**1. Introduction**

234 Technical Problems in Patients on Hemodialysis

Muñoz R.I , Montenegro J, Salcedo A, et al. Effect of acetate-free biofiltration with a

Severi S, Vecchietti S, Cavalcanti S, et al. Electrocardiographic changes during

a pilot study. Hemodialysis International 2008; 12:108-113.

Phipps L, Harris D. Review: modelling the dialysate. Nephrology 2010; 15: 393-398.

381-388.

potassium-profiled dialysate on the control of cardiac arrhytmias in patients at risk:

hemodiafiltration with different potassium removal rates. Blood Purif 2003; 21:

Cardiovascular death remains the most frequent cause of mortality of dialysis population. Among the risk factor of cardiovascular mortality low blood pressure has been investigated by several authors even though not all the author conclude that low blood pressure per se, not adjusted for concomitant factors like age, and the presence of diabetes can be seen as independent predictor of mortality (Tisler et al., 2003; Iseki et al., 1997; Port et al., 1999).

In particular, the presence of low blood pressure and the intradialytic fall in the systolic and diastolic blood pressure seems to predict mortality in the dialysis population (Sohji et al., 2004). Some other authors found that not only low predialysis systolic blood pressure but also high values (J shape theory) expose patients at higher risk of mortality (Port et al, 1999).

Nevertheless, the fall in systolic blood pressure seems to be to date the most common intradialytic complication accounting for up to 30 % of dialysis related symptoms, despite the several improvement of dialysis techniques in terms of biocompatibility of the material used, more convection and quality of dialysis fluid. But hypertension is indeed the most frequent chronic co morbidity affecting dialysis patients.

Predialysis hypertension does not prevent hypotension episodes and not having targets for blood pressure control will not necessarily reduce its onset. When improved control of blood pressure is desired, modifications to the dialysis treatment itself should be considered as part of the management strategy (Davenport et al., 2008).

Hypo- as well hypertension are then the most challenging complications which dialysis has to face with. Several strategies are available to prevent hypo and hypertension like longer or more frequent dialysis regimen, diffusive-convective therapies, and assessment of dry body weight. Not all of them can be used in routine practice due to infrastructural or financial constraints while some others give in a certain extent opposite's effects like assessment of dry body weight beneficial for the hypertensive status, but which can lead to an increase of intradialytic symptoms (Davenport et al., 2008). Among the tools today available, the biofeedback systems, those devices able to adapt the operative condition of dialysis to the dynamic changes of the patients status along the dialysis, seems to be promising for contributing a step forward in the patients well-being (Locatelli et al., 2005) Among these devices the Hemocontrol Biofeedback System (HBS) has been extensively studied by several

Blood Volume Regulation 237

sodium to keep all the controlled variables as close to the target values as possible, however within pre-specified degree of tolerance. The desired blood volume reduction is set according to the total amount of fluid to be removed during dialysis (BV/UF volume). In fact, the blood volume reduction is a function of the overall water removal (Mancini et al, 1995), then its reduction per liters of water removal should be an invariant for any patient and then can take into account for the long or short interdialytic period. Since the actual dialysate sodium level is adjusted from the controller to regulate the blood volume time pattern, another parameter is set at the beginning of dialysis, that is the average dialysate sodium level. This value must prevent any sodium overload or abrupt depletion

All the actuators, that is the ultrafiltration and actual dialysate sodium, are always limited within safety operating regions. In particular the ultrafiltration rate is upper limited from the max UF rate at the beginning of dialysis to the average weight loss rate at the end and it is lower limited by the minimum UF rate (100 mL/h). The maximal initial UF rate is prescribed by the nurse according to patients characteristics and it is usually set at a value 20% to 50% higher than the average UF rate. The dialysate sodium is again upper limited from a maxi initial sodium content to the lowest value at the end of the treatment and the

Fig. 2. Example of a HBS dialysis. The time course of the controlled variables (BV, %, UF volume, L), are shown on the top panel, while the actuators (DC, mS/cm and Ultrafiltration rate rate, L/h) are shown at the bottom. According to the desired BV and UF vol time pattern (the dashed mid line in 1 and 2) the UFR and DC are adjusted time by time by the controller to minimize the error between the desired and actual values. At any time their maximum and minimum values are maintained within safety limits (dashed lines in the bottom panel) As long as the BV, UF Vol and average DC are within the pre-set tolerances (dashed lines in the top panel) there is no need to change any set-point (Santoro et al., 2002).

during the treatment (Moret et al., 2002).

user can set the most appropriate range for the patients (Figure 2).

authors as a tool to contribute in lowering the hypotension incidence but also as a further utility to control or normalize the blood pressure.

