- Email: [email protected]
Equal levels of blood pressure control in ESRD patients receiving high- efficiency hemodialysis and conventional hemodialysis
Equal Levels of Blood Pressure Control in ESRD Patients Receiving High-Efficiency Hemodialysis and Conventional Hemodialysis Manuel T. Velasquez, MD, Beat von Albertini, MD, Susie Q. Lew, MD, Gary J. Mishkin, MS, and Juan P. Bosch, MD ● The present study compared the status of hypertension and adequacy of blood pressure control in 73 end-stage renal disease (ESRD) patients treated with four different modalities of hemodialysis, namely, conventional hemodialysis (CHD) with cuprophan 1.1 m2 at a blood flow rate of 300 mL/min, high-efficiency hemodialysis (HED) with cuprophan 1.6 m2 at a blood flow rate of 450 to 500 mL/min, high-flux hemodialysis (HFD) with F80 polysulfone 1.8 m2 at a blood flow rate 500 mL/min, and high-flux hemodiafiltration (HDF) with F80 2 ⴛ 1.8 m2 in series at a blood flow rate of 600 to 650 mL/min. Thirty of the 73 patients (41%) were receiving one or more antihypertensive agents to control their hypertension. The percentage of patients taking antihypertensive medication was less in the groups treated with HED, HFD, and HDF compared with the CHD group: 38%, 39%, and 39%, respectively, in the HED, HFD, and HDF groups versus 56% in the CHD group. Control of systolic and diastolic hypertension was achieved in a higher percentage of patients treated with HED, HFD, and HDF compared with patients treated with CHD. Sixty-two percent of HED, 58% of HFD, and 61% of HDF patients compared with 44% of CHD patients had systolic blood pressure less than 150 mm Hg, whereas 77 % of HED, 76% of HFD, and 78% of HDF patients compared 56% of CHD patients had diastolic blood pressure less than 90 mm Hg. However, the differences in the use of antihypertensive medication and control rates of hypertension did not reach statistical significance. The average blood pressure of all patients was 144/89 mm Hg; this did not differ significantly between the four groups. There also were no significant differences in etiology of ESRD, hematocrit, biochemical data, as well as use and dose of recombinant human erythropoietin between the four groups. Compared with the CHD patients, the average treatment times with high-efficiency treatments were shorter, with HDF patients showing the shortest mean treatment time of 157 ⴞ 41 minutes per hemodialysis session. The mean Kt/V was higher in the groups treated with HED, HFD, or HDF (1.31 ⴞ 0.3, 1.30 ⴞ 0.4, and 1.43 ⴞ 0.3, respectively) than in the CHD group (1.12 ⴞ 0.3; P F 0.05). Interdialytic weight gain also did not differ among the four groups. There was no correlation between predialysis mean arterial pressure and either treatment time (r ⴝ 0.04, P ⴝ NS), Kt/V (r ⴝ 0.03, P ⴝ NS), ultrafiltration rate (r ⴝ 0.06, P ⴝ NS), or interdialytic weight gain (r ⴝ ⴚ0.08, P ⴝ NS). There also was no significant association between Kt/V and use of antihypertensive medications (chi-square ⴝ 1.76, P ⴝ NS). There was, however, a significant positive correlation between interdialytic weight gain and treatment time (r ⴝ 0.33, P F 0.01). We conclude that the use of short dialysis sessions with efficient hemodialysis treatments, namely, HFD and HDF, was associated with similar levels of blood pressure control in ESRD patients. r 1998 by the National Kidney Foundation, Inc. INDEX WORDS: Blood pressure control; high-efficiency hemodialysis; high-flux hemodialysis; high-flux hemodiafiltration; hypertension; interdialytic weight gain; treatment time.
I
N RECENT YEARS, the introduction of highly efficient hemodialytic techniques that use large surface area membranes and high blood flow rates, namely, high-flux hemodialysis (HFD) and high-flux hemodiafiltration (HDF), has added significant improvements in dialysis delivery in patients with end-stage renal disease (ESRD).1-5
From the Division of Renal Diseases and Hypertension, Department of Medicine, George Washington University Medical Center, Washington, DC. Received April 21, 1997; accepted in revised form October 10, 1997. Address reprint requests to Manuel T. Velasquez, MD, Division of Renal Diseases and Hypertension, Department of Medicine, George Washington University Medical Center, 2150 Pennsylvania Ave, NW, Room 4-426, Washington, DC 20037.
r 1998 by the National Kidney Foundation, Inc. 0272-6386/98/3104-0007$3.00/0 618
These techniques combine conventional diffusive hemodialysis with the large convective component of hemofiltration, and improve both efficacy and efficiency of hemodialysis treatment by increasing the rate of solute removal, as well as the range or molecular size of solutes removed. The major advantage of these therapies is that they provide adequate dialysis in a shorter time. However, it has been suggested that the shorter treatment times may limit the rate of ultrafiltration (UF) and lead to inadequate removal of fluid during hemodialysis. Such failure to completely remove excess fluid is thought to account for the high incidence or persistence of hypertension in hemodialysis patients. This hypothesis is based on the observation by Charra et al6 in Tassin, France, that long hemodialysis treatment 8 hours three times a week is associated with wellcontrolled hypertension. However, studies are
American Journal of Kidney Diseases, Vol 31, No 4 (April), 1998: pp 618-623
HYPERTENSION CONTROL IN HEMODIALYSIS PATIENTS
