- Email: [email protected]
Automatic Adaptive System Dialysis for Hemodialysis-Associated Hypotension and Intolerance: A Noncontrolled Multicenter Trial
Original Investigation Automatic Adaptive System Dialysis for Hemodialysis-Associated Hypotension and Intolerance: A Noncontrolled Multicenter Trial Luigi Colì, MD,1 Gaetano La Manna, MD,1 Giorgia Comai, MD,1 Mauro Ursino, MD,2 Davide Ricci, MD,1 Matteo Piccari, MD,1 Francesco Locatelli, MD,3 Salvatore Di Filippo, MD,3 Luciano Cristinelli, MD,4 Massimo Bacchi, MD,4 Alessandro Balducci, MD,5 Filippo Aucella, MD,6 Vincenzo Panichi, MD,7 Francesco Paolo Ferrandello, MD,7 Renzo Tarchini, MD,8 Domenica Lambertini, MD,8 Carlo Mura, MD,9 Giancarlo Marinangeli, MD,10 Ermanno Di Loreto, MD,10 Francesco Quarello, MD,11 Giacomo Forneris, MD,11 Maurizio Tancredi, MD,12 Massimo Morosetti, MD,13 Giuditta Palombo, MD,13 Marina Di Luca, MD,14 Mauro Martello, MD,14 Giuseppe Emiliani, MD,15 Roberto Bellazzi, MD,16 and Sergio Stefoni, MD1 Background: Hemodialysis is complicated by a high incidence of intradialytic hypotension and disequilibrium symptoms caused by hypovolemia and a decrease in extracellular osmolarity. Automatic adaptive system dialysis (AASD) is a proprietary dialysis system that provides automated elaboration of dialysate and ultrafiltration profiles based on the prescribed decrease in body weight and sodium content. Study Design: A noncontrolled (single arm), multicenter, prospective, clinical trial. Setting & Participants: 55 patients with intradialytic hypotension or disequilibrium syndrome in 15 dialysis units were studied over a 1-month interval using standard treatment (642 sessions) followed by 6 months using AASD (2,376 sessions). Intervention: AASD (bicarbonate dialysis with dialysate sodium concentration and ultrafiltration rate profiles determined by the automated procedure). Outcomes: Primary and major secondary outcomes were the frequency of intradialytic hypotension and symptoms (hypotensive events, headache, nausea, vomiting, and cramps), respectively. Results: More stable intradialytic systolic and diastolic blood pressures with lower heart rate were found using AASD compared with standard treatment. Sessions complicated by hypotension decreased from 58.7% ⫾ 7.3% to 0.9% ⫾ 0.6% (P ⬍ 0.001). The incidence of other disequilibrium syndrome symptoms was lower in patients receiving AASD. There were no differences in end-session body weight, interdialytic weight gain, or presession natremia between the standard and AASD treatment periods. Limitations: A noncontrolled (single arm) study, no crossover from AASD to standard treatment. Conclusions: This study shows the long-term clinical efficacy of AASD for intradialytic hypotension and disequilibrium symptoms in a large number of patients and dialysis sessions. Am J Kidney Dis. 58(1):93-100. © 2011 by the National Kidney Foundation, Inc. INDEX WORDS: Hemodialysis; hypotension; profiler; sodium.
T
he dialytic population continues to increase in age and number of comorbid conditions.1,2 The current incidence of intradialytic hypotension is high, affecting 10%-30% of sessions, as is the incidence of
other disequilibrium symptoms (cramps, 5%-20%; nausea and vomiting, 5%-15%; and headache, 5%).3-8 The major cause of intradialytic hypotension is hypovolemia due to (1) loss of vascular fluids by
From the 1U.O. di Nefrologia, Dialisi e Trapianto, Policlinico S. Orsola; 2Dipartimento di Elettronica, Informatica e Sistemistica, Università di Bologna, Bologna; 3Dipartimento di Nefrologia, Dialisi e Trapianto, Ospedale “Alessandro Manzoni”, Lecco; 4 U.O. Nefrologia e Dialisi, Ospedale “Guglielmo da Saliceto,” Piacenza; 5U.O.C. Nefrologia e Dialisi, Azienda Ospedaliera S. Giovanni-Addolorata, Roma; 6Divisione di Nefrologia e Dialisi, IRCCS “Casa Sollievo della Sofferenza,” S. Giovanni Rotondo; 7 U.O.C. di Nefrologia e Dialisi, Ospedale Versilia, Viareggio; 8U.O. di Nefrologia e Dialisi, Ospedale “Carlo Poma,” Mantova; 9Dipartimento di Nefrologia, Ospedale S. Donato, Arezzo; 10U.O.C. di Nefrologia e Dialisi, Ospedale Mona Santissima dello Splendore, Giulianova; 11Struttura Complessa di Nefrologia e Dialisi, Ospedale S. Giovanni in Bosco, Torino; 12U.O. di Nefrologia e Dialisi, Ospedale S. Liberatore, Atri; 13U.O. di Nefrologia e Dialisi, Ospedale “G.B.
Grassi,” Ostia; 14U.O. di Nefrologia e Dialisi, Azienda Ospedaliera S. Salvatore, Pesaro; 15U.O.A. Nefrologia e Dialisi, Azienda USL-Ravenna; and 16U.O. di Nefrologia e Dialisi, Ospedale di Vigevano, Vigevano, Italy. Received April 21, 2010. Accepted in revised form February 27, 2011. Originally published online May 23, 2011. Trial registration: www.ClinicalTrials.gov; study number: NCT01241994. Address correspondence to Sergio Stefoni, MD, Nephrology, Dialysis and Renal Transplantation Unit, S. Orsola University Hospital, Via Massarenti 9, 40138 Bologna, Italy. E-mail: [email protected] © 2011 by the National Kidney Foundation, Inc. 0272-6386/$36.00 doi:10.1053/j.ajkd.2011.01.030
Am J Kidney Dis. 2011;58(1):93-100
93
Colì et al
ultrafiltration and (2) decreased plasma osmolarity. This leads to both a decrease in fluid refilling from the extravascular to the intravascular space and a shift of fluids from the extracellular to the intracellular space. The extracellular osmolarity decrease caused by the rapid removal of solutes also is the main factor responsible for disequilibrium symptoms.3,8-13 Plasma refilling is the effect of 2 main forces acting at the capillary wall: one induced by the oncotic pressure gradient, and the other, by the hydrostatic pressure gradient. Both can be improved by acting on extracellular osmolarity. An increase in extracellular osmolarity primarily leads to increased extracellular volume by a fluid shift from the intracellular to the extracellular space. In turn, the increase in extracellular-interstitial volume improves plasma refilling through a decrease in oncotic pressure in the extracellular-interstitial fluid and an increase in extracellularinterstitial fluid hydrostatic pressure.14 Recently, dialysate sodium and ultrafiltration profiling also has been proposed as a better method to preserve intradialytic blood volume.15-21 However, studies of this approach did not base profiles on a standardized elaboration strategy and did not take into account patients’ sodium balance, yielding discordant clinical outcomes and increasing the risk of sodium/ water overload in the long term.17,18,22-28 Our experience with sodium and ultrafiltration profiles led to the development of a dialysis technique that we termed automatic adaptive system dialysis (AASD). This is the first profiled dialysis treatment based on a mathematical model aimed at the careful a priori automated determination of dialysate sodium and ultrafiltration profiles, taking into account the session’s sodium balance.29-33 To further evaluate the efficacy of AASD, we conducted a noncontrolled multicenter clinical trial, described here.
METHODS Setting and Participants The study was a noncontrolled (single arm) multicenter prospective clinical trial evaluating a group of patients receiving standard hemodialysis or hemodiafiltration (treatment at baseline) and presenting with dialysis intolerance. Patients were monitored for 1 month using the treatment they were receiving at baseline (the run-in phase) followed by a 6-month treatment with AASD divided into two 3-month periods. Patients selected for the study had intradialytic hypotension or disequilibrium syndrome with the onset of hypotensive events or other symptoms of disequilibrium (headache, nausea, vomiting, and cramps) during at least 1 session per week in the 6 months before starting the study. Intradialytic hypotension was defined as: (1) systolic blood pressure (SBP) at dialysis start ⱖ100 mm Hg and subsequent SBP ⱕ90 mm Hg, without symptoms (condition 1); (2) SBP at dialysis start ⬍100 mm Hg, followed by a decrease of at least 10%, symptomatic (condition 2); and (3) decrease in SBP after dialysis start of at least 25 mm Hg, symptomatic, with therapeutic intervention (condition 3).34 Inclusion criteria, evalu94
ated in the run-in phase, were: (1) appearance of conditions 2 or 3 in at least 30% of dialysis sessions and/or appearance of condition 1 in at least 1 session per week; and (2) appearance of symptoms (hypotensive events, headache, nausea, vomiting, and cramps) in at least 1 session per week. Exclusion criteria were chronic heart failure; poorly functioning vascular access (blood flow rate ⬍250 mL/min, recirculation ⬎5%, and difficult puncture); a permanent catheter as vascular access; unstable diabetes; presence of acute vascular, immunologic, or infectious disease or pregnancy; life expectancy less than 12 months; and equilibrated Kt/V ⬍1.2. The study design comprised 1 month of run-in using standard treatment (642 sessions evaluated) followed by 6 months using AASD (2,376 sessions evaluated).