This review will summarize all the contributions given by the several studies about the cardiovascular stability in HBS dialysis.

#### **2. HemocontrolTM biofeedback system**

It is well known that blood volume behavior during dialysis is affected by several factors among which the ultrafiltration and changes in dialysate sodium level are the most relevant. Ultrafiltration should be adapted to the rhythm of plasma water removal respect to the patient's refilling capabilities, but the main limitation relies in the inability to maintain the total planned weight loss within the preset dialysis time except for long lasting dialysis treatments. The dialysate sodium can also promote the mobilization of water from the extravascular space reconstituting the plasma lost during ultrafiltration. Moreover, the modification of the intravascular sodium level can increase the activity of the autonomic nervous system, with a better hemodynamic response from the peripheral vascular resistance. However, the intradialytic sodium balance must be taken under tight control to avoid any sodium overload leading in turn a fluid overload especially in the interdialytic period. The blood volume control system (Hemocontrol) has been developed bearing in mind these three aspects (hemodynamic stability, fluid and sodium balance). The system is designed as feedback controller with three controlled variables (blood volume, total weight loss and average dialysate sodium level) and two actuators, the ultrafiltration and actual dialysate sodium level (Paolini & Bosetto, 1999) (Figure 1).

Fig. 1. Schematic representation of the Hemocontrol Biofeedback System. The MIMO controller processes three different controlled variables (blood volume reduction, average dialysate conductivity, body weight loss) through the errors between the actual and the desired values. The two actuators, reacting to the errors, are the actual dialysate conductivity and the weight loss rate (Santoro et al., 1998).

In Particular, the multi-input multi-output controller processes the errors between the actual continuously measured values of BV, weight loss and average dialysate sodium and their desired values, and reacts by adjusting the ultrafiltration and actual dialysate

authors as a tool to contribute in lowering the hypotension incidence but also as a further

This review will summarize all the contributions given by the several studies about the

It is well known that blood volume behavior during dialysis is affected by several factors among which the ultrafiltration and changes in dialysate sodium level are the most relevant. Ultrafiltration should be adapted to the rhythm of plasma water removal respect to the patient's refilling capabilities, but the main limitation relies in the inability to maintain the total planned weight loss within the preset dialysis time except for long lasting dialysis treatments. The dialysate sodium can also promote the mobilization of water from the extravascular space reconstituting the plasma lost during ultrafiltration. Moreover, the modification of the intravascular sodium level can increase the activity of the autonomic nervous system, with a better hemodynamic response from the peripheral vascular resistance. However, the intradialytic sodium balance must be taken under tight control to avoid any sodium overload leading in turn a fluid overload especially in the interdialytic period. The blood volume control system (Hemocontrol) has been developed bearing in mind these three aspects (hemodynamic stability, fluid and sodium balance). The system is designed as feedback controller with three controlled variables (blood volume, total weight loss and average dialysate sodium level) and two actuators, the ultrafiltration and actual

Fig. 1. Schematic representation of the Hemocontrol Biofeedback System. The MIMO controller processes three different controlled variables (blood volume reduction, average dialysate conductivity, body weight loss) through the errors between the actual and the

In Particular, the multi-input multi-output controller processes the errors between the actual continuously measured values of BV, weight loss and average dialysate sodium and their desired values, and reacts by adjusting the ultrafiltration and actual dialysate

desired values. The two actuators, reacting to the errors, are the actual dialysate

conductivity and the weight loss rate (Santoro et al., 1998).

utility to control or normalize the blood pressure.