lacking regarding the status of blood pressure control in hemodialysis patients treated with the newer high-flux dialyzers and shorter dialysis sessions. We therefore compared the status of hypertension and adequacy of blood pressure control in ESRD patients receiving four different modalities of hemodialysis treatments and sought to determine whether the shorter treatment times used in highly efficient dialysis techniques have an adverse effect on blood pressure control in these patients. PATIENTS AND METHODS Seventy-three adult patients with ESRD who were receiving chronic hemodialysis treatment at the George Washington University Ambulatory Dialysis Center were studied. All patients had been on maintenance hemodialysis for at least 3 months before the study. Patients were divided into four groups according to the modality of hemodialysis treatment they received: conventional hemodialysis (CHD) with the use of cuprophan membranes having a 1.1 m2 surface area at a blood flow rate of 300 mL/min, high-efficiency hemodialysis (HED) with cuprophan ⫻ 1.6 m2 surface area at a blood flow rate of 450 to 500 mL/min, HFD with a F80 polysulfone membrane ⫻ 1.8 m2 surface area (Fresenius, Walnut Creek, CA) at a blood flow rate of 500 mL/min, and HDF with two F80 dialyzers 2 ⫻ 1.8 m2 surface area in series at a blood flow rate of 600 to 650 mL/min. All hemodialysis treatments were conducted three times a week using bicarbonate dialysate with constant sodium concentration. For both HFD and HDF, dialyzers were reused up to 12 times after reprocessing with 4% formaldehyde. During hemodialysis, UF was performed in each patient to achieve dry weight. Dry weight was defined as the weight below which further fluid removal during dialysis results in hypotension and/or symptoms of dizziness or muscle cramps and above which signs and symptoms of fluid expansion, such as rales in the chest and pedal edema, are evident. Blood samples for routine hematologic and biochemical evaluations were obtained monthly. Monthly predialysis and postdialysis blood samples for urea determination were also obtained at the beginning of the week. Adequacy of dialysis was determined monthly by urea kinetic analysis using a posttreatment blood sample for urea obtained at 10 minutes after the end of treatment. The formula of Daugirdas7 was used to calculate Kt/V urea. In all patients, demographic data were collected that included age, gender, race, primary ESRD diagnosis, body weight, blood pressure, hematocrit, biochemical data, and current use of antihypertensive agents and recombinant human erythropoietin (rHuEPO). Measurements of predialysis and postdialysis body weight and blood pressure were routinely obtained by the dialysis technician or nurse at each dialysis session. Interdialytic weight gain was calculated as the amount of weight gained between the end of a preceding dialysis session and the start of the next dialysis session. Predialysis systolic blood pres-
619
sure (SBP) and diastolic blood pressure (DBP) were measured with a standard mercury sphygmomanometer and taken with the patient in the sitting position. Mean arterial pressure (MAP) was calculated as the sum of DBP and one third of the difference in SBP and DBP or pulse pressure. Patients were considered to be hypertensive if they were receiving antihypertensive medications during the study period to control their hypertension. Any patient with an SBP less than 150 mm Hg and a DBP less than 90 mm Hg (based on the weekly average of predialysis blood pressure measurements taken over a 1-month period) was considered to have adequately controlled hypertension.
Statistical Analysis Results are expressed as mean values ⫾ SD unless otherwise indicated. Comparisons among groups were performed using ANOVA. Chi-square analysis was used for comparison of nominal data. Simple regression analysis was used for calculation of the correlation coefficients. Statistical differences were considered significant at P ⬍ 0.05.
RESULTS
The demographic characteristics of the four groups of patients are summarized in Table 1. The patients treated with HED, HFD, and HDF tended to be older compared with the patients in the CHD group. Note that well over 75% of the patients in each treatment group were black. Hypertension and diabetes were the two most common causes of renal disease in all groups. Of the 73 hemodialysis patients, 30 (41%) were receiving one or more antihypertensive drugs to control their hypertension and 43 (59%) were not (Table 2). The percentage of patients receiving antihypertensive medication was less in the HED, HFD, and HDF groups compared with the CHD group: 38%, 39%, and 39%, respectively, versus 56%. However, the difference between the four dialytic groups did not reach statistical significance (P ⫽ NS, chi-square test). Table 2 also shows that the percentage of patients achieving an SBP less than 150 mm Hg and a DBP less than 90 mm Hg was higher in the groups receiving high-efficiency treatments compared with the CHD group, but this difference was not statistically significant (P ⫽ NS, chisquare test). Fifty-six (77%) of the 73 patients were also receiving rHuEPO therapy (Table 3). However, the proportion of patients on rHuEPO did not differ among the four groups. The dose of rHuEPO tended to be lower in the HED, HFD, and HDF groups compared with the CHD group, but the difference between groups was not statis-
620
VELASQUEZ ET AL Table 1. Demographic Characteristics of Study Patients CHD (%) HED (%) HFD (%) HDF (%)
No. of patients 9 13 33 18 Mean age ⫾ SE (yr) 40 ⫾ 16 55 ⫾ 18 55 ⫾ 12 48 ⫾ 14 Race Black 7 (78) 12 (92) 27 (82) 16 (89) White 1 (11) 1 (8) 6 (18) 2 (11) Asian 1 (11) 0 (0) 0 (0) 0 (0) Gender Male 7 (78) 6 (46) 15 (45) 16 (89) Female 2 (22) 7 (54) 18 (55) 2 (11) Cause of renal disease Hypertension 4 (44) 5 (39) 13 (40) 6 (33) Diabetes 1 (11) 3 (23) 9 (27) 3 (17) Chronic glomerulonephritis 1 (11) 0 (0) 5 (15) 3 (17) Chronic interstitial nephritis 0 (0) 2 (15) 1 (3) 1 (5) Adult polycystic kidney disease 0 (0) 1 (8) 1 (3) 2 (11) Other 3 (34) 2 (15) 4 (12) 3 (17)
tically significant. Similarly, there was no significant difference in mean hematocrit and biochemical data among the four groups. Table 4 summarizes the average treatment time per dialysis session, UF rate, interdialytic weight gain, adequacy of dialysis, and blood pressure in each of the four treatment groups. Compared with the CHD group, the average Table 2. Use of Antihypertensive Medication and Control of Blood Pressure in ESRD Patients Treated With Four Different Hemodialysis Techniques
No. (%) of patients on antihypertensive medication One drug Two drugs Three or more drugs Total No. (%) of patients with SBP ⬍150 mm Hg DBP ⬍90 mm Hg
CHD, n⫽9 (%)
HED, n ⫽ 13 (%)
HFD, n ⫽ 33 (%)
HDF, n ⫽ 18 (%)
2 (23) 3 (33)
3 (23) 2 (15)
8 (24) 4 (12)
3 (17) 4 (22)
0 (0) 5 (56)
0 (0) 5 (38)
1 (3) 13 (39)
0 (0) 7 (39)
4 (44) 5 (56)
8 (62) 10 (77)
19 (58) 25 (76)
11 (61) 14 (78)
NOTE. Differences between groups are not significant (chi-square test).