Intervention and Measurements During the 6 months of AASD, the Formula 2000 Plus machine (Bellco Group Co, www.bellco.net) and the NC 1785 dialyzer (Bellco Group Co; synthetically modified cellulose membrane, 1.7 m2 surface area, and ␥-ray sterilization) were used. AASD is a proprietary dialysis system that provides automated determination of dialysate and ultrafiltration profiles based on the prescribed decrease in body weight and sodium content. Blood sodium was measured monthly before each session using a Modular ISE analyzer (Roche Diagnostics, www.roche.com), which uses indirect potentiometry. The mathematical model used in AASD is a recently updated version of our previous models that has been validated in vivo.35-37 Patients’ dry weights were evaluated using a clinical method used in everyday practice and took into account the last case history, dialysis summary of the last few sessions, onset of specific clinical signs (dyspnea, headache, edema, thirst, etc), chest radiograph (vascular peduncle), and laboratory data for hemodilution or hemoconcentration.38 The sodium mass to be removed for each session was evaluated using 2 methods. The sodium to be removed was imposed a priori in 47 patients (85.4%) for subsequent profile determination by AASD in the range of 80-100 mmol/kg of ultrafiltration goal, according to a clinical evaluation comprising variation of thirst, interdialytic body weight gain, dry body weight, intra- and interdialytic blood pressure, intradialytic symptoms, postdialysis wellness (dizziness, cramps, and asthenia), laboratory parameters (hemoglobin, protein, and albumin), and others.38 The sodium mass to be removed in the remaining 8 patients (14.6%) was calculated by AASD as a function of end-session natremia (the natremia target). The natremia target was determined as mean blood natremia in the last 12 dialysis sessions using standard treatment, measured immediately before the disconnection phase. In both cases, AASD calculates the sodium and ultrafiltration profile able to satisfy the constraints of the algorithm, respecting the input data, which are session length, weight loss, and sodium mass to be removed. Pre- and postsession body weight and body weight decrease were collected for each study period (1-month run-in period and 6 months of AASD): SBP, diastolic blood pressure (DBP), and heart rate presession, intrasession (hourly), and 1 hour after the end of the session. Also, the incidence of disequilibrium syndrome symptoms (hypotension, headache, nausea, vomiting, and cramps) was recorded as a percentage of symptomatic sessions (sessions with at least 1 event).
Outcomes The primary outcome of this study was the efficacy of AASD on intradialytic blood pressure. Secondary outcomes were the efficacy of AASD on other symptoms, including hypotensive events, headache, nausea, vomiting, and cramps. Am J Kidney Dis. 2011;58(1):93-100
Automatic Adaptive System Dialysis for Hypotension The following parameters were monitored: (1) blood pressure and heart rate pre- (at the beginning), intra- (hourly), and post session (1 hour after); and (2) disequilibrium symptoms (hypotensive events, headache, nausea, vomiting, and cramps), expressed as a percentage of symptomatic sessions. At the same time, patients’ sodium/water balance was assessed monthly, monitoring dry body weight, mean interdialytic weight gain, and mean presession blood sodium level.
Statistical Analysis All data are expressed as mean ⫾ standard error. Differences between baseline treatment and the first and second 3-month periods of AASD treatment were tested using analysis of variance test at each time of dialysis session. Longitudinal mixed models with an unstructured covariance matrix were used to analyze repeated measures of SBP, DBP, and heart rate. The advantage of linear mixed models is that they permit examination of variations among patients while adjusting for correlation within individuals. Our analysis included random effects for intercepts and for the effect of time. This implies that change over time may vary between patients. Treatment was included in the analysis as a covariate along with the time ⫻ treatment interaction term. Similarly, a longitudinal mixed model was implemented for weight and sodium. Variability in percentage of intradialytic symptoms during the study period was assessed using repeated-measure analysis of variance followed by post hoc comparison with Tukey test. Significant differences were defined as P ⬍ 0.05. Data were analyzed using SPSS for Windows software package (version 9.0.1; SPSS, www.spss.com) and SAS software (SAS Institute Inc, www.sas.com). Informed consent and the approval of the study protocol by the ethical committee of each center were collected. The EudraCT registry number for the protocol is 106/2004/U/Oss.
RESULTS An Italian noncontrolled multicenter (15 dialysis units) prospective clinical trial was carried out from September 2007 to September 2008 to confirm the clinical efficacy of AASD in the long term (6 months) in 55 patients during 2,376 dialysis sessions. Patients were all older than 18 years and at baseline had been undergoing hemodialysis 3 times a week for at least 6 months. Patient and dialysis session characteristics are listed in Table 1. Seven patients dropped out of the study: 1 patient died during the run-in period, and another, after 4 months of AASD treatment (from cardiac complications unrelated to the treatment itself); 2 patients dropped out for kidney transplant after 2 months of AASD; 1 patient stopped the treatment of their own volition after 1 month of AASD because of difficulty adapting to changes in dialysis shift/room; and 2 patients dropped out 3 months after AASD for vascular access problems. Systolic Blood Pressure Average predialysis SBPs were 119.6 ⫾ 3.8 mm Hg during the run-in phase, 126.6 ⫾ 1.5 mm Hg in the first 3 months of AASD (P ⫽ 0.08), and 130.5 ⫾ 1.5 mm Hg in the second 3 months of AASD (P ⫽ 0.005). Intradialytic SBP for the first 3 months of AASD was Am J Kidney Dis. 2011;58(1):93-100
Table 1. Patient and Dialysis Session Characteristics Characteristic
Value
No. of patients Men/women Age (y) Dialysis vintage (mo) Mean residual glomerular filtration (mL/min) Origin of kidney disease Diabetes Chronic glomerulonephritis Vascular nephropathy Chronic interstitial nephritis ESKD
55 24:31 61 ⫾ 11 53 ⫾ 18 ⬍3
No. of sessions studied No. of standard treatment sessions (run-in phase) No. of AASD sessions
3,018 642
No. of patients in each treatment group at baseline Standard bicarbonate dialysis (low-flux synthetic membrane) Hemodiafiltration (high-flux synthetic membrane)
18 (33) 5 (9) 14 (25) 3 (5) 15 (27)
2,376
37 (67) 18 (33)
No. of patients in each session length category 3.5 h 4h 4.5 h
11 (20) 39 (70.9) 5 (9.1)
No. of patients in each sodium mass removal method group Set a prioria By natremia targetb
47 (85.5) 8 (14.5)
Note: Unless otherwise indicated, values shown are number, mean ⫾ standard error, or number (percentage). Abbreviations: AASD, automatic adaptive system dialysis; ESKD, end-stage kidney disease. a Subsequent profile elaboration by AASD in the range of 80-100 mmol/kg of ultrafiltration goal, according to a clinical evaluation. b Calculated by AASD as a function of end-session natremia.
significantly higher than in the run-in phase at 1 hour of treatment (119.3 ⫾ 1.5 vs 109.8 ⫾ 3.6 mm Hg; P ⫽ 0.02) and at 2 hours (115.3 ⫾ 1.6 vs 104.1 ⫾ 3.7 mm Hg; P ⫽ 0.008). SBP in the second 3 months was higher than during the run-in phase at 1 (124.3 ⫾ 1.7 vs 113.4 ⫾ 3.9 mm Hg; P ⫽ 0.01), 2 (121.0 ⫾ 1.4 vs 109.8 ⫾ 3.6 mm Hg; P ⫽ 0.004), and 3 hours of treatment (118.8 ⫾ 1.6 vs 104.1 ⫾ 3.7 mm Hg; P ⬍ 0.001). Compared with the run-in phase, postdialytic SBP was higher for both the first and second 3-month AASD treatment periods: 122.0 ⫾ 1.8 versus 109.9 ⫾ 3.7 mm Hg (P ⫽ 0.006) and 124.8 ⫾ 1.7 versus 109.9 ⫾ 3.7 mm Hg (P ⬍ 0.001), respectively. Patients started on average with an SBP of 124.11 mm Hg, which changed by ⫺3.06 mm Hg (P ⬍ 0.001) each hour. Compared with the run-in phase, 95
Colì et al
Figure 1. Comparison of (A) systolic blood pressure (SBP), (B) diastolic blood pressure (DBP), and (C) heart rate in the 2 automatic adaptive system dialysis (AASD) treatment periods (first and second 3 months) versus the run-in phase (standard treatment with dialysis or hemodiafilration). Times shown are predialytic; 1-, 2-, 3-, and 4-hour times of dialysis; and postdialytic. (A) *P ⫽ 0.01 for second 3-month period of AASD versus the run-in phase. (B) *P ⫽ 0.04. (C) **P ⬍ 0.001 for the first and second 3-month periods versus the run-in phase (longitudinal mixed models).
in the first and second 3-month periods of AASD treatment, patients had an increase in SBP over time of 2.7 ⫾ 2.1 (P ⫽ 0.2) and 5.4 ⫾ 2.1 mm Hg (P ⫽ 0.01), respectively (Fig 1A). Diastolic Blood Pressure Predialysis DBP values did not show significant differences between the run-in phase and the first and second 3-month periods of AASD treatment. In the first 3 months of AASD, intradialytic DBP was higher than during the run-in phase at 2 hours (68.1 ⫾ 0.9 vs 63.2 ⫾ 96
2.0 mm Hg; P ⫽ 0.04). In the second 3 months of AASD, intradialytic DBP was higher than during the run-in phase at 2 (69.4 ⫾ 0.8 vs 64.3 ⫾ 1.9 mm Hg; P ⫽ 0.03), 3 (69.4 ⫾ 0.8 vs 63.2 ⫾ 2.0 mm Hg; P ⫽ 0.006), and 4 hours (69.4 ⫾ 1.0 vs 62.9 ⫾ 2.5 mm Hg; P ⫽ 0.01). Postdialysis DBPs were significantly different between AASD and the run-in phase during the second 3 months (72.1 ⫾ 0.7 vs 66.6 ⫾ 2.0 mm Hg; P ⫽ 0.01). Patients started on average with a DBP of 68.29 mm Hg, which changed by ⫺0.98 mm Hg (P ⫽ 0.001) each hour. In the first and second 3-month Am J Kidney Dis. 2011;58(1):93-100
Automatic Adaptive System Dialysis for Hypotension
Figure 2. Symptoms of disequilibrium syndrome in each month of automatic adaptive system dialysis (AASD) treatment versus the run-in phase (standard treatment).