**2. HemocontrolTM biofeedback system** 

dialysate sodium level (Paolini & Bosetto, 1999) (Figure 1).

cardiovascular stability in HBS dialysis.

sodium to keep all the controlled variables as close to the target values as possible, however within pre-specified degree of tolerance. The desired blood volume reduction is set according to the total amount of fluid to be removed during dialysis (BV/UF volume). In fact, the blood volume reduction is a function of the overall water removal (Mancini et al, 1995), then its reduction per liters of water removal should be an invariant for any patient and then can take into account for the long or short interdialytic period. Since the actual dialysate sodium level is adjusted from the controller to regulate the blood volume time pattern, another parameter is set at the beginning of dialysis, that is the average dialysate sodium level. This value must prevent any sodium overload or abrupt depletion during the treatment (Moret et al., 2002).

All the actuators, that is the ultrafiltration and actual dialysate sodium, are always limited within safety operating regions. In particular the ultrafiltration rate is upper limited from the max UF rate at the beginning of dialysis to the average weight loss rate at the end and it is lower limited by the minimum UF rate (100 mL/h). The maximal initial UF rate is prescribed by the nurse according to patients characteristics and it is usually set at a value 20% to 50% higher than the average UF rate. The dialysate sodium is again upper limited from a maxi initial sodium content to the lowest value at the end of the treatment and the user can set the most appropriate range for the patients (Figure 2).

Fig. 2. Example of a HBS dialysis. The time course of the controlled variables (BV, %, UF volume, L), are shown on the top panel, while the actuators (DC, mS/cm and Ultrafiltration rate rate, L/h) are shown at the bottom. According to the desired BV and UF vol time pattern (the dashed mid line in 1 and 2) the UFR and DC are adjusted time by time by the controller to minimize the error between the desired and actual values. At any time their maximum and minimum values are maintained within safety limits (dashed lines in the bottom panel) As long as the BV, UF Vol and average DC are within the pre-set tolerances (dashed lines in the top panel) there is no need to change any set-point (Santoro et al., 2002).

Blood Volume Regulation 239

observed in 32.16% of all patients, Glomerulonephritis in 29.03%, interstitial nephropathy

**Author Year Journal Type of** 

Santoro A et al 1994 ASAIO Trans Clinical Trial Santoro A 1995 Nephrol Dial Transplant Editorial Mancini E et al 1995 Int J Artif Organs Clinical Trial Santoro A et al 1996 Nephrol Dial Transplant Review Santoro A et al 1997 Int J Artif Organs Editorial Santoro A et al 1998 Am J Kidney Dis Clinical Trial Santoro A et al 1998 Nephrol Dial Transplant Review Paolini F et al 1999 Adv Ren Replace Ther Review Ronco C et al 2000 Kidney Int Clinical Trial Basile C et al 2001 Nephrol Dial Transplant Clinical Trial Zucchelli P et al 2001 Semin Nephrol Review Begin V et al 2002 ASAIO J Clinical Trial Pastore C et al 2002 EDTNA ERCA J Clinical Trial Santoro A et al 2002 Contrib Nephrol Review Santoro A et al 2002 Kidney Int Clinical Trial Wolkotte C et al 2002 Nephron Clinical Trial McIntyre CW et al 2003 Clin Nephrol Clinical Trial Santoro A et al 2003 J Nephrol Review Cavalcanti S et al 2004 Kidney Int Clinical Trial Franssen CF et al 2005 Hemodial Int Clinical Trial Selby NM et al 2006 Am J Kidney Dis Clinical Trial Moret K et al 2006 Nephrol Dial Transplant Clinical Trial Severi S et al 2006 Hemodial Int Clinical Trial Dasselaar JJ et al 2007 ASAIO J Clinical Trial Dasselaar JJ et al 2007 J Ren Care Clinical Trial Deziel C et al 2007 Clin J Am Soc Nephrol Clinical Trial Azar AT 2008 Saudi J Kidney Dis Transpl Review McIntyre CW et al 2008 Clin J Am Soc Nephrol Clinical Trial Nesrallah GE et al 2008 ASAIO J Clinical Trial Santoro A et al 2008 Contrib Nephrol Review Winkler RE et al 2008 Contrib Nephrol Clinical Trial

**publication** 

14.3%, IgA nephropathy 7.1%, chronic pyelonephritis 3.6%, other 13.81%.

Table 1. Full list of peer reviewed papers on HBS.

Fig. 3. State diagram of two out of the three controlled variables (BV and UF vol) during the sample dialysis shown in figure 2.