treatment times in the HED and HFD groups were shorter, with HDF showing the shortest mean treatment time of 157 ⫾ 41 minutes per session. There also was a trend for higher UF rates in HED and HFD treatments compared with CHD. The mean UF rate in the HDF group was 19.8 ⫾ 6.1 mL/min compared with a mean of 15.4 ⫾ 4.7 mL/min in the CHD group (P ⬍ 0.05). The mean Kt/V was significantly higher in the groups treated with HED, HFD, or HDF (1.31 ⫾ 0.3, 1.30 ⫾ 0.4, and 1.43 ⫾ 0.3, respectively) compared with the CHD group (1.12 ⫾ 0.3; P ⬍ 0.05). The average predialysis blood pressure of all patients was 144/89 mm Hg, which did not differ between the four groups. There also were no differences in magnitude of interdialytic weight gain and predialysis MAP among the four groups. The relationships between hemodialysis treatment time, UF, Kt/V, interdialytic weight gain, and blood pressure for all patients were also examined using correlation analysis. The results show no significant correlation between predialysis MAP and dialysis treatment time (r ⫽ 0.04, P ⫽ NS), as shown in Fig 1. There was no correlation between predialysis MAP and either Kt/V (r ⫽ 0.03, P ⫽ NS), UF (r ⫽ 0.06, P ⫽ NS), or interdialytic weight gain (r ⫽ ⫺0.08, P ⫽ NS). There also was no significant association between Kt/V and use of antihypertensive medications (chi-square ⫽ 1.76, P ⫽ NS). There was, however, a significant positive correlation between interdialytic weight gain and treatment time (r ⫽ 0.33, P ⬍ 0.01; Fig 2). DISCUSSION
To date, many dialysis centers commonly prescribe dialysis treatment of 4 hours three times a week to most patients undergoing maintenance hemodialysis. However, for patients receiving HED treatments, the length of dialysis time per session is usually reduced. Although these highefficiency treatments provide adequate dialysis, the issue has been raised that the relatively shorter treatment times achieved with these therapies may adversely affect control of blood pressure. It is postulated that reducing dialysis treatment time may limit removal of excess fluid during hemodialysis and lead to poor control of hypertension. On the other hand, increasing treatment time allows slow and complete fluid removal and
HYPERTENSION CONTROL IN HEMODIALYSIS PATIENTS
621
Table 3. Use and Dose of rHuEPO, Hematocrit, and Biochemical Data in ESRD Patients Treated With Four Different Hemodialysis Techniques
No. of patients on rHuEpo (%) Dose of rHuEPO (U/kg/wk) Hematocrit (%) Creatinine (mg/dL) Glucose (mg/dL) Cholesterol (mg/dL) Albumin (g/dL) Calcium (mg/dL) Phosphorus (mg/dL) Alkaline phosphatase
CHD (n ⫽ 9)
HED (n ⫽ 13)
HFD (n ⫽ 33)
HDF (n ⫽ 18)
7 (78) 80.2 ⫾ 40.0 28.5 ⫾ 4.0 15.0 ⫾ 4.1 93 ⫾ 33 158 ⫾ 29 4.0 ⫾ 0.6 8.9 ⫾ 0.8 6.5 ⫾ 1.9 144 ⫾ 89
10 (77) 74.3 ⫾ 32.0 30.2 ⫾ 6.0 12.1 ⫾ 4.2 93 ⫾ 29 164 ⫾ 34 3.7 ⫾ 0.9 9.1 ⫾ 0.5 5.9 ⫾ 2.1 136 ⫾ 73
25 (76) 73.1 ⫾ 39.0 31.3 ⫾ 3.0 13.8 ⫾ 3.6 100 ⫾ 55 164 ⫾ 27 3.9 ⫾ 0.3 9.4 ⫾ 1.3 6.3 ⫾ 2.4 104 ⫾ 49
14 (78) 61.6 ⫾ 33.0 31.6 ⫾ 3.0 15.5 ⫾ 3.9 93 ⫾ 30 152 ⫾ 29 4.1 ⫾ 0.3 9.3 ⫾ 1.2 5.6 ⫾ 1.9 95 ⫾ 45
NOTE. Data are given as mean values ⫾ SD. Differences between groups are not significant (ANOVA).
results in improved control of blood pressure. This is supported by the studies of Charra et al6 in Tassin, France, which showed that long hemodialysis treatment sessions with Kiil dialyzers (8 hours three times a week) is associated with good control of hypertension and discontinuation of antihypertensive drug therapy in approximately 98% of their dialysis patients. In the present study, we addressed this issue by comparing the status of blood pressure control in our ESRD patients receiving either CHD or high-efficiency treatments. We found that highefficiency treatments with shorter dialysis times, namely, HED, HFD, and HDF, were not associated with a higher incidence of hypertension (use of antihypertensive medication) compared with CHD. Furthermore, control of hypertension (defined as SBP ⬍150 mm Hg or DBP ⬍90 mm Hg) was not worse in patients receiving high-effi-
ciency treatments compared with patients treated with CHD. Indeed, the majority of the patients receiving high-efficiency treatments achieved adequate control of hypertension, with over 55% of patients having an SBP less than 150 mm Hg and over 75% showing a DBP less than 90 mm Hg. Thus, it appears that control of hypertension in our patients was not negatively impacted by high-efficiency treatments compared with CHD. Previous studies on the prevalence rates and control of hypertension in ESRD patients undergoing hemodialysis have yielded variable results. Ritz et al8 reported that 96% of diabetic patients and 93% of nondiabetic patients receiving chronic hemodialysis required antihypertensive medication to control their hypertension. In this study, only 13% of diabetic patients and 28% of nondiabetic patients had SBPs below 160 mm Hg. Cheigh et al9 found that approximately half
Table 4. Comparison of Dialysis Treatment Time, Body Weight, Interdialytic Weight Gain, Adequacy of Dialysis, and Blood Pressure in Hemodialysis Patients
Treatment time (min) Predialysis body weight (kg) Ultrafiltration rate (mL/min) Interdialytic weight gain (kg) Kt/V Predialysis blood pressure (mm Hg) SBP DBP MAP
CHD (n ⫽ 9)
HED (n ⫽ 13)
HFD (n ⫽ 33)
HDF (n ⫽ 18)
230 ⫾ 21 68.4 ⫾ 18.0 15.4 ⫾ 4.7 3.3 ⫾ 1.8 1.12 ⫾ 0.30
184 ⫾ 31* 68.8 ⫾ 16.0 17.0 ⫾ 6.2 3.3 ⫾ 1.1 1.31 ⫾ 0.30*
186 ⫾ 52* 73.1 ⫾ 39.0 18.2 ⫾ 5.9 3.6 ⫾ 1.7 1.30 ⫾ 0.40*
157 ⫾ 41* 71.6 ⫾ 33 19.8 ⫾ 6.1* 3.2 ⫾ 1.6 1.43 ⫾ 0.30*
146 ⫾ 16 89 ⫾ 14 104 ⫾ 12
139 ⫾ 21 86 ⫾ 15 96 ⫾ 12
148 ⫾ 15 92 ⫾ 14 101 ⫾ 15
148 ⫾ 14 88 ⫾ 15 102 ⫾ 28
NOTE. Data are given as mean values ⫾ SD. *Significantly different from CHD; P ⬍ 0.05.