periods of AASD treatment, patients had a slight increase in DBP over time compared with the run-in phase, of 1.05 ⫾ 1.02 (P ⫽ 0.3) and 2.05 ⫾ 1.03 mm Hg (P ⫽ 0.04), respectively (Fig 1B). Heart Rate Intradialytic heart rate values were lower in the first 3 months with AASD than in the run-in phase, a difference that was statistically significant at the 2-hour time (73.9 ⫾ 0.8 vs 78.4 ⫾ 1.5 beats/min; P ⫽ 0.03, ANOVA). In the second 3 months of AASD, significant differences at the 1-, 2-, 3-, and 4-hour times were noted. Postdialysis heart rate values were significantly lower in the second 3 months (77.3 ⫾ 1.2 vs 72.6 ⫾ 0.7 beats/min; P ⫽ 0.01). By means of longitudinal mixed models, patients started on average with a heart rate of 75.2 beats/min, which changed by 0.6 beat/min (P ⫽ 0.02) each hour. In the first and second 3-month periods of AASD treatment, changes in patients’ heart rates over time were statistically significantly different compared with the run-in phase, with changes of ⫺2.8 ⫾ 0.9 (P ⬍ 0.001) and ⫺4.0 ⫾ 0.9 beats/min (P ⬍ 0.001), respectively (Fig 1C). Intradialytic Hypotension Events and Symptoms The frequency of each symptom tracked for the study decreased in a statistically significant fashion during the AASD treatment period. The percentage of sessions affected by hypotension events decreased from 58.7% ⫾ 7.3% to 32.7% ⫾ 6.1% in the first month of AASD (P ⫽ 0.008) and to 0.9% ⫾ 0.6% in the sixth month (P ⬍ 0.001), with Am J Kidney Dis. 2011;58(1):93-100
significant differences throughout the AASD period. The incidence of headaches decreased from 32.6% ⫾ 8.1% of sessions to 18.7% ⫾ 5.0% in the first month of AASD, although this change was not statistically significant (P ⫽ 0.4). However, the percentage of sessions with headaches was significantly lower in the second (10.0% ⫾ 3.1%; P ⫽ 0.01), third (6.6% ⫾ 2.7%; P ⬍ 0.001), fourth (1.4% ⫾ 1.0%; P ⬍ 0.001), and fifth months (0.4% ⫾ 0.4%; P ⬍ 0.001) of AASD compared with the run-in phase, and no headaches were reported in the sixth month. The percentage of sessions with nausea decreased from 26.8% ⫾ 6.7% to 11.2% ⫾ 4.1% in the first month of AASD (P ⫽ 0.07) to 6.6% ⫾ 3.2% in the second month (P ⫽ 0.005), to 3.3% ⫾ 1.8% in the third month (P ⬍ 0.001), and to 0.5% ⫾ 0.5% in the fourth month (P ⬍ 0.001); no instances of nausea were reported in the fifth and sixth months. Vomiting went from an incidence of 13.1% ⫾ 4.3% in the run-in phase to 5.9% ⫾ 2.4% in the first month of AASD, but this change was not statistically significant (P ⫽ 0.3). However, the percentage of sessions with vomiting was 2.3% ⫾ 1.1% in the second month (P ⫽ 0.01 compared with the run-in phase), 0.5% ⫾ 0.5% in the third and fourth months (P ⬍ 0.001), and none in the fifth month. The percentage of sessions with cramps decreased to 3.5% ⫾ 1.3% in the first month of AASD (P ⫽ 0.002) to 1.7% ⫾ 1.0% in the second month (P ⬍ 0.001), to 0.5% ⫾ 0.5% in the third month (P ⬍ 0.001), to 0.8% ⫾ 0.5% in the fourth month (P ⬍ 0.001), and to 0.4% ⫾ 0.4% in the fifth and sixth months (P ⬍ 0.001; Fig 2). 97
Colì et al
Sodium Balance and Body Weight Assessment Results from the mixed model showed no significant improvement in sodium over time for patients with AASD treatment. The average value of sodium removed was the same during the run-in phase and the AASD treatment period at 378 ⫾ 95 and 367 ⫾ 84 mmol, respectively (P ⫽ 0.08). In patients treated according to natremia target in the AASD phase, the average amount of sodium removed was 384 ⫾ 91 mmol, which did not differ significantly from the value in the run-in phase (P ⫽ 0.09). Average predialysis sodium level for the run-in phase was 137 ⫾ 0.97 mEq/L, whereas in the AASD phase, levels were 137.54 ⫾ 0.49 mEq/L (P ⫽ 0.08) in the third month and 136.77 ⫾ 0.81 mEq/L in the sixth month (P ⫽ 0.08). There were no significant differences in end-ofsession body weights, which were 68.32 ⫾ 1.4 kg in the run-in phase, 68.49 ⫾ 1.4 kg in the first month of AASD, 68.45 ⫾ 1.4 kg in the second month, 68.31 ⫾ 1.4 kg in the third month, 68.7 ⫾ 1.5 kg in the fourth month, 68.4 ⫾ 1.5 kg in the fifth month, and 67.7 ⫾ 1.6 kg in the sixth month. Mean interdialytic weight gain did not vary significantly, being 3.23 ⫾ 0.1 kg in the run-in phase, 3.16 ⫾ 0.1 kg in the first month of AASD, 3.12 ⫾ 0.1 kg in the second month, 3.18 ⫾ 0.1 kg in the third month, 3.18 ⫾ 0.1 kg in the fourth month, 3.07 ⫾ 0.1 kg in the fifth month, and 2.96 ⫾ 0.1 in the sixth month.
DISCUSSION The pathogenesis of intradialytic complications is multifactorial. The main factor is plasma water loss through ultrafiltration, with a decrease in extracellular space and blood volume. The decrease in intradialytic osmolarity causes an additional loss of plasma water due to the shift from the extracellular to the intracellular space.11 The extracellular space decrease induces hypotension, whereas the intracellular space increase induces disequilibrium symptoms.3,12,13 Many investigators agree that preservation of blood volume has a pivotal role in intradialytic hemodynamic stability.3,9,10,39-44 Because sodium is the major determinant of extracellular space osmolarity, dialysate sodium concentration can be manipulated to decrease symptoms during dialysis.45 More recently, this assumption led to the introduction of dialysate sodium profiling. Our technical approach to profiled dialysis is based on the clinical application of an automatic computer-based program, which in turn is based on a new mathematical kinetic model for processing intradialytic sodium and ultrafiltration profiles. The aim of this strategy is to preserve blood volume during dialysis by boosting refilling and decreasing the shift from the extracellu98
lar to the intracellular space. AASD allows for the elaboration of profiles at the beginning of each dialysis session while simultaneously taking into account the patient’s sodium balance at each dialysis session. Results of this study show that AASD significantly improves intradialytic blood pressure, decreases the incidence of hypotensive events and disequilibrium symptoms compared with basal treatment, and does not induce symptoms of sodium/water overload. With AASD, the intradialytic curves of SBP and DBP were more stable from the first 3-month period of treatment, with quicker postsession recovery to baseline values for SBP. AASD also decreased the incidence of intradialytic hypotensive events. The stabilization of intradialytic blood pressure observed in the study confirms the efficacy of AASD in enhancing compensation of blood volume decrease. This water drainage is exploited by the simultaneous ultrafiltration profile, which also reaches its peak between the first 40 and 60 minutes of the session, with a lower ultrafiltration rate in the second half of the session. Heart rate, an index of sympathetic system response, was more stable with AASD compared with the standard treatment during the run-in phase. Heart rate increases during hemodialysis, and Redaelli et al46 showed that it reached the peak value 1 hour after the end of the session, recovering the baseline value after 8 hours. Krepel et al47 showed significant correlation between relative blood volume and heart rate during a dialysis session, suggesting that a decrease in relative blood volume stimulates the autonomic nervous system to minimize the decrease in blood pressure by increasing heart rate. There is still much discordance about the correlation between heart rate and blood volume decrease. Several investigators have noted that intradialytic hypotension–prone patients more frequently are older and have diabetes and may have a minor adaptive response of heart rate due to impaired autonomic control.43,48,49 Of note, from the first 3-month period of AASD, postsession heart rate values were not significantly different from presession values. This result is related to lower intra- and postdialytic sympathetic activity, confirming that there is less need for compensation. Several studies have shown that use of profiled dialysate sodium can decrease the incidence of disequilibrium-symptomatic dialysis sessions by decreasing the magnitude of the intradialytic decrease in plasma osmolarity and thereby reducing the shift from the extracellular to the intracellular space.50-52 Our study showed that the percentage of symptomatic sessions was decreased significantly from the first month with AASD in terms of muscle cramps and from the second month in terms of nausea, vomiting, and headache. Am J Kidney Dis. 2011;58(1):93-100
Automatic Adaptive System Dialysis for Hypotension
When processing profiles, AASD takes into account the quantity of sodium to be removed, which is an input datum for AASD at the beginning of each session. This sodium balance of the AASD session remained the same as in the run-in phase using standard treatment, and no signs of sodium/water overload appeared in patients during the AASD treatment period: dry body weight remained stable, as did interdialytic body weight gain and natremia levels at the start of each session. Our study has several limitations. It lacks blinding and is not a crossover randomized study. Patients enrolled in the study were clinically monitored during the run-in and AASD treatment periods, each patient serving as his or her own control. However, the study design lacks any carryover effects of AASD on the subsequent basal treatment. In conclusion, our multicenter study confirmed the long-term clinical efficacy of AASD on dialysis intolerance in a large number of patients. In particular, AASD maintained stable intradialytic blood pressure and decreased the appearance of intradialytic disequilibrium symptoms, taking into account the patient’s sodium balance. Furthermore, the technique’s automatic determination of profiles by the AASD program proved both reliable and easy to implement in clinical practice in all 15 dialysis units participating in the study.
