Improved Urea Reduction Ration and Kt/V in Large Hemodialysis Patients Using Two Dialyzers in Parallel

Kathleen M. Powers N.P., Michael J. Wilkowski M.D., Allen W. Helmandollar, Karl G. Koenig M.D., W. Kline Bolton M.D.
Division of Nephrology, University of Virginia Health Sciences Center, Charlottesville, Virginia, 22908 USA

Kathleen M. Powers, R.N., N.P.
Nephrology Nurse Practitioner
Division of Nephrology, Box 133
University of Virginia Heath Sciences Center
Charlottesville, VA 22908
Karl G. Koenig, M.D.
Assistant Professor of Medicine
Division of Nephrology, Box 133
Michael J. Wilkowski, M.D.
Associate Professor of Medicine
Division of Nephrology, Box 133
University of Virginia Health Sciences Center
Charlottesville, VA 22908
W. Kline Bolton, M.D.
Professor of Medicine
Chief, Division of Nephrology
Box 133
Univeristy of Virginia Health Sciences Center
Charlottesville, VA 22908
Allen W. Helmandollar, B.S.
Chief Technician
Department of Renal Services, Box 447-51
University of Virginia Health Sciences Center
Charlottesville, VA 22908


Address requests for reprints to:
          W. KLINE BOLTON, M.D.
          BOX 133
          TELEPHONE:    804-924-5125
          FAX:                   804-924-5848


Running Head: Hemodialysis Using Two Dialyzers in Parallel


Delivered dose of hemodialysis (HD) in large patients with end stage renal disease is often less than adequate. Fourteen chronic HD patients with weights > 80 kg participated in a prospective, cross-over study comparing urea reduction ratio (URR± SEM) and the fractional clearance index for urea (eKt/Vurea± SEM ) on a single polysulfone dialyzer for a control (HDC) period of 4 wks versus clearances obtained with two dialyzers in parallel during an intervention (HDP) period of 4 wks. Clearance of the surrogate middle molecule iohexol (CIo) was also measured. Health status was assessed with the SF-36. Blood and dialysate flow rates and duration of HD sessions were constant. URR increased from 0.67 +/-0.006 during HDC to 0.72 +/-0.006 with HDP (p<0.0001). eKt/Vurea increased from 1.16 +/-0.021 to 1.34 +/-0.021 (p<0.0001). Increased URR and eKt/Vurea occurred in all 14 during HDP (p < 0.05). CIo during HDP averaged 182 +/- 7.7 ml/min compared to 131 +/- 5.4 ml/min in HDC sessions (p<0.00001). Health status improved in six of eight categories. Expense increased approximately $14.27 per dialysis with HDP. In 11/14 patients continued on two dialyzers in parallel for one year, monthly eKt/V averaged 1.46 +/-0.066 and health status further improved in five of eight categories. In large patients, two dialyzers in parallel increased urea and iohexol clearance. Increased urea clearance was maintained for one year and health status improved.


          The delivered dose of hemodialysis (HD) in large patients with end stage renal disease (ESRD), as measured by urea reduction ratio (URR) and fractional clearance index for urea (eKt/Vurea), is often less than adequate.[1] Use of large surface area dialyzers, high blood flow rates, high dialysate flow rates and increased dialysis time are maneuvers that can increase the delivered dose of hemodialysis. Despite optimization of these parameters, patients weighing greater than 80 kg are at risk of receiving less than adequate doses of HD.[2-4] In an effort to optimize dialysis in this group of patients, we sought to improve the delivered dose of HD, measured by URR, eKt/Vurea and iohexol clearance, using two dialyzers inserted into the HD circuit in parallel. We compared this technique to a single dialyzer in the same patients using the identical blood and dialysate flow rates and duration of HD sessions.


          Fourteen patients with ESRD on maintenance HD three times per week at the University of Virginia Kidney Center were enrolled in this study. All were greater than 80 kg body weight and were medically stable for at least 90 days preceding the initiation of the study. All had HD vascular access (PTFE grafts) which provided blood flow rates of 400 ml/min and used 15 gauge needles. All used a dialysate flow rate of 800 ml/min. All patients dialyzed for four hours. All patients used the Fresenius 2008E and H dialysis machines and Fresenius F-80A (hollow fiber, polysulfone, 2.1 m2 surface area, high flux) dialyzers.[5] All dialyzers were reprocessed using heat and citric acid with a reuse limit of ten.[6,7]

          This was a prospective, crossover study. Seven patients used the single dialyzer (HDC) for twelve consecutive sessions (four weeks) and seven patients used the double dialyzers in parallel (HDP) for twelve consecutive sessions (four weeks) with measurement of URR and eKt/Vurea after every session. The next twelve consecutive sessions (four weeks) each group switched to the opposite dialyzer configuration and URR and eKt/Vurea were measured after each session.[8] Patients were randomly assigned to start the study in the control or intervention period. Pre- and post-dialysis urea samples were drawn from the arterial vascular access needle and the post dialysis sample was drawn after the cessation of dialysis.[9] Calculation of eKt/Vurea used the single pool second generation formula of Daugirdas: sp Kt/Vurea = -ln(R - 0.008t) + (4 - 3.5R)UF/W, where R = postBUN/preBUN, t = dialysis session length in hours, UF = weight loss in kilograms and W = post-dialysis weight in kilograms.[10] Equilibrated Kt/Vurea (eKt/Vurea) was calculated using spKt/Vurea above and the rate equation which adjusts for urea rebound: eKt/Vurea = spKt/Vurea – (0.6 X spKt/Vurea )/t + 0.03, where t is dialysis time in hours.[11] Calculation of URR used the formula URR = (preBUN – postBUN)/preBUN.[12]

          During HDP, two Fresenius F-80A artificial kidneys were used in a parallel configuration during each HD session (Figure 1). The parallel configuration used a Y-connector (MPC-800, Molded Products, Harlan, IA) at the end of each standard blood line to attach to the blood ports of the artificial kidneys. To deliver dialysate in the parallel configuration, the following were attached to the ends of the dialysate lines: a male Hanson barbed connector (Part # 641985, Fresenius, Inc., Walnut