622
Fig 1. Correlation between dialysis treatment time and predialysis MAP.
(44%) of the 244 chronic hemodialysis patients in their series were receiving antihypertensive medications. They also observed a high incidence of uncontrolled hypertension (defined as the percentage of ambulatory SBP ⬎150 mm Hg and DBP ⬎90 mm Hg) in this study, with only 15% of patients maintaining a blood pressure below 150/90 mm Hg at all times, despite antihypertensive drug therapy. Recently, Salem,10 in a survey of 649 patients receiving hemodialysis with cuprophan dialyzer 4 hours three times a week, showed a 72% prevalence rate of hypertension defined as MAP greater than 114 mm Hg or blood pressure greater than 160/90 mm Hg. Approximately 59% of the total patients were receiving antihypertensive medication. More recently, Dhakal et al11 showed that 67% of 96 patients on chronic hemodialysis had uncontrolled hyperten-
Fig 2. Correlation between dialysis treatment time and interdialytic weight gain.
VELASQUEZ ET AL
sion defined as SBP greater than 150 mm Hg or DBP greater than 90 mm Hg. The differences in the prevalence rates and control of hypertension between these reports and our study are difficult to interpret, but they may relate to differences in demographics, the type and amount of hemodialysis treatment, and volume status of the study patients. In the present study, the majority of our patients were blacks in whom hypertension and diabetes were identified as the two most frequent probable causes of ESRD. However, there were no significant differences in either race or etiology of renal disease among the four groups of patients. The majority of our patients were also receiving rHuEPO treatment for maintenance treatment of anemia. Since an increase in blood pressure is a well-documented side effect of rHuEPO therapy,12-14 this could have added to sustained hypertension and increased need for antihypertensive drug therapy in some of these patients. However, we did not find significant differences in the use of rHuEPO among the four groups of patients. Furthermore, the average weekly doses of rHuEPO used and mean hematocrit values did not differ between the four groups. Therefore, it seems unlikely that the use of rHuEPO and the degree (or correction) of anemia had a different influence on the magnitude of blood pressure control among the four dialytic groups. Salt and water retention is considered the dominant factor contributing to hypertension in ESRD patients. In hemodialysis, the persistence of hypertension in some patients may relate to volume overload that occurs during the interdialytic period. In the present study, interdialytic weight gain, which probably reflects increased fluid volume, did not differ among the four groups of patients. Moreover, there was no clear relationship between predialysis MAP and interdialytic weight gain, a finding that has been observed by other groups in hemodialysis patients.10,15,16 Therefore, differences in blood pressure control between high-efficiency treatments and CHD cannot be attributed to differences in interdialytic weight gain. It is interesting that we observed a direct correlation between treatment time and interdialytic weight gain. The significance of this finding is unclear. Even though treatment time was reduced in patients treated
HYPERTENSION CONTROL IN HEMODIALYSIS PATIENTS
with HED, HFD, and HDF, the amount of dialysis delivered with these therapies was adequate and more than that achieved with CHD, as indicated by the higher Kt/V urea of ⱖ1.3, in HED, HFD, and HDF patients compared with patients treated with CHD. It is possible that highefficiency treatments provide better clearance of vasoconstrictor substances or middle molecules potentially capable of sustaining hypertension in ESRD patients.17 Additional studies are needed to examine this point. Our results do not necessarily contradict the view that long hemodialysis treatment can provide better control of blood pressure because we did not evaluate the effects of increasing treatment time with high-efficiency techniques in the same group of patients. However, Luik et al18 prospectively studied the effect of increasing dialysis treatment time on blood pressure control in a small group of patients. These investigators found that increasing treatment time by 2 hours without a change in dry weight resulted in a decrease in the use of antihypertensive medication as well as a reduction in interdialytic SBP. Further studies involving a larger number of hemodialysis patients are needed to confirm these findings. On the other hand, other prospective studies are also needed to evaluate the effects on blood pressure control of reducing treatment time with high-efficiency treatments but keeping all other variables (eg, type of dialyzer, dry weight, and Kt/V) constant in the same groups of subjects. In summary, the present study has shown that the use of short dialysis sessions with highly efficient hemodialysis treatments, namely, HFD and HDF, that provide adequate dialysis (as evidenced by a Kt/V ⱖ1.3) was associated with similar levels of blood pressure control in ESRD patients. Prospective randomized studies are needed to determine the effects of high-efficiency treatments and shorter dialysis times on blood pressure control in the same groups of subjects. REFERENCES 1. von Albertini B, Miller JH, Gardner PW, Shinaberger JH: High-flux hemodiafiltration under six hours/week treatment. Trans Am Soc Artif Int Organs 30:227-231, 1984 2. Miller JH, von Albertini B, Gardner PW, Shinaberger
623
JH: Technical aspects of high flux hemodiafiltration for adequate short (under 2 hours) treatment. Trans Am Soc Artif Int Organs 30:377-381, 1984 3. Collins A, Ilstrup K, Hanson G, Berkseth R, Keshaviah P: Rapid high-efficiency hemodialysis. Artif Organs 10:185188, 1986 4. Acchiardo S, Burk L, Banister D: High-flux hemodialysis. Kidney Int 31:226-228, 1987 5. von Albertini B, Bosch JP: Short hemodialysis. Am J Nephrol 11:169-173, 1991 6. Charra B, Calemard E, Ruffet M, Chazot C, Terrat JC, Vanel T, Laurent G: Survival as an index of adequacy of dialysis. Kidney Int 41:1286-1291, 1992 7. Daugirdas JT: The pre:post dialysis plasma urea nitrogen ratio to estimate Kt/V and NPCR: Validation. Int J Artif Organs 12:420-427, 1989 8. Ritz E, Strumpf C, Katz F, Wing AJ, Quellhorst E: Hypertension and cardiovascular risk factors in hemodialyzed diabetic patients. Hypertension 7:II-118-II-124, 1985 (suppl II) 9. Cheigh JS, Milite C, Sullivan JF, Rubin AL, Stenzel KH: Hypertension is not adequately controlled in hemodialysis patients. Am J Kidney Dis 19:453-459, 1992 10. Salem MM: Hypertension in the hemodialysis population: A survey of 649 patients. Am J Kidney Dis 26:461468, 1995 11. Dhakal M, Sloand JA, Schiff MJ: Prevalence of hypertension (HTN) and adequacy of blood pressure (BP) control in hemodialysis (HD) patients (Pts) in the 90’s. J Am Soc Nephrol 7:1444, 1996 (abstr) 12. Eschbach JW, Abdulhadi