Thanks to the continuous adjustment of these parameters according to the dynamically changing clinical condition of the patient, the system is also useful to understand the refilling capacity of the patients. In fact, it can display the deviation from the set and forecasted end dialysis values respectively of the blood volume and UF volume to their set points (Figure 3). In particular, the diagram shows, on the horizontal axis, the difference between the forecasted end-dialysis BV and its set-point at any minute, while, on the vertical axis the difference between the forecasted end-dialysis UF volume and its set-point at any minute. It is worth to note that the deviation between the forecasted and set-point lies almost always in the upper right quadrant of the diagram. In this case, despite a forecasted higher UF volume, the blood volume does not fall down correspondingly, meaning that the patient's refilling capacity is higher than expected. On the contrary, when the state vector is in the lower left quadrant, the patient's refilling capacity seems to be less than expected

How much this can be attributable to the hydration status (over-hydration or underhydration) is still matter of discussion even though there should be a physiological dependency behind.

#### **3. Clinical studies on the hemocontrol™ biofeedback system**

Since the first appearance of an abstract talking about the blood volume monitoring and control in the far 1991 Santoro et al., 1991, several studies and papers have been undertaken and written. The overall body of this literature is reported in Table 1.

Many of them are studies showing the clinical benefit of HBS in particular in the reduction of hypotension events, some others have been published to explain the underlying mechanisms for such a kind of results (Cavalcanti et al, 2004, Severi et al, 2006). We will review the results of original clinical trials, which are summarized in Table 2.

Fifteen studies were prospective, randomised, controlled trials lasting from four weeks to 48 weeks. In general, study objectives addressed haemodynamic stability during dialysis, cardiovascular effects, dialysis efficiency, interdialytic events and quality-of-life aspects. In total, the HBS has been used in more than 280 patients who were hypertensive, hypotensionprone or non-hypotension-prone. The studies were run in Canada and several European countries. Actually, 287 patients were treated in the demonstrated clinical studies, ratio between male and female was 2.2, the mean age 66.3 years +/- 6.3 years. Diabetes mellitus was

Fig. 3. State diagram of two out of the three controlled variables (BV and UF vol) during the

Thanks to the continuous adjustment of these parameters according to the dynamically changing clinical condition of the patient, the system is also useful to understand the refilling capacity of the patients. In fact, it can display the deviation from the set and forecasted end dialysis values respectively of the blood volume and UF volume to their set points (Figure 3). In particular, the diagram shows, on the horizontal axis, the difference between the forecasted end-dialysis BV and its set-point at any minute, while, on the vertical axis the difference between the forecasted end-dialysis UF volume and its set-point at any minute. It is worth to note that the deviation between the forecasted and set-point lies almost always in the upper right quadrant of the diagram. In this case, despite a forecasted higher UF volume, the blood volume does not fall down correspondingly, meaning that the patient's refilling capacity is higher than expected. On the contrary, when the state vector is in the lower left quadrant, the patient's refilling capacity seems to be less than expected How much this can be attributable to the hydration status (over-hydration or underhydration) is still matter of discussion even though there should be a physiological

Since the first appearance of an abstract talking about the blood volume monitoring and control in the far 1991 Santoro et al., 1991, several studies and papers have been undertaken

Many of them are studies showing the clinical benefit of HBS in particular in the reduction of hypotension events, some others have been published to explain the underlying mechanisms for such a kind of results (Cavalcanti et al, 2004, Severi et al, 2006). We will review the results

Fifteen studies were prospective, randomised, controlled trials lasting from four weeks to 48 weeks. In general, study objectives addressed haemodynamic stability during dialysis, cardiovascular effects, dialysis efficiency, interdialytic events and quality-of-life aspects. In total, the HBS has been used in more than 280 patients who were hypertensive, hypotensionprone or non-hypotension-prone. The studies were run in Canada and several European countries. Actually, 287 patients were treated in the demonstrated clinical studies, ratio between male and female was 2.2, the mean age 66.3 years +/- 6.3 years. Diabetes mellitus was

**3. Clinical studies on the hemocontrol™ biofeedback system** 

and written. The overall body of this literature is reported in Table 1.

of original clinical trials, which are summarized in Table 2.

sample dialysis shown in figure 2.

dependency behind.


observed in 32.16% of all patients, Glomerulonephritis in 29.03%, interstitial nephropathy 14.3%, IgA nephropathy 7.1%, chronic pyelonephritis 3.6%, other 13.81%.