ACKNOWLEDGEMENTS Support: The study was supported in part by the Fondazione Cassa di Risparmio in Bologna, project no. 2007/0234 “Innovazioni diagnostiche e terapeutiche nell’approccio alle criticità immunologiche del paziente nel percorso Dialisi-Trapianto di Rene” (Principal Investigator, Prof Stefoni). Financial Disclosure: The authors declare that they have no relevant financial interests.
REFERENCES 1. de Jager DJ, Grootendorst DC, Jager KJ, et al. Cardiovascular and noncardiovascular mortality among patients starting dialysis. JAMA. 2009;302(16):1782-1789. 2. Collins AJ, Foley RN, Gilbertson DT, Chen SC. The state of chronic kidney disease, ESRD, and morbidity and mortality in the first year of dialysis. Clin J Am Soc Nephrol. 2009;4(suppl 1):S511. 3. Daugirdas JT. Pathophysiology of dialysis hypotension: an update. Am J Kidney Dis. 2001;38(4)(suppl 4):S11-17. 4. Donauer J. Hemodialysis-induced hypotension: impact of technologic advances. Semin Dial. 2004;17(5):333-335. 5. Sulowicz W, Radziszewski A. Dialysis induced hypotension—a serious clinical problem in renal replacement therapy. Med Pregl. 2007;60(2):14-20. 6. Germin Petrovic D. Effect of automated blood volume control on the incidence of intra-dialysis hypotension. Acta Med Croatica. 2003;57(1):17-22. 7. Stegmayr BG. Ultrafiltration and dry weight—what are the cardiovascular effects? Artif Organs. 2003;27(3):227-229. 8. Daugirdas JT, Blake PG, Ing TS. Handbook of Dialysis. Philadelphia, PA: Lippincott Williams & Wilkins; 2007. Am J Kidney Dis. 2011;58(1):93-100
9. Henrich WL. Hemodynamic stability during dialysis and cardiovascular disease in end-stage renal disease patients. Am J Kidney Dis. 1999;33(6):XLIX-LII. 10. Daugirdas JT. Dialysis hypotension: a hemodynamic analysis. Kidney Int. 1991;39(2):233-246. 11. Ebel H, Laage C, Keuchel M, et al. Impact of profile haemodialysis on intra-/extracellular fluid shifts and the release of vasoactive hormones in elderly patients on regular dialysis treatment. Nephron. 1997;75(3):264-271. 12. Trinh-Trang-Tan MM, Cartron JP, Bankir L. Molecular basis for the dialysis disequilibrium syndrome: altered aquaporin and urea transporter expression in the brain. Nephrol Dial Transplant. 2005;20(9):1984-1988. 13. Chen CL, Lai PH, Chou KJ, et al. A preliminary report of brain edema in patients with uremia at first hemodialysis: evaluation by diffusion-weighted MR imaging. AJNR Am J Neuroradiol. 2007;28(1):68-71. 14. Guyton AC. Textbook of Medical Physiology. Philadelphia, PA: W. B. Saunders Co; 1991. 15. Di Filippo S, Manzoni C, Andrulli S, et al. Sodium removal during pre-dilution haemofiltration. Nephrol Dial Transplant. 2003; 18(suppl 7):vii31-36; discussion vii57-58. 16. Rabindranath KS, Strippoli GF, Daly C, et al. Haemodiafiltration, haemofiltration and haemodialysis for end-stage kidney disease. Cochrane Database Syst Rev. 2006;4:CD006258. 17. Tang HL, Wong SH, Chu KH, et al. Sodium ramping reduces hypotension and symptoms during haemodialysis. Hong Kong Med J. 2006;12(1):10-14. 18. Raja RM. Sodium profiling in elderly haemodialysis patients. Nephrol Dial Transplant. 1996;11(suppl 8):42-45. 19. Straver B, De Vries PM, Donker AJ, et al. The effect of profiled hemodialysis on intra-dialytic hemodynamics when a proper sodium balance is applied. Blood Purif. 2002;20(4):364369. 20. Song JH, Park GH, Lee SY, et al. Effect of sodium balance and the combination of ultrafiltration profile during sodium profiling hemodialysis on the maintenance of the quality of dialysis and sodium and fluid balances. J Am Soc Nephrol. 2005;16(1):237246. 21. Agarwal R, Kelley K, Light RP. Diagnostic utility of blood volume monitoring in haemodialysis patients. Am J Kidney Dis. 2008;51:242-254. 22. Movilli E, Camerini C, Viola BF, et al. Blood volume changes during three different profiles of dialysate sodium variation with similar intra-dialytic sodium balances in chronic hemodialyzed patients. Am J Kidney Dis. 1997;30(1):58-63. 23. Zhou YL, Liu HL, Duan XF, et al. Impact of sodium and ultrafiltration profiling on haemodialysis-related hypotension. Nephrol Dial Transplant. 2006;21(11):3231-3237. 24. Kooman JP, van der Sande F, Leunissen K, et al. Sodium balance in hemodialysis therapy. Semin Dial. 2003;16(5):351-355. 25. Locatelli F, Covic A, Chazot C, et al. Optimal composition of the dialysate, with emphasis on its influence on blood pressure. Nephrol Dial Transplant. 2004;19(4):785-796. 26. Moret K, Aalten J, van den Wall Bake W, et al. The effect of sodium profiling and feedback technologies on plasma conductivity and ionic mass balance: a study in hypotension-prone dialysis patients. Nephrol Dial Transplant. 2006;21(1):138-144. 27. Oliver MJ, Edwards LJ, Churchill DN. Impact of sodium and ultrafiltration profiling on hemodialysis-related symptoms. J Am Soc Nephrol. 2001;12(1):151-156. 28. Davenport A. Can advances in hemodialysis machine technology prevent intradialytic hypotension? Semin Dial. 2009;22(3): 231-236. 29. Ursino M, Colì L, La Manna G, et al. A simple mathematical model of intradialytic sodium kinetics: “in vivo” validation 99
Colì et al during hemodialysis with constant or variable sodium. Int J Artif Organs. 1996;19(7):393-403. 30. Colì L, Ursino M, Dalmastri V, et al. A simple mathematical model applied to selection of the sodium profile during profiled haemodialysis. Nephrol Dial Transplant. 1998;13(2):404-416. 31. Colì L, Ursino M, De Pascalis A, et al. Evaluation of intradialytic solute and fluid kinetics. Setting up a predictive mathematical model. Blood Purif. 2000;18(1):37-49. 32. Ursino M, Colì L, Brighenti C. Prediction of solute kinetics, acid base status, and blood volume changes during profiled hemodialysis. Ann Biomed Eng. 2000;28:204-216. 33. Colì L, Ursino M, Donati G, et al. Clinical application of sodium profiling in the treatment of intradialytic hypotension. Int J Artif Organs. 2003;26:715-722. 34. Santoro A, Mancini E, Basile C, et al. Blood volume controlled hemodialysis in hypotension-prone patients: a randomized, multicenter controlled trial. Kidney Int. 2002;62(3):1034-1045. 35. Ursino M, Colì L, Dalmastri V, et al. An algorithm for the rational choice of sodium profile during hemodialysis. Int J Artif Organs. 1997;20(12):659-672. 36. Colì L, Bonomini M, La Manna G, et al. Clinical use of profiled hemodialysis. Artif Organs. 1998;22(9):724-730. 37. Ursino M, Colì L, Magosso E, et al. A mathematical model for the prediction of solute kinetics, osmolarity and fluid volume changes during hemodiafiltration with on-line regeneration of ultrafiltrate (HFR). Int J Artif Organs. 2006;29(11):10311041. 38. Charra B. Fluid balance, dry weight, and blood pressure in dialysis. Hemodial Int. 2007;11:21-31. 39. Beige J, Sone J, Sharma AM, et al. Computational analysis of blood volume curves and risk of intradialytic morbid events in hemodialysis. Kidney Int. 2000;58(4):1805-1809. 40. Locatelli F, Buoncristiani U, Canaud B, et al. Haemodialysis with on-line monitoring equipment: tools or toys? Nephrol Dial Transplant. 2005;20(1):22-33. 41. Dasselaar JJ, Huisman RM, de Jong PE, et al. Measurement of relative blood volume changes during haemodialysis:
100
merits and limitations. Nephrol Dial Transplant. 2005;20(10): 2043-2049. 42. Van der Sande FM, Kooman JP, Leunissen KM. Intradialytic hypotension—new concepts on an old problem. Nephrol Dial Transplant. 2000;15(11):1746-1748. 43. Andrulli S, Colzani S, Mascia F, et al. The role of blood volume reduction in the genesis of intradialytic hypotension. Am J Kidney Dis. 2002;40(6):1244-1254. 44. Van Kuijk WH, Wirtz JJ, Grave W, et al. Vascular reactivity during combined ultrafiltration-haemodialysis: influence of dialysate sodium. Nephrol Dial Transplant. 1996;11(2):323-328. 45. Mann H, Stiller S. Sodium modeling. Kidney Int Suppl. 2000;76:S79-88. 46. Redaelli B, Locatelli F, Limido D, et al. Effect of a new model of hemodialysis potassium removal on the control of ventricular arrhythmias. Kidney Int. 1996;50(2):609-617. 47. Krepel HP, Nette RW, Akçahüseyin E, et al. Variability of relative blood volume during haemodialysis. Nephrol Dial Transplant. 2000;15(5):673-679. 48. Lin YF, Wang JY, Shum AY, et al. Role of plasma catecholamines, autonomic, and left ventricular function in normotensive and hypotension prone dialysis patients. ASAIO J. 1993;39(4): 946-953. 49. Jassal SV, Douglas JF, Stout RW. Prevalence of central autonomic neuropathy in elderly dialysis patients. Nephrol Dial Transplant. 1998;13(7):1702-1708. 50. De Vries PM, Olthof CG, Solf A, et al. Fluid balance during haemodialysis and haemofiltration: the effect of dialysate sodium and a variable ultrafiltration rate. Nephrol Dial Transplant. 1991; 6(4):257-263. 51. Swartz RD, Somermeyer MG, Hsu CH. Preservation of plasma volume during hemodialysis depends on dialysate osmolality. Am J Nephrol. 1982;2(4):189-194. 52. Petitclerc T, Jacobs C. Dialysis sodium concentration: what is optimal and can it be individualized? Nephrol Dial Transplant. 1995;10(5):596-599.