MH, Browne JK, Delano BG, Downing MR, Egrie JC, Evans RW, Friedman EA, Graber SE, Haley NR, Korbet S, Krantz SB, Lundin AP, Nissenson AR, Ogden DA, Paganini EP, Rader B, Rutsky EA, Stivelman J, Stone WJ, Teschan P, Van Stone JC, Van Wyck DB, Zuckerman K, Adamson JW: Recombinant human erythropoietin in anemic patients with end-stage renal disease: Results of a phase III multicenter clinical trial. Ann Intern Med 111:992-1000, 1989 13. Canadian Erythropoietin Study Group: Effect of recombinant human erythropoietin therapy on blood pressure in hemodialysis patients. Am J Nephrol 11:23-26, 1991 14. Abraham PA, Macres MG: Blood pressure in hemodialysis patients during amelioration of anemia with erythropoietin. J Am Soc Nephrol 2:927-936, 1991 15. Kooman JP, Galdziwa U, Bocker G, Wijnen JA, Bortel L, Luik AJ, de Leeuw PW, van Hoff JP, Leunissen KM: Blood pressure during the interdialytic period in hemodialysis patients: Estimation of representative blood pressure values. Nephrol Dial Transplant 7:917-923, 1992 16. Sherman RA, Daniel A, Cody RP: The effect of interdialytic weight gain on predialysis blood pressure. Artif Organs 17:770-774, 1993 17. Victor RG: Reflex control of sympathetic nerve discharge in uremia. Blood Purif 15:43-45, 1997 (suppl 1) 18. Luik AJ, vd Sande FM, Weideman P, Cherieux EC, Leunissen KML: The influence of dry weight and dialysis treatment time on blood pressure (BP) control in hemodialysis patients. J Am Soc Nephrol 7:1444, 1996 (abstr)
I
N RECENT YEARS, the introduction of highly efficient hemodialytic techniques that use large surface area membranes and high blood flow rates, namely, high-flux hemodialysis (HFD) and high-flux hemodiafiltration (HDF), has added significant improvements in dialysis delivery in patients with end-stage renal disease (ESRD).1-5
From the Division of Renal Diseases and Hypertension, Department of Medicine, George Washington University Medical Center, Washington, DC. Received April 21, 1997; accepted in revised form October 10, 1997. Address reprint requests to Manuel T. Velasquez, MD, Division of Renal Diseases and Hypertension, Department of Medicine, George Washington University Medical Center, 2150 Pennsylvania Ave, NW, Room 4-426, Washington, DC 20037.
r 1998 by the National Kidney Foundation, Inc. 0272-6386/98/3104-0007$3.00/0 618
These techniques combine conventional diffusive hemodialysis with the large convective component of hemofiltration, and improve both efficacy and efficiency of hemodialysis treatment by increasing the rate of solute removal, as well as the range or molecular size of solutes removed. The major advantage of these therapies is that they provide adequate dialysis in a shorter time. However, it has been suggested that the shorter treatment times may limit the rate of ultrafiltration (UF) and lead to inadequate removal of fluid during hemodialysis. Such failure to completely remove excess fluid is thought to account for the high incidence or persistence of hypertension in hemodialysis patients. This hypothesis is based on the observation by Charra et al6 in Tassin, France, that long hemodialysis treatment 8 hours three times a week is associated with wellcontrolled hypertension. However, studies are
American Journal of Kidney Diseases, Vol 31, No 4 (April), 1998: pp 618-623
HYPERTENSION CONTROL IN HEMODIALYSIS PATIENTS
lacking regarding the status of blood pressure control in hemodialysis patients treated with the newer high-flux dialyzers and shorter dialysis sessions. We therefore compared the status of hypertension and adequacy of blood pressure control in ESRD patients receiving four different modalities of hemodialysis treatments and sought to determine whether the shorter treatment times used in highly efficient dialysis techniques have an adverse effect on blood pressure control in these patients. PATIENTS AND METHODS Seventy-three adult patients with ESRD who were receiving chronic hemodialysis treatment at the George Washington University Ambulatory Dialysis Center were studied. All patients had been on maintenance hemodialysis for at least 3 months before the study. Patients were divided into four groups according to the modality of hemodialysis treatment they received: conventional hemodialysis (CHD) with the use of cuprophan membranes having a 1.1 m2 surface area at a blood flow rate of 300 mL/min, high-efficiency hemodialysis (HED) with cuprophan ⫻ 1.6 m2 surface area at a blood flow rate of 450 to 500 mL/min, HFD with a F80 polysulfone membrane ⫻ 1.8 m2 surface area (Fresenius, Walnut Creek, CA) at a blood flow rate of 500 mL/min, and HDF with two F80 dialyzers 2 ⫻ 1.8 m2 surface area in series at a blood flow rate of 600 to 650 mL/min. All hemodialysis treatments were conducted three times a week using bicarbonate dialysate with constant sodium concentration. For both HFD and HDF, dialyzers were reused up to 12 times after reprocessing with 4% formaldehyde. During hemodialysis, UF was performed in each patient to achieve dry weight. Dry weight was defined as the weight below which further fluid removal during dialysis results in hypotension and/or symptoms of dizziness or muscle cramps and above which signs and symptoms of fluid expansion, such as rales in the chest and pedal edema, are evident. Blood samples for routine hematologic and biochemical evaluations were obtained monthly. Monthly predialysis and postdialysis blood samples for urea determination were also obtained at the beginning of the week. Adequacy of dialysis was determined monthly by urea kinetic analysis using a posttreatment blood sample for urea obtained at 10 minutes after the end of treatment. The formula of Daugirdas7 was used to calculate Kt/V urea. In all patients, demographic data were collected that included age, gender, race, primary ESRD diagnosis, body weight, blood pressure, hematocrit, biochemical data, and current use of antihypertensive agents and recombinant human erythropoietin (rHuEPO). Measurements of predialysis and postdialysis body weight and blood pressure were routinely obtained by the dialysis technician or nurse at each dialysis session. Interdialytic weight gain was calculated as the amount of weight gained between the end of a preceding dialysis session and the start of the next dialysis session. Predialysis systolic blood pres-
619
sure (SBP) and diastolic blood pressure (DBP) were measured with a standard mercury sphygmomanometer and taken with the patient in the sitting position. Mean arterial pressure (MAP) was calculated as the sum of DBP and one third of the difference in SBP and DBP or pulse pressure. Patients were considered to be hypertensive if they were receiving antihypertensive medications during the study period to control their hypertension. Any patient with an SBP less than 150 mm Hg and a DBP less than 90 mm Hg (based on the weekly average of predialysis blood pressure measurements taken over a 1-month period) was considered to have adequately controlled hypertension.