Santoro A et al 2002 Contrib Nephrol Review Santoro A et al 2002 Kidney Int Clinical Trial Wolkotte C et al 2002 Nephron Clinical Trial McIntyre CW et al 2003 Clin Nephrol Clinical Trial Santoro A et al 2003 J Nephrol Review Cavalcanti S et al 2004 Kidney Int Clinical Trial Franssen CF et al 2005 Hemodial Int Clinical Trial Selby NM et al 2006 Am J Kidney Dis Clinical Trial Moret K et al 2006 Nephrol Dial Transplant Clinical Trial Severi S et al 2006 Hemodial Int Clinical Trial Dasselaar JJ et al 2007 ASAIO J Clinical Trial Dasselaar JJ et al 2007 J Ren Care Clinical Trial Deziel C et al 2007 Clin J Am Soc Nephrol Clinical Trial Azar AT 2008 Saudi J Kidney Dis Transpl Review McIntyre CW et al 2008 Clin J Am Soc Nephrol Clinical Trial Nesrallah GE et al 2008 ASAIO J Clinical Trial Santoro A et al 2008 Contrib Nephrol Review

Winkler RE et al 2008 Contrib Nephrol Clinical Trial Table 1. Full list of peer reviewed papers on HBS.

Blood Volume Regulation 241

**Compared to HD** 

fluid volume

eKt/V ↑ LVMI ↓ EF ↑

systolic and diastolic BP ↓ intradialytic interventions ↓ QoL/KDQOL-SF (burden of kidney disease) ↑

no change in extracellular

intradialytic hypotension ↓ no change in QoL/dialysisrelated symptoms quest.

hypotensive episodes ↓ muscle cramps ↓

Antihypertensive drugs ↓

*P*


<0.01 <0.01 <0.05 <0.01 n.s n.s

n.s. 0.04 0.004

**Author Patients Study Design Major Results with HBS** 

\*wash-out/run-in phase; § SP, sodium profiling; # PC, plasma conductivity controlled feedback; ↓,

1) during first HBS phase post-HD weight remained stable, during second phase reduced target weight 2) 11 consecutive treatments for each modality followed by 1 week of treatment with standard dialysis Abbreviations: RCT, randomised controlled trial; HD: haemodialysis; HBS: HemocontrolTM biofeedback system; SAP, systolic arterial blood pressure; n.s., not significant; BP, blood pressure, LV(MI), left

Based on the European Best Practice Guidelines (EBPG) for hemodynamic instability (Kooman et al., 2007) the average incidence of hypotensive episodes during dialysis therapy is 20%. In some cohort studies, hypotensive events were observed in up to 33% of cases during dialysis therapy. Hypotensive episodes during the course of dialysis therapy are closely correlated with morbidity and mortality and play a fundamental role in the development of myocardial and cerebral ischemia. Frequent occurrences of hypotension lead to chronic over-hydration and have a negative impact on the clearance of dissolved

The first study on HBS was run by Santoro in 1994 which was addressed to check the feasibility of the system. They included a small sample of five patients in a HD-HBS-HD experimental set-up. The occurrence of dialysis complicated by severe hypotension was 8 in

Subsequently, they did another clinical investigation on 8 hypotension-prone patients in a prospective crossover study including 8 hypotension-prone haemodialysis patients (Santoro et al, 1994). They compared conventional haemodialysis (HD) to the HBS, following a protocol with an HD1-HBS-HD2 sequence, with each treatment period lasting one month. Changes in predialysis to postdialysis systolic arterial pressure were lower in the HBS

Prospective RCT HD(4wks\*+6mo) vs. HBS(4wks\*+6mo)

Prospective RCT HD(4wks\*+6mo) vs. HBS(4wks\*+6mo)

Retrospective HDHBS(48wks)

Table 2. Summary of the main results of the clinical trias on HBS.

Déziel et al., 200711

Nesrallah et al., 200812

Winkler et al., 200813

reduced; ↑, improved

50% hypotensionprone 50% hypertensive n=36

hypotensionprone n=60

hypotensionprone n=18

ventricular (mass index); EF, cardiac ejection fraction.