Am J Kidney Dis. 2011;58(1):93-100
T
he dialytic population continues to increase in age and number of comorbid conditions.1,2 The current incidence of intradialytic hypotension is high, affecting 10%-30% of sessions, as is the incidence of
other disequilibrium symptoms (cramps, 5%-20%; nausea and vomiting, 5%-15%; and headache, 5%).3-8 The major cause of intradialytic hypotension is hypovolemia due to (1) loss of vascular fluids by
From the 1U.O. di Nefrologia, Dialisi e Trapianto, Policlinico S. Orsola; 2Dipartimento di Elettronica, Informatica e Sistemistica, Università di Bologna, Bologna; 3Dipartimento di Nefrologia, Dialisi e Trapianto, Ospedale “Alessandro Manzoni”, Lecco; 4 U.O. Nefrologia e Dialisi, Ospedale “Guglielmo da Saliceto,” Piacenza; 5U.O.C. Nefrologia e Dialisi, Azienda Ospedaliera S. Giovanni-Addolorata, Roma; 6Divisione di Nefrologia e Dialisi, IRCCS “Casa Sollievo della Sofferenza,” S. Giovanni Rotondo; 7 U.O.C. di Nefrologia e Dialisi, Ospedale Versilia, Viareggio; 8U.O. di Nefrologia e Dialisi, Ospedale “Carlo Poma,” Mantova; 9Dipartimento di Nefrologia, Ospedale S. Donato, Arezzo; 10U.O.C. di Nefrologia e Dialisi, Ospedale Mona Santissima dello Splendore, Giulianova; 11Struttura Complessa di Nefrologia e Dialisi, Ospedale S. Giovanni in Bosco, Torino; 12U.O. di Nefrologia e Dialisi, Ospedale S. Liberatore, Atri; 13U.O. di Nefrologia e Dialisi, Ospedale “G.B.
Grassi,” Ostia; 14U.O. di Nefrologia e Dialisi, Azienda Ospedaliera S. Salvatore, Pesaro; 15U.O.A. Nefrologia e Dialisi, Azienda USL-Ravenna; and 16U.O. di Nefrologia e Dialisi, Ospedale di Vigevano, Vigevano, Italy. Received April 21, 2010. Accepted in revised form February 27, 2011. Originally published online May 23, 2011. Trial registration: www.ClinicalTrials.gov; study number: NCT01241994. Address correspondence to Sergio Stefoni, MD, Nephrology, Dialysis and Renal Transplantation Unit, S. Orsola University Hospital, Via Massarenti 9, 40138 Bologna, Italy. E-mail: [email protected] © 2011 by the National Kidney Foundation, Inc. 0272-6386/$36.00 doi:10.1053/j.ajkd.2011.01.030
Am J Kidney Dis. 2011;58(1):93-100
93
Colì et al
ultrafiltration and (2) decreased plasma osmolarity. This leads to both a decrease in fluid refilling from the extravascular to the intravascular space and a shift of fluids from the extracellular to the intracellular space. The extracellular osmolarity decrease caused by the rapid removal of solutes also is the main factor responsible for disequilibrium symptoms.3,8-13 Plasma refilling is the effect of 2 main forces acting at the capillary wall: one induced by the oncotic pressure gradient, and the other, by the hydrostatic pressure gradient. Both can be improved by acting on extracellular osmolarity. An increase in extracellular osmolarity primarily leads to increased extracellular volume by a fluid shift from the intracellular to the extracellular space. In turn, the increase in extracellular-interstitial volume improves plasma refilling through a decrease in oncotic pressure in the extracellular-interstitial fluid and an increase in extracellularinterstitial fluid hydrostatic pressure.14 Recently, dialysate sodium and ultrafiltration profiling also has been proposed as a better method to preserve intradialytic blood volume.15-21 However, studies of this approach did not base profiles on a standardized elaboration strategy and did not take into account patients’ sodium balance, yielding discordant clinical outcomes and increasing the risk of sodium/ water overload in the long term.17,18,22-28 Our experience with sodium and ultrafiltration profiles led to the development of a dialysis technique that we termed automatic adaptive system dialysis (AASD). This is the first profiled dialysis treatment based on a mathematical model aimed at the careful a priori automated determination of dialysate sodium and ultrafiltration profiles, taking into account the session’s sodium balance.29-33 To further evaluate the efficacy of AASD, we conducted a noncontrolled multicenter clinical trial, described here.
METHODS Setting and Participants The study was a noncontrolled (single arm) multicenter prospective clinical trial evaluating a group of patients receiving standard hemodialysis or hemodiafiltration (treatment at baseline) and presenting with dialysis intolerance. Patients were monitored for 1 month using the treatment they were receiving at baseline (the run-in phase) followed by a 6-month treatment with AASD divided into two 3-month periods. Patients selected for the study had intradialytic hypotension or disequilibrium syndrome with the onset of hypotensive events or other symptoms of disequilibrium (headache, nausea, vomiting, and cramps) during at least 1 session per week in the 6 months before starting the study. Intradialytic hypotension was defined as: (1) systolic blood pressure (SBP) at dialysis start ⱖ100 mm Hg and subsequent SBP ⱕ90 mm Hg, without symptoms (condition 1); (2) SBP at dialysis start ⬍100 mm Hg, followed by a decrease of at least 10%, symptomatic (condition 2); and (3) decrease in SBP after dialysis start of at least 25 mm Hg, symptomatic, with therapeutic intervention (condition 3).34 Inclusion criteria, evalu94
ated in the run-in phase, were: (1) appearance of conditions 2 or 3 in at least 30% of dialysis sessions and/or appearance of condition 1 in at least 1 session per week; and (2) appearance of symptoms (hypotensive events, headache, nausea, vomiting, and cramps) in at least 1 session per week. Exclusion criteria were chronic heart failure; poorly functioning vascular access (blood flow rate ⬍250 mL/min, recirculation ⬎5%, and difficult puncture); a permanent catheter as vascular access; unstable diabetes; presence of acute vascular, immunologic, or infectious disease or pregnancy; life expectancy less than 12 months; and equilibrated Kt/V ⬍1.2. The study design comprised 1 month of run-in using standard treatment (642 sessions evaluated) followed by 6 months using AASD (2,376 sessions evaluated).
Intervention and Measurements During the 6 months of AASD, the Formula 2000 Plus machine (Bellco Group Co, www.bellco.net) and the NC 1785 dialyzer (Bellco Group Co; synthetically modified cellulose membrane, 1.7 m2 surface area, and ␥-ray sterilization) were used. AASD is a proprietary dialysis system that provides automated determination of dialysate and ultrafiltration profiles based on the prescribed decrease in body weight and sodium content. Blood sodium was measured monthly before each session using a Modular ISE analyzer (Roche Diagnostics, www.roche.com), which uses indirect potentiometry. The mathematical model used in AASD is a recently updated version of our previous models that has been validated in vivo.35-37 Patients’ dry weights were evaluated using a clinical method used in everyday practice and took into account the last case history, dialysis summary of the last few sessions, onset of specific clinical signs (dyspnea, headache, edema, thirst, etc), chest radiograph (vascular peduncle), and laboratory data for hemodilution or hemoconcentration.38 The sodium mass to be removed for each session was evaluated using 2 methods. The sodium to be removed was imposed a priori in 47 patients (85.4%) for subsequent profile determination by AASD in the range of 80-100 mmol/kg of ultrafiltration goal, according to a clinical evaluation comprising variation of thirst, interdialytic body weight gain, dry body weight, intra- and interdialytic blood pressure, intradialytic symptoms, postdialysis wellness (dizziness, cramps, and asthenia), laboratory parameters (hemoglobin, protein, and albumin), and others.38 The sodium mass to be removed in the remaining 8 patients (14.6%) was calculated by AASD as a function of end-session natremia (the natremia target). The natremia target was determined as mean blood natremia in the last 12 dialysis sessions using standard treatment, measured immediately before the disconnection phase. In both cases, AASD calculates the sodium and ultrafiltration profile able to satisfy the constraints of the algorithm, respecting the input data, which are session length, weight loss, and sodium mass to be removed. Pre- and postsession body weight and body weight decrease were collected for each study period (1-month run-in period and 6 months of AASD): SBP, diastolic blood pressure (DBP), and heart rate presession, intrasession (hourly), and 1 hour after the end of the session. Also, the incidence of disequilibrium syndrome symptoms (hypotension, headache, nausea, vomiting, and cramps) was recorded as a percentage of symptomatic sessions (sessions with at least 1 event).