Statistical Analysis Results are expressed as mean values ⫾ SD unless otherwise indicated. Comparisons among groups were performed using ANOVA. Chi-square analysis was used for comparison of nominal data. Simple regression analysis was used for calculation of the correlation coefficients. Statistical differences were considered significant at P ⬍ 0.05.
RESULTS
The demographic characteristics of the four groups of patients are summarized in Table 1. The patients treated with HED, HFD, and HDF tended to be older compared with the patients in the CHD group. Note that well over 75% of the patients in each treatment group were black. Hypertension and diabetes were the two most common causes of renal disease in all groups. Of the 73 hemodialysis patients, 30 (41%) were receiving one or more antihypertensive drugs to control their hypertension and 43 (59%) were not (Table 2). The percentage of patients receiving antihypertensive medication was less in the HED, HFD, and HDF groups compared with the CHD group: 38%, 39%, and 39%, respectively, versus 56%. However, the difference between the four dialytic groups did not reach statistical significance (P ⫽ NS, chi-square test). Table 2 also shows that the percentage of patients achieving an SBP less than 150 mm Hg and a DBP less than 90 mm Hg was higher in the groups receiving high-efficiency treatments compared with the CHD group, but this difference was not statistically significant (P ⫽ NS, chisquare test). Fifty-six (77%) of the 73 patients were also receiving rHuEPO therapy (Table 3). However, the proportion of patients on rHuEPO did not differ among the four groups. The dose of rHuEPO tended to be lower in the HED, HFD, and HDF groups compared with the CHD group, but the difference between groups was not statis-
620
VELASQUEZ ET AL Table 1. Demographic Characteristics of Study Patients CHD (%) HED (%) HFD (%) HDF (%)
No. of patients 9 13 33 18 Mean age ⫾ SE (yr) 40 ⫾ 16 55 ⫾ 18 55 ⫾ 12 48 ⫾ 14 Race Black 7 (78) 12 (92) 27 (82) 16 (89) White 1 (11) 1 (8) 6 (18) 2 (11) Asian 1 (11) 0 (0) 0 (0) 0 (0) Gender Male 7 (78) 6 (46) 15 (45) 16 (89) Female 2 (22) 7 (54) 18 (55) 2 (11) Cause of renal disease Hypertension 4 (44) 5 (39) 13 (40) 6 (33) Diabetes 1 (11) 3 (23) 9 (27) 3 (17) Chronic glomerulonephritis 1 (11) 0 (0) 5 (15) 3 (17) Chronic interstitial nephritis 0 (0) 2 (15) 1 (3) 1 (5) Adult polycystic kidney disease 0 (0) 1 (8) 1 (3) 2 (11) Other 3 (34) 2 (15) 4 (12) 3 (17)
tically significant. Similarly, there was no significant difference in mean hematocrit and biochemical data among the four groups. Table 4 summarizes the average treatment time per dialysis session, UF rate, interdialytic weight gain, adequacy of dialysis, and blood pressure in each of the four treatment groups. Compared with the CHD group, the average Table 2. Use of Antihypertensive Medication and Control of Blood Pressure in ESRD Patients Treated With Four Different Hemodialysis Techniques
No. (%) of patients on antihypertensive medication One drug Two drugs Three or more drugs Total No. (%) of patients with SBP ⬍150 mm Hg DBP ⬍90 mm Hg
CHD, n⫽9 (%)
HED, n ⫽ 13 (%)
HFD, n ⫽ 33 (%)
HDF, n ⫽ 18 (%)
2 (23) 3 (33)
3 (23) 2 (15)
8 (24) 4 (12)
3 (17) 4 (22)
0 (0) 5 (56)
0 (0) 5 (38)
1 (3) 13 (39)
0 (0) 7 (39)
4 (44) 5 (56)
8 (62) 10 (77)
19 (58) 25 (76)
11 (61) 14 (78)
NOTE. Differences between groups are not significant (chi-square test).
treatment times in the HED and HFD groups were shorter, with HDF showing the shortest mean treatment time of 157 ⫾ 41 minutes per session. There also was a trend for higher UF rates in HED and HFD treatments compared with CHD. The mean UF rate in the HDF group was 19.8 ⫾ 6.1 mL/min compared with a mean of 15.4 ⫾ 4.7 mL/min in the CHD group (P ⬍ 0.05). The mean Kt/V was significantly higher in the groups treated with HED, HFD, or HDF (1.31 ⫾ 0.3, 1.30 ⫾ 0.4, and 1.43 ⫾ 0.3, respectively) compared with the CHD group (1.12 ⫾ 0.3; P ⬍ 0.05). The average predialysis blood pressure of all patients was 144/89 mm Hg, which did not differ between the four groups. There also were no differences in magnitude of interdialytic weight gain and predialysis MAP among the four groups. The relationships between hemodialysis treatment time, UF, Kt/V, interdialytic weight gain, and blood pressure for all patients were also examined using correlation analysis. The results show no significant correlation between predialysis MAP and dialysis treatment time (r ⫽ 0.04, P ⫽ NS), as shown in Fig 1. There was no correlation between predialysis MAP and either Kt/V (r ⫽ 0.03, P ⫽ NS), UF (r ⫽ 0.06, P ⫽ NS), or interdialytic weight gain (r ⫽ ⫺0.08, P ⫽ NS). There also was no significant association between Kt/V and use of antihypertensive medications (chi-square ⫽ 1.76, P ⫽ NS). There was, however, a significant positive correlation between interdialytic weight gain and treatment time (r ⫽ 0.33, P ⬍ 0.01; Fig 2). DISCUSSION
To date, many dialysis centers commonly prescribe dialysis treatment of 4 hours three times a week to most patients undergoing maintenance hemodialysis. However, for patients receiving HED treatments, the length of dialysis time per session is usually reduced. Although these highefficiency treatments provide adequate dialysis, the issue has been raised that the relatively shorter treatment times achieved with these therapies may adversely affect control of blood pressure. It is postulated that reducing dialysis treatment time may limit removal of excess fluid during hemodialysis and lead to poor control of hypertension. On the other hand, increasing treatment time allows slow and complete fluid removal and
HYPERTENSION CONTROL IN HEMODIALYSIS PATIENTS
621
Table 3. Use and Dose of rHuEPO, Hematocrit, and Biochemical Data in ESRD Patients Treated With Four Different Hemodialysis Techniques
No. of patients on rHuEpo (%) Dose of rHuEPO (U/kg/wk) Hematocrit (%) Creatinine (mg/dL) Glucose (mg/dL) Cholesterol (mg/dL) Albumin (g/dL) Calcium (mg/dL) Phosphorus (mg/dL) Alkaline phosphatase
CHD (n ⫽ 9)
HED (n ⫽ 13)
HFD (n ⫽ 33)
HDF (n ⫽ 18)
7 (78) 80.2 ⫾ 40.0 28.5 ⫾ 4.0 15.0 ⫾ 4.1 93 ⫾ 33 158 ⫾ 29 4.0 ⫾ 0.6 8.9 ⫾ 0.8 6.5 ⫾ 1.9 144 ⫾ 89
10 (77) 74.3 ⫾ 32.0 30.2 ⫾ 6.0 12.1 ⫾ 4.2 93 ⫾ 29 164 ⫾ 34 3.7 ⫾ 0.9 9.1 ⫾ 0.5 5.9 ⫾ 2.1 136 ⫾ 73
25 (76) 73.1 ⫾ 39.0 31.3 ⫾ 3.0 13.8 ⫾ 3.6 100 ⫾ 55 164 ⫾ 27 3.9 ⫾ 0.3 9.4 ⫾ 1.3 6.3 ⫾ 2.4 104 ⫾ 49
14 (78) 61.6 ⫾ 33.0 31.6 ⫾ 3.0 15.5 ⫾ 3.9 93 ⫾ 30 152 ⫾ 29 4.1 ⫾ 0.3 9.3 ⫾ 1.2 5.6 ⫾ 1.9 95 ⫾ 45
NOTE. Data are given as mean values ⫾ SD. Differences between groups are not significant (ANOVA).
