5, 6 and 7), the bioimpedance measures predicted fluid overload in the intracellular compartment, with a smaller fall in TBW than in extracellular water. This might be related to a "reverse refilling" phenomenon. For patient

Bioimpedance Measurement in the Kidney Disease Patient 179

Fig. 11. Cole-Cole time curves after kidney transplantation. In blue at Day 4 post transplantation (OH +0.5), in red at Day 10 post transplantation (OH +0.6)

We studied three patients with acute reversible kidney disease. Body composition monitoring was started at the time of the acute disease and continued to the recovery period. Results (impedance, reactance, phase shift, Cole-Cole curve) showed an overhydration period that disappeared with the normalization of kidney function. The time course of improved hydration status in this patient is represented in Figure 12. Overhydration and weight variations followed the same pattern, rising and declining in

0

1

2

Weight [kg]

OH [L]

3

4

**3.3.3 Body composition monitoring in acute kidney failure** 

J1 J2 J5 J6 J7 J8 J8 J9 J10 J11

Fig. 12. Hydration status time course in a patient recovering normal kidney function.

parallel.

76

80

84

88

92


n° 8 who had a pacemaker, fluid removal seemed to occur exclusively from the intracellular compartment.

Table 2. Body composition monitoring (BCM) diagnosis of over-hydration (OH) status

The over-hydration data presented in Table 2 show that, according to the BCM findings, all patients except two (n° 8 and 10), were over-hydrated at dialysis onset and underhydrated at dialysis end. Comparing the clinical diagnosis of over-hydration established by the nephrologist programming the ultrafiltration at the beginning of the session (Table 1) with the over-hydration status diagnosis established by BCM before dialysis (Table 2) revealed different trends. The nephrologist tended to overestimate over-hydration before dialysis while the BCM tended towards an underestimation. At the end of the ultrafiltration six of the patients were dehydrated, but not far from their dry weight, with a difference of ≤0.5 liters from normo-hydration.