Outcomes The primary outcome of this study was the efficacy of AASD on intradialytic blood pressure. Secondary outcomes were the efficacy of AASD on other symptoms, including hypotensive events, headache, nausea, vomiting, and cramps. Am J Kidney Dis. 2011;58(1):93-100
Automatic Adaptive System Dialysis for Hypotension The following parameters were monitored: (1) blood pressure and heart rate pre- (at the beginning), intra- (hourly), and post session (1 hour after); and (2) disequilibrium symptoms (hypotensive events, headache, nausea, vomiting, and cramps), expressed as a percentage of symptomatic sessions. At the same time, patients’ sodium/water balance was assessed monthly, monitoring dry body weight, mean interdialytic weight gain, and mean presession blood sodium level.
Statistical Analysis All data are expressed as mean ⫾ standard error. Differences between baseline treatment and the first and second 3-month periods of AASD treatment were tested using analysis of variance test at each time of dialysis session. Longitudinal mixed models with an unstructured covariance matrix were used to analyze repeated measures of SBP, DBP, and heart rate. The advantage of linear mixed models is that they permit examination of variations among patients while adjusting for correlation within individuals. Our analysis included random effects for intercepts and for the effect of time. This implies that change over time may vary between patients. Treatment was included in the analysis as a covariate along with the time ⫻ treatment interaction term. Similarly, a longitudinal mixed model was implemented for weight and sodium. Variability in percentage of intradialytic symptoms during the study period was assessed using repeated-measure analysis of variance followed by post hoc comparison with Tukey test. Significant differences were defined as P ⬍ 0.05. Data were analyzed using SPSS for Windows software package (version 9.0.1; SPSS, www.spss.com) and SAS software (SAS Institute Inc, www.sas.com). Informed consent and the approval of the study protocol by the ethical committee of each center were collected. The EudraCT registry number for the protocol is 106/2004/U/Oss.
RESULTS An Italian noncontrolled multicenter (15 dialysis units) prospective clinical trial was carried out from September 2007 to September 2008 to confirm the clinical efficacy of AASD in the long term (6 months) in 55 patients during 2,376 dialysis sessions. Patients were all older than 18 years and at baseline had been undergoing hemodialysis 3 times a week for at least 6 months. Patient and dialysis session characteristics are listed in Table 1. Seven patients dropped out of the study: 1 patient died during the run-in period, and another, after 4 months of AASD treatment (from cardiac complications unrelated to the treatment itself); 2 patients dropped out for kidney transplant after 2 months of AASD; 1 patient stopped the treatment of their own volition after 1 month of AASD because of difficulty adapting to changes in dialysis shift/room; and 2 patients dropped out 3 months after AASD for vascular access problems. Systolic Blood Pressure Average predialysis SBPs were 119.6 ⫾ 3.8 mm Hg during the run-in phase, 126.6 ⫾ 1.5 mm Hg in the first 3 months of AASD (P ⫽ 0.08), and 130.5 ⫾ 1.5 mm Hg in the second 3 months of AASD (P ⫽ 0.005). Intradialytic SBP for the first 3 months of AASD was Am J Kidney Dis. 2011;58(1):93-100
Table 1. Patient and Dialysis Session Characteristics Characteristic
Value
No. of patients Men/women Age (y) Dialysis vintage (mo) Mean residual glomerular filtration (mL/min) Origin of kidney disease Diabetes Chronic glomerulonephritis Vascular nephropathy Chronic interstitial nephritis ESKD
55 24:31 61 ⫾ 11 53 ⫾ 18 ⬍3
No. of sessions studied No. of standard treatment sessions (run-in phase) No. of AASD sessions
3,018 642
No. of patients in each treatment group at baseline Standard bicarbonate dialysis (low-flux synthetic membrane) Hemodiafiltration (high-flux synthetic membrane)
18 (33) 5 (9) 14 (25) 3 (5) 15 (27)
2,376
37 (67) 18 (33)
No. of patients in each session length category 3.5 h 4h 4.5 h
11 (20) 39 (70.9) 5 (9.1)
No. of patients in each sodium mass removal method group Set a prioria By natremia targetb
47 (85.5) 8 (14.5)
Note: Unless otherwise indicated, values shown are number, mean ⫾ standard error, or number (percentage). Abbreviations: AASD, automatic adaptive system dialysis; ESKD, end-stage kidney disease. a Subsequent profile elaboration by AASD in the range of 80-100 mmol/kg of ultrafiltration goal, according to a clinical evaluation. b Calculated by AASD as a function of end-session natremia.
significantly higher than in the run-in phase at 1 hour of treatment (119.3 ⫾ 1.5 vs 109.8 ⫾ 3.6 mm Hg; P ⫽ 0.02) and at 2 hours (115.3 ⫾ 1.6 vs 104.1 ⫾ 3.7 mm Hg; P ⫽ 0.008). SBP in the second 3 months was higher than during the run-in phase at 1 (124.3 ⫾ 1.7 vs 113.4 ⫾ 3.9 mm Hg; P ⫽ 0.01), 2 (121.0 ⫾ 1.4 vs 109.8 ⫾ 3.6 mm Hg; P ⫽ 0.004), and 3 hours of treatment (118.8 ⫾ 1.6 vs 104.1 ⫾ 3.7 mm Hg; P ⬍ 0.001). Compared with the run-in phase, postdialytic SBP was higher for both the first and second 3-month AASD treatment periods: 122.0 ⫾ 1.8 versus 109.9 ⫾ 3.7 mm Hg (P ⫽ 0.006) and 124.8 ⫾ 1.7 versus 109.9 ⫾ 3.7 mm Hg (P ⬍ 0.001), respectively. Patients started on average with an SBP of 124.11 mm Hg, which changed by ⫺3.06 mm Hg (P ⬍ 0.001) each hour. Compared with the run-in phase, 95
Colì et al
Figure 1. Comparison of (A) systolic blood pressure (SBP), (B) diastolic blood pressure (DBP), and (C) heart rate in the 2 automatic adaptive system dialysis (AASD) treatment periods (first and second 3 months) versus the run-in phase (standard treatment with dialysis or hemodiafilration). Times shown are predialytic; 1-, 2-, 3-, and 4-hour times of dialysis; and postdialytic. (A) *P ⫽ 0.01 for second 3-month period of AASD versus the run-in phase. (B) *P ⫽ 0.04. (C) **P ⬍ 0.001 for the first and second 3-month periods versus the run-in phase (longitudinal mixed models).
in the first and second 3-month periods of AASD treatment, patients had an increase in SBP over time of 2.7 ⫾ 2.1 (P ⫽ 0.2) and 5.4 ⫾ 2.1 mm Hg (P ⫽ 0.01), respectively (Fig 1A). Diastolic Blood Pressure Predialysis DBP values did not show significant differences between the run-in phase and the first and second 3-month periods of AASD treatment. In the first 3 months of AASD, intradialytic DBP was higher than during the run-in phase at 2 hours (68.1 ⫾ 0.9 vs 63.2 ⫾ 96
2.0 mm Hg; P ⫽ 0.04). In the second 3 months of AASD, intradialytic DBP was higher than during the run-in phase at 2 (69.4 ⫾ 0.8 vs 64.3 ⫾ 1.9 mm Hg; P ⫽ 0.03), 3 (69.4 ⫾ 0.8 vs 63.2 ⫾ 2.0 mm Hg; P ⫽ 0.006), and 4 hours (69.4 ⫾ 1.0 vs 62.9 ⫾ 2.5 mm Hg; P ⫽ 0.01). Postdialysis DBPs were significantly different between AASD and the run-in phase during the second 3 months (72.1 ⫾ 0.7 vs 66.6 ⫾ 2.0 mm Hg; P ⫽ 0.01). Patients started on average with a DBP of 68.29 mm Hg, which changed by ⫺0.98 mm Hg (P ⫽ 0.001) each hour. In the first and second 3-month Am J Kidney Dis. 2011;58(1):93-100
Automatic Adaptive System Dialysis for Hypotension
Figure 2. Symptoms of disequilibrium syndrome in each month of automatic adaptive system dialysis (AASD) treatment versus the run-in phase (standard treatment).