results in improved control of blood pressure. This is supported by the studies of Charra et al6 in Tassin, France, which showed that long hemodialysis treatment sessions with Kiil dialyzers (8 hours three times a week) is associated with good control of hypertension and discontinuation of antihypertensive drug therapy in approximately 98% of their dialysis patients. In the present study, we addressed this issue by comparing the status of blood pressure control in our ESRD patients receiving either CHD or high-efficiency treatments. We found that highefficiency treatments with shorter dialysis times, namely, HED, HFD, and HDF, were not associated with a higher incidence of hypertension (use of antihypertensive medication) compared with CHD. Furthermore, control of hypertension (defined as SBP ⬍150 mm Hg or DBP ⬍90 mm Hg) was not worse in patients receiving high-effi-
ciency treatments compared with patients treated with CHD. Indeed, the majority of the patients receiving high-efficiency treatments achieved adequate control of hypertension, with over 55% of patients having an SBP less than 150 mm Hg and over 75% showing a DBP less than 90 mm Hg. Thus, it appears that control of hypertension in our patients was not negatively impacted by high-efficiency treatments compared with CHD. Previous studies on the prevalence rates and control of hypertension in ESRD patients undergoing hemodialysis have yielded variable results. Ritz et al8 reported that 96% of diabetic patients and 93% of nondiabetic patients receiving chronic hemodialysis required antihypertensive medication to control their hypertension. In this study, only 13% of diabetic patients and 28% of nondiabetic patients had SBPs below 160 mm Hg. Cheigh et al9 found that approximately half
Table 4. Comparison of Dialysis Treatment Time, Body Weight, Interdialytic Weight Gain, Adequacy of Dialysis, and Blood Pressure in Hemodialysis Patients
Treatment time (min) Predialysis body weight (kg) Ultrafiltration rate (mL/min) Interdialytic weight gain (kg) Kt/V Predialysis blood pressure (mm Hg) SBP DBP MAP
CHD (n ⫽ 9)
HED (n ⫽ 13)
HFD (n ⫽ 33)
HDF (n ⫽ 18)
230 ⫾ 21 68.4 ⫾ 18.0 15.4 ⫾ 4.7 3.3 ⫾ 1.8 1.12 ⫾ 0.30
184 ⫾ 31* 68.8 ⫾ 16.0 17.0 ⫾ 6.2 3.3 ⫾ 1.1 1.31 ⫾ 0.30*
186 ⫾ 52* 73.1 ⫾ 39.0 18.2 ⫾ 5.9 3.6 ⫾ 1.7 1.30 ⫾ 0.40*
157 ⫾ 41* 71.6 ⫾ 33 19.8 ⫾ 6.1* 3.2 ⫾ 1.6 1.43 ⫾ 0.30*
146 ⫾ 16 89 ⫾ 14 104 ⫾ 12
139 ⫾ 21 86 ⫾ 15 96 ⫾ 12
148 ⫾ 15 92 ⫾ 14 101 ⫾ 15
148 ⫾ 14 88 ⫾ 15 102 ⫾ 28
NOTE. Data are given as mean values ⫾ SD. *Significantly different from CHD; P ⬍ 0.05.
622
Fig 1. Correlation between dialysis treatment time and predialysis MAP.
(44%) of the 244 chronic hemodialysis patients in their series were receiving antihypertensive medications. They also observed a high incidence of uncontrolled hypertension (defined as the percentage of ambulatory SBP ⬎150 mm Hg and DBP ⬎90 mm Hg) in this study, with only 15% of patients maintaining a blood pressure below 150/90 mm Hg at all times, despite antihypertensive drug therapy. Recently, Salem,10 in a survey of 649 patients receiving hemodialysis with cuprophan dialyzer 4 hours three times a week, showed a 72% prevalence rate of hypertension defined as MAP greater than 114 mm Hg or blood pressure greater than 160/90 mm Hg. Approximately 59% of the total patients were receiving antihypertensive medication. More recently, Dhakal et al11 showed that 67% of 96 patients on chronic hemodialysis had uncontrolled hyperten-
Fig 2. Correlation between dialysis treatment time and interdialytic weight gain.