**4. Intradialytic haemodynamic stability** 

substances due to water retention.

HD, 1 in HBS and 5 in the second HD period.


**Compared to HD** 

Intradialytic symptoms ↓

hypotensive episodes ↓ saline infusions ↓ rebound ↓ (eKt/V↑)

post-HD-asthenia ↓(other

interdialysis symptoms ↓

symptomatic episodes ↓ reductions in systolic BP ↓ RBV falling >10% ↓ interdialytic weight gain ↓ eKt/V ↑, urea clearance ↑

intradialytic hypotension ↓

no effect on post-HD dry

hypotensive episodes least

LV regional wall motion

haemodynamic data ↑

hypotensive episodes ↓ brain natriuretic peptide

predialysis systolic BP ↓

frequent with HBS

abnormalities ↓

intradialytic stability (SAP)↑ hypotensive episodes ↓ intradialytic events ↓ isotonic saline ↓

Intradialytic hypotension ↓

symptomatic hypotension ↓ muscle cramps ↓

symptoms n.s.) vascular refilling ↓

Intradialytic hypotension ↓ Intradialytic symptoms ↓

intradialytic hypotension ↓

systolic BP ↑

weight

EF ↑

levels ↓

event-free sessions ↑ mean postdialysis BP ↑ *P*

<0.05 n.s.

<0.05 <0.05 - <0.05

<0.001 <0.001 <0.001

<0.002 <0.02 <0.0001 <0.05

<0.01 -

0.033 0.039

0.004 <0.001

<0.001 - <0.001 0.009 <0.01

<0.01 <0.05

n.s.

<0.05 n.s. <0.05


**Author Patients Study Design Major Results with HBS** 

Prospective, cross-over HD(2wks) – HBS(2wks) –

Prospective, cross-over HD(4wks)HBS(4wks)HD(4

Prospective, cross-over HD(6mo)HBS(14-30mo) HD(4\*+3wks)HBS(4\*+3wks)

Prospective, cross-over 3x[HD(2wks)HBS(2wks)]

Prospective, cross-over HD(3wks) - HBS(3wks)

Prospective, cross-over HD(3wks)HBS(2\*+3wks)

Prospective, cross-over HD(3wks)HBS(3wks)HBS(

Prospective RCT, cross-over 4 phases: HD, HBS, SP§, PC#2

Prospective RCT, cross-over HD(1\*+2wks)HBS(1\*+2wks)

HBS(1\*+2wks)HD(1\*+2wks)

Prospective RCT HD(4\*+12wks) vs. HBS(4\*+12wks)

6wks)1

vs.

Prospective RCT, cross-over 2x[HD(4wks)HBS(4wks)] vs. 2x[HBS(4wks)HD(4wks)]

Prospective RCT, cross-over HD(2wks)HBS(2wks) vs. HBS(2wks)HD(2wks)

HD(2wks)

wks)

Santoro et al, 1994

Santoro et al., 19981

Ronco et al., 20002

Basile et al., 20013

Bégin et al., 20024

Wolkotte et al., 2002

Santoro et al., 20025

McIntyre et al., 20036

Franssen et al., 20057

Moret et al., 20068

Selby et al., 20069

Dasselaar et al., 200710

hypotension prone n=5

hypotensionprone n=8

hypotensionprone n=12

hypotensionprone n=19

hypotensionprone n=7

Unselected sample N=16

hypotensionprone n=36

hypotensionprone n=15

hypotensionprone n=7

hypotensionprone n=12

hypotensionprone n=8

hypertensive n=28

non-


\*wash-out/run-in phase; § SP, sodium profiling; # PC, plasma conductivity controlled feedback; ↓, reduced; ↑, improved

1) during first HBS phase post-HD weight remained stable, during second phase reduced target weight 2) 11 consecutive treatments for each modality followed by 1 week of treatment with standard dialysis Abbreviations: RCT, randomised controlled trial; HD: haemodialysis; HBS: HemocontrolTM biofeedback system; SAP, systolic arterial blood pressure; n.s., not significant; BP, blood pressure, LV(MI), left ventricular (mass index); EF, cardiac ejection fraction.

Table 2. Summary of the main results of the clinical trias on HBS.