periods of AASD treatment, patients had a slight increase in DBP over time compared with the run-in phase, of 1.05 ⫾ 1.02 (P ⫽ 0.3) and 2.05 ⫾ 1.03 mm Hg (P ⫽ 0.04), respectively (Fig 1B). Heart Rate Intradialytic heart rate values were lower in the first 3 months with AASD than in the run-in phase, a difference that was statistically significant at the 2-hour time (73.9 ⫾ 0.8 vs 78.4 ⫾ 1.5 beats/min; P ⫽ 0.03, ANOVA). In the second 3 months of AASD, significant differences at the 1-, 2-, 3-, and 4-hour times were noted. Postdialysis heart rate values were significantly lower in the second 3 months (77.3 ⫾ 1.2 vs 72.6 ⫾ 0.7 beats/min; P ⫽ 0.01). By means of longitudinal mixed models, patients started on average with a heart rate of 75.2 beats/min, which changed by 0.6 beat/min (P ⫽ 0.02) each hour. In the first and second 3-month periods of AASD treatment, changes in patients’ heart rates over time were statistically significantly different compared with the run-in phase, with changes of ⫺2.8 ⫾ 0.9 (P ⬍ 0.001) and ⫺4.0 ⫾ 0.9 beats/min (P ⬍ 0.001), respectively (Fig 1C). Intradialytic Hypotension Events and Symptoms The frequency of each symptom tracked for the study decreased in a statistically significant fashion during the AASD treatment period. The percentage of sessions affected by hypotension events decreased from 58.7% ⫾ 7.3% to 32.7% ⫾ 6.1% in the first month of AASD (P ⫽ 0.008) and to 0.9% ⫾ 0.6% in the sixth month (P ⬍ 0.001), with Am J Kidney Dis. 2011;58(1):93-100
significant differences throughout the AASD period. The incidence of headaches decreased from 32.6% ⫾ 8.1% of sessions to 18.7% ⫾ 5.0% in the first month of AASD, although this change was not statistically significant (P ⫽ 0.4). However, the percentage of sessions with headaches was significantly lower in the second (10.0% ⫾ 3.1%; P ⫽ 0.01), third (6.6% ⫾ 2.7%; P ⬍ 0.001), fourth (1.4% ⫾ 1.0%; P ⬍ 0.001), and fifth months (0.4% ⫾ 0.4%; P ⬍ 0.001) of AASD compared with the run-in phase, and no headaches were reported in the sixth month. The percentage of sessions with nausea decreased from 26.8% ⫾ 6.7% to 11.2% ⫾ 4.1% in the first month of AASD (P ⫽ 0.07) to 6.6% ⫾ 3.2% in the second month (P ⫽ 0.005), to 3.3% ⫾ 1.8% in the third month (P ⬍ 0.001), and to 0.5% ⫾ 0.5% in the fourth month (P ⬍ 0.001); no instances of nausea were reported in the fifth and sixth months. Vomiting went from an incidence of 13.1% ⫾ 4.3% in the run-in phase to 5.9% ⫾ 2.4% in the first month of AASD, but this change was not statistically significant (P ⫽ 0.3). However, the percentage of sessions with vomiting was 2.3% ⫾ 1.1% in the second month (P ⫽ 0.01 compared with the run-in phase), 0.5% ⫾ 0.5% in the third and fourth months (P ⬍ 0.001), and none in the fifth month. The percentage of sessions with cramps decreased to 3.5% ⫾ 1.3% in the first month of AASD (P ⫽ 0.002) to 1.7% ⫾ 1.0% in the second month (P ⬍ 0.001), to 0.5% ⫾ 0.5% in the third month (P ⬍ 0.001), to 0.8% ⫾ 0.5% in the fourth month (P ⬍ 0.001), and to 0.4% ⫾ 0.4% in the fifth and sixth months (P ⬍ 0.001; Fig 2). 97
Colì et al
Sodium Balance and Body Weight Assessment Results from the mixed model showed no significant improvement in sodium over time for patients with AASD treatment. The average value of sodium removed was the same during the run-in phase and the AASD treatment period at 378 ⫾ 95 and 367 ⫾ 84 mmol, respectively (P ⫽ 0.08). In patients treated according to natremia target in the AASD phase, the average amount of sodium removed was 384 ⫾ 91 mmol, which did not differ significantly from the value in the run-in phase (P ⫽ 0.09). Average predialysis sodium level for the run-in phase was 137 ⫾ 0.97 mEq/L, whereas in the AASD phase, levels were 137.54 ⫾ 0.49 mEq/L (P ⫽ 0.08) in the third month and 136.77 ⫾ 0.81 mEq/L in the sixth month (P ⫽ 0.08). There were no significant differences in end-ofsession body weights, which were 68.32 ⫾ 1.4 kg in the run-in phase, 68.49 ⫾ 1.4 kg in the first month of AASD, 68.45 ⫾ 1.4 kg in the second month, 68.31 ⫾ 1.4 kg in the third month, 68.7 ⫾ 1.5 kg in the fourth month, 68.4 ⫾ 1.5 kg in the fifth month, and 67.7 ⫾ 1.6 kg in the sixth month. Mean interdialytic weight gain did not vary significantly, being 3.23 ⫾ 0.1 kg in the run-in phase, 3.16 ⫾ 0.1 kg in the first month of AASD, 3.12 ⫾ 0.1 kg in the second month, 3.18 ⫾ 0.1 kg in the third month, 3.18 ⫾ 0.1 kg in the fourth month, 3.07 ⫾ 0.1 kg in the fifth month, and 2.96 ⫾ 0.1 in the sixth month.
DISCUSSION The pathogenesis of intradialytic complications is multifactorial. The main factor is plasma water loss through ultrafiltration, with a decrease in extracellular space and blood volume. The decrease in intradialytic osmolarity causes an additional loss of plasma water due to the shift from the extracellular to the intracellular space.11 The extracellular space decrease induces hypotension, whereas the intracellular space increase induces disequilibrium symptoms.3,12,13 Many investigators agree that preservation of blood volume has a pivotal role in intradialytic hemodynamic stability.3,9,10,39-44 Because sodium is the major determinant of extracellular space osmolarity, dialysate sodium concentration can be manipulated to decrease symptoms during dialysis.45 More recently, this assumption led to the introduction of dialysate sodium profiling. Our technical approach to profiled dialysis is based on the clinical application of an automatic computer-based program, which in turn is based on a new mathematical kinetic model for processing intradialytic sodium and ultrafiltration profiles. The aim of this strategy is to preserve blood volume during dialysis by boosting refilling and decreasing the shift from the extracellu98
lar to the intracellular space. AASD allows for the elaboration of profiles at the beginning of each dialysis session while simultaneously taking into account the patient’s sodium balance at each dialysis session. Results of this study show that AASD significantly improves intradialytic blood pressure, decreases the incidence of hypotensive events and disequilibrium symptoms compared with basal treatment, and does not induce symptoms of sodium/water overload. With AASD, the intradialytic curves of SBP and DBP were more stable from the first 3-month period of treatment, with quicker postsession recovery to baseline values for SBP. AASD also decreased the incidence of intradialytic hypotensive events. The stabilization of intradialytic blood pressure observed in the study confirms the efficacy of AASD in enhancing compensation of blood volume decrease. This water drainage is exploited by the simultaneous ultrafiltration profile, which also reaches its peak between the first 40 and 60 minutes of the session, with a lower ultrafiltration rate in the second half of the session. Heart rate, an index of sympathetic system response, was more stable with AASD compared with the standard treatment during the run-in phase. Heart rate increases during hemodialysis, and Redaelli et al46 showed that it reached the peak value 1 hour after the end of the session, recovering the baseline value after 8 hours. Krepel et al47 showed significant correlation between relative blood volume and heart rate during a dialysis session, suggesting that a decrease in relative blood volume stimulates the autonomic nervous system to minimize the decrease in blood pressure by increasing heart rate. There is still much discordance about the correlation between heart rate and blood volume decrease. Several investigators have noted that intradialytic hypotension–prone patients more frequently are older and have diabetes and may have a minor adaptive response of heart rate due to impaired autonomic control.43,48,49 Of note, from the first 3-month period of AASD, postsession heart rate values were not significantly different from presession values. This result is related to lower intra- and postdialytic sympathetic activity, confirming that there is less need for compensation. Several studies have shown that use of profiled dialysate sodium can decrease the incidence of disequilibrium-symptomatic dialysis sessions by decreasing the magnitude of the intradialytic decrease in plasma osmolarity and thereby reducing the shift from the extracellular to the intracellular space.50-52 Our study showed that the percentage of symptomatic sessions was decreased significantly from the first month with AASD in terms of muscle cramps and from the second month in terms of nausea, vomiting, and headache. Am J Kidney Dis. 2011;58(1):93-100
Automatic Adaptive System Dialysis for Hypotension
When processing profiles, AASD takes into account the quantity of sodium to be removed, which is an input datum for AASD at the beginning of each session. This sodium balance of the AASD session remained the same as in the run-in phase using standard treatment, and no signs of sodium/water overload appeared in patients during the AASD treatment period: dry body weight remained stable, as did interdialytic body weight gain and natremia levels at the start of each session. Our study has several limitations. It lacks blinding and is not a crossover randomized study. Patients enrolled in the study were clinically monitored during the run-in and AASD treatment periods, each patient serving as his or her own control. However, the study design lacks any carryover effects of AASD on the subsequent basal treatment. In conclusion, our multicenter study confirmed the long-term clinical efficacy of AASD on dialysis intolerance in a large number of patients. In particular, AASD maintained stable intradialytic blood pressure and decreased the appearance of intradialytic disequilibrium symptoms, taking into account the patient’s sodium balance. Furthermore, the technique’s automatic determination of profiles by the AASD program proved both reliable and easy to implement in clinical practice in all 15 dialysis units participating in the study.
ACKNOWLEDGEMENTS Support: The study was supported in part by the Fondazione Cassa di Risparmio in Bologna, project no. 2007/0234 “Innovazioni diagnostiche e terapeutiche nell’approccio alle criticità immunologiche del paziente nel percorso Dialisi-Trapianto di Rene” (Principal Investigator, Prof Stefoni). Financial Disclosure: The authors declare that they have no relevant financial interests.