VELASQUEZ ET AL
sion defined as SBP greater than 150 mm Hg or DBP greater than 90 mm Hg. The differences in the prevalence rates and control of hypertension between these reports and our study are difficult to interpret, but they may relate to differences in demographics, the type and amount of hemodialysis treatment, and volume status of the study patients. In the present study, the majority of our patients were blacks in whom hypertension and diabetes were identified as the two most frequent probable causes of ESRD. However, there were no significant differences in either race or etiology of renal disease among the four groups of patients. The majority of our patients were also receiving rHuEPO treatment for maintenance treatment of anemia. Since an increase in blood pressure is a well-documented side effect of rHuEPO therapy,12-14 this could have added to sustained hypertension and increased need for antihypertensive drug therapy in some of these patients. However, we did not find significant differences in the use of rHuEPO among the four groups of patients. Furthermore, the average weekly doses of rHuEPO used and mean hematocrit values did not differ between the four groups. Therefore, it seems unlikely that the use of rHuEPO and the degree (or correction) of anemia had a different influence on the magnitude of blood pressure control among the four dialytic groups. Salt and water retention is considered the dominant factor contributing to hypertension in ESRD patients. In hemodialysis, the persistence of hypertension in some patients may relate to volume overload that occurs during the interdialytic period. In the present study, interdialytic weight gain, which probably reflects increased fluid volume, did not differ among the four groups of patients. Moreover, there was no clear relationship between predialysis MAP and interdialytic weight gain, a finding that has been observed by other groups in hemodialysis patients.10,15,16 Therefore, differences in blood pressure control between high-efficiency treatments and CHD cannot be attributed to differences in interdialytic weight gain. It is interesting that we observed a direct correlation between treatment time and interdialytic weight gain. The significance of this finding is unclear. Even though treatment time was reduced in patients treated
HYPERTENSION CONTROL IN HEMODIALYSIS PATIENTS
with HED, HFD, and HDF, the amount of dialysis delivered with these therapies was adequate and more than that achieved with CHD, as indicated by the higher Kt/V urea of ⱖ1.3, in HED, HFD, and HDF patients compared with patients treated with CHD. It is possible that highefficiency treatments provide better clearance of vasoconstrictor substances or middle molecules potentially capable of sustaining hypertension in ESRD patients.17 Additional studies are needed to examine this point. Our results do not necessarily contradict the view that long hemodialysis treatment can provide better control of blood pressure because we did not evaluate the effects of increasing treatment time with high-efficiency techniques in the same group of patients. However, Luik et al18 prospectively studied the effect of increasing dialysis treatment time on blood pressure control in a small group of patients. These investigators found that increasing treatment time by 2 hours without a change in dry weight resulted in a decrease in the use of antihypertensive medication as well as a reduction in interdialytic SBP. Further studies involving a larger number of hemodialysis patients are needed to confirm these findings. On the other hand, other prospective studies are also needed to evaluate the effects on blood pressure control of reducing treatment time with high-efficiency treatments but keeping all other variables (eg, type of dialyzer, dry weight, and Kt/V) constant in the same groups of subjects. In summary, the present study has shown that the use of short dialysis sessions with highly efficient hemodialysis treatments, namely, HFD and HDF, that provide adequate dialysis (as evidenced by a Kt/V ⱖ1.3) was associated with similar levels of blood pressure control in ESRD patients. Prospective randomized studies are needed to determine the effects of high-efficiency treatments and shorter dialysis times on blood pressure control in the same groups of subjects. REFERENCES 1. von Albertini B, Miller JH, Gardner PW, Shinaberger JH: High-flux hemodiafiltration under six hours/week treatment. Trans Am Soc Artif Int Organs 30:227-231, 1984 2. Miller JH, von Albertini B, Gardner PW, Shinaberger
623
JH: Technical aspects of high flux hemodiafiltration for adequate short (under 2 hours) treatment. Trans Am Soc Artif Int Organs 30:377-381, 1984 3. Collins A, Ilstrup K, Hanson G, Berkseth R, Keshaviah P: Rapid high-efficiency hemodialysis. Artif Organs 10:185188, 1986 4. Acchiardo S, Burk L, Banister D: High-flux hemodialysis. Kidney Int 31:226-228, 1987 5. von Albertini B, Bosch JP: Short hemodialysis. Am J Nephrol 11:169-173, 1991 6. Charra B, Calemard E, Ruffet M, Chazot C, Terrat JC, Vanel T, Laurent G: Survival as an index of adequacy of dialysis. Kidney Int 41:1286-1291, 1992 7. Daugirdas JT: The pre:post dialysis plasma urea nitrogen ratio to estimate Kt/V and NPCR: Validation. Int J Artif Organs 12:420-427, 1989 8. Ritz E, Strumpf C, Katz F, Wing AJ, Quellhorst E: Hypertension and cardiovascular risk factors in hemodialyzed diabetic patients. Hypertension 7:II-118-II-124, 1985 (suppl II) 9. Cheigh JS, Milite C, Sullivan JF, Rubin AL, Stenzel KH: Hypertension is not adequately controlled in hemodialysis patients. Am J Kidney Dis 19:453-459, 1992 10. Salem MM: Hypertension in the hemodialysis population: A survey of 649 patients. Am J Kidney Dis 26:461468, 1995 11. Dhakal M, Sloand JA, Schiff MJ: Prevalence of hypertension (HTN) and adequacy of blood pressure (BP) control in hemodialysis (HD) patients (Pts) in the 90’s. J Am Soc Nephrol 7:1444, 1996 (abstr) 12. Eschbach JW, Abdulhadi MH, Browne JK, Delano BG, Downing MR, Egrie JC, Evans RW, Friedman EA, Graber SE, Haley NR, Korbet S, Krantz SB, Lundin AP, Nissenson AR, Ogden DA, Paganini EP, Rader B, Rutsky EA, Stivelman J, Stone WJ, Teschan P, Van Stone JC, Van Wyck DB, Zuckerman K, Adamson JW: Recombinant human erythropoietin in anemic patients with end-stage renal disease: Results of a phase III multicenter clinical trial. Ann Intern Med 111:992-1000, 1989 13. Canadian Erythropoietin Study Group: Effect of recombinant human erythropoietin therapy on blood pressure in hemodialysis patients. Am J Nephrol 11:23-26, 1991 14. Abraham PA, Macres MG: Blood pressure in hemodialysis patients during amelioration of anemia with erythropoietin. J Am Soc Nephrol 2:927-936, 1991 15. Kooman JP, Galdziwa U, Bocker G, Wijnen JA, Bortel L, Luik AJ, de Leeuw PW, van Hoff JP, Leunissen KM: Blood pressure during the interdialytic period in hemodialysis patients: Estimation of representative blood pressure values. Nephrol Dial Transplant 7:917-923, 1992 16. Sherman RA, Daniel A, Cody RP: The effect of interdialytic weight gain on predialysis blood pressure. Artif Organs 17:770-774, 1993 17. Victor RG: Reflex control of sympathetic nerve discharge in uremia. Blood Purif 15:43-45, 1997 (suppl 1) 18. Luik AJ, vd Sande FM, Weideman P, Cherieux EC, Leunissen KML: The influence of dry weight and dialysis treatment time on blood pressure (BP) control in hemodialysis patients. J Am Soc Nephrol 7:1444, 1996 (abstr)