REFERENCES 1. de Jager DJ, Grootendorst DC, Jager KJ, et al. Cardiovascular and noncardiovascular mortality among patients starting dialysis. JAMA. 2009;302(16):1782-1789. 2. Collins AJ, Foley RN, Gilbertson DT, Chen SC. The state of chronic kidney disease, ESRD, and morbidity and mortality in the first year of dialysis. Clin J Am Soc Nephrol. 2009;4(suppl 1):S511. 3. Daugirdas JT. Pathophysiology of dialysis hypotension: an update. Am J Kidney Dis. 2001;38(4)(suppl 4):S11-17. 4. Donauer J. Hemodialysis-induced hypotension: impact of technologic advances. Semin Dial. 2004;17(5):333-335. 5. Sulowicz W, Radziszewski A. Dialysis induced hypotension—a serious clinical problem in renal replacement therapy. Med Pregl. 2007;60(2):14-20. 6. Germin Petrovic D. Effect of automated blood volume control on the incidence of intra-dialysis hypotension. Acta Med Croatica. 2003;57(1):17-22. 7. Stegmayr BG. Ultrafiltration and dry weight—what are the cardiovascular effects? Artif Organs. 2003;27(3):227-229. 8. Daugirdas JT, Blake PG, Ing TS. Handbook of Dialysis. Philadelphia, PA: Lippincott Williams & Wilkins; 2007. Am J Kidney Dis. 2011;58(1):93-100
9. Henrich WL. Hemodynamic stability during dialysis and cardiovascular disease in end-stage renal disease patients. Am J Kidney Dis. 1999;33(6):XLIX-LII. 10. Daugirdas JT. Dialysis hypotension: a hemodynamic analysis. Kidney Int. 1991;39(2):233-246. 11. Ebel H, Laage C, Keuchel M, et al. Impact of profile haemodialysis on intra-/extracellular fluid shifts and the release of vasoactive hormones in elderly patients on regular dialysis treatment. Nephron. 1997;75(3):264-271. 12. Trinh-Trang-Tan MM, Cartron JP, Bankir L. Molecular basis for the dialysis disequilibrium syndrome: altered aquaporin and urea transporter expression in the brain. Nephrol Dial Transplant. 2005;20(9):1984-1988. 13. Chen CL, Lai PH, Chou KJ, et al. A preliminary report of brain edema in patients with uremia at first hemodialysis: evaluation by diffusion-weighted MR imaging. AJNR Am J Neuroradiol. 2007;28(1):68-71. 14. Guyton AC. Textbook of Medical Physiology. Philadelphia, PA: W. B. Saunders Co; 1991. 15. Di Filippo S, Manzoni C, Andrulli S, et al. Sodium removal during pre-dilution haemofiltration. Nephrol Dial Transplant. 2003; 18(suppl 7):vii31-36; discussion vii57-58. 16. Rabindranath KS, Strippoli GF, Daly C, et al. Haemodiafiltration, haemofiltration and haemodialysis for end-stage kidney disease. Cochrane Database Syst Rev. 2006;4:CD006258. 17. Tang HL, Wong SH, Chu KH, et al. Sodium ramping reduces hypotension and symptoms during haemodialysis. Hong Kong Med J. 2006;12(1):10-14. 18. Raja RM. Sodium profiling in elderly haemodialysis patients. Nephrol Dial Transplant. 1996;11(suppl 8):42-45. 19. Straver B, De Vries PM, Donker AJ, et al. The effect of profiled hemodialysis on intra-dialytic hemodynamics when a proper sodium balance is applied. Blood Purif. 2002;20(4):364369. 20. Song JH, Park GH, Lee SY, et al. Effect of sodium balance and the combination of ultrafiltration profile during sodium profiling hemodialysis on the maintenance of the quality of dialysis and sodium and fluid balances. J Am Soc Nephrol. 2005;16(1):237246. 21. Agarwal R, Kelley K, Light RP. Diagnostic utility of blood volume monitoring in haemodialysis patients. Am J Kidney Dis. 2008;51:242-254. 22. Movilli E, Camerini C, Viola BF, et al. Blood volume changes during three different profiles of dialysate sodium variation with similar intra-dialytic sodium balances in chronic hemodialyzed patients. Am J Kidney Dis. 1997;30(1):58-63. 23. Zhou YL, Liu HL, Duan XF, et al. Impact of sodium and ultrafiltration profiling on haemodialysis-related hypotension. Nephrol Dial Transplant. 2006;21(11):3231-3237. 24. Kooman JP, van der Sande F, Leunissen K, et al. Sodium balance in hemodialysis therapy. Semin Dial. 2003;16(5):351-355. 25. Locatelli F, Covic A, Chazot C, et al. Optimal composition of the dialysate, with emphasis on its influence on blood pressure. Nephrol Dial Transplant. 2004;19(4):785-796. 26. Moret K, Aalten J, van den Wall Bake W, et al. The effect of sodium profiling and feedback technologies on plasma conductivity and ionic mass balance: a study in hypotension-prone dialysis patients. Nephrol Dial Transplant. 2006;21(1):138-144. 27. Oliver MJ, Edwards LJ, Churchill DN. Impact of sodium and ultrafiltration profiling on hemodialysis-related symptoms. J Am Soc Nephrol. 2001;12(1):151-156. 28. Davenport A. Can advances in hemodialysis machine technology prevent intradialytic hypotension? Semin Dial. 2009;22(3): 231-236. 29. Ursino M, Colì L, La Manna G, et al. A simple mathematical model of intradialytic sodium kinetics: “in vivo” validation 99
Colì et al during hemodialysis with constant or variable sodium. Int J Artif Organs. 1996;19(7):393-403. 30. Colì L, Ursino M, Dalmastri V, et al. A simple mathematical model applied to selection of the sodium profile during profiled haemodialysis. Nephrol Dial Transplant. 1998;13(2):404-416. 31. Colì L, Ursino M, De Pascalis A, et al. Evaluation of intradialytic solute and fluid kinetics. Setting up a predictive mathematical model. Blood Purif. 2000;18(1):37-49. 32. Ursino M, Colì L, Brighenti C. Prediction of solute kinetics, acid base status, and blood volume changes during profiled hemodialysis. Ann Biomed Eng. 2000;28:204-216. 33. Colì L, Ursino M, Donati G, et al. Clinical application of sodium profiling in the treatment of intradialytic hypotension. Int J Artif Organs. 2003;26:715-722. 34. Santoro A, Mancini E, Basile C, et al. Blood volume controlled hemodialysis in hypotension-prone patients: a randomized, multicenter controlled trial. Kidney Int. 2002;62(3):1034-1045. 35. Ursino M, Colì L, Dalmastri V, et al. An algorithm for the rational choice of sodium profile during hemodialysis. Int J Artif Organs. 1997;20(12):659-672. 36. Colì L, Bonomini M, La Manna G, et al. Clinical use of profiled hemodialysis. Artif Organs. 1998;22(9):724-730. 37. Ursino M, Colì L, Magosso E, et al. A mathematical model for the prediction of solute kinetics, osmolarity and fluid volume changes during hemodiafiltration with on-line regeneration of ultrafiltrate (HFR). Int J Artif Organs. 2006;29(11):10311041. 38. Charra B. Fluid balance, dry weight, and blood pressure in dialysis. Hemodial Int. 2007;11:21-31. 39. Beige J, Sone J, Sharma AM, et al. Computational analysis of blood volume curves and risk of intradialytic morbid events in hemodialysis. Kidney Int. 2000;58(4):1805-1809. 40. Locatelli F, Buoncristiani U, Canaud B, et al. Haemodialysis with on-line monitoring equipment: tools or toys? Nephrol Dial Transplant. 2005;20(1):22-33. 41. Dasselaar JJ, Huisman RM, de Jong PE, et al. Measurement of relative blood volume changes during haemodialysis:
100
merits and limitations. Nephrol Dial Transplant. 2005;20(10): 2043-2049. 42. Van der Sande FM, Kooman JP, Leunissen KM. Intradialytic hypotension—new concepts on an old problem. Nephrol Dial Transplant. 2000;15(11):1746-1748. 43. Andrulli S, Colzani S, Mascia F, et al. The role of blood volume reduction in the genesis of intradialytic hypotension. Am J Kidney Dis. 2002;40(6):1244-1254. 44. Van Kuijk WH, Wirtz JJ, Grave W, et al. Vascular reactivity during combined ultrafiltration-haemodialysis: influence of dialysate sodium. Nephrol Dial Transplant. 1996;11(2):323-328. 45. Mann H, Stiller S. Sodium modeling. Kidney Int Suppl. 2000;76:S79-88. 46. Redaelli B, Locatelli F, Limido D, et al. Effect of a new model of hemodialysis potassium removal on the control of ventricular arrhythmias. Kidney Int. 1996;50(2):609-617. 47. Krepel HP, Nette RW, Akçahüseyin E, et al. Variability of relative blood volume during haemodialysis. Nephrol Dial Transplant. 2000;15(5):673-679. 48. Lin YF, Wang JY, Shum AY, et al. Role of plasma catecholamines, autonomic, and left ventricular function in normotensive and hypotension prone dialysis patients. ASAIO J. 1993;39(4): 946-953. 49. Jassal SV, Douglas JF, Stout RW. Prevalence of central autonomic neuropathy in elderly dialysis patients. Nephrol Dial Transplant. 1998;13(7):1702-1708. 50. De Vries PM, Olthof CG, Solf A, et al. Fluid balance during haemodialysis and haemofiltration: the effect of dialysate sodium and a variable ultrafiltration rate. Nephrol Dial Transplant. 1991; 6(4):257-263. 51. Swartz RD, Somermeyer MG, Hsu CH. Preservation of plasma volume during hemodialysis depends on dialysate osmolality. Am J Nephrol. 1982;2(4):189-194. 52. Petitclerc T, Jacobs C. Dialysis sodium concentration: what is optimal and can it be individualized? Nephrol Dial Transplant. 1995;10(5):596-599.
Am J Kidney Dis. 2011;58(1):93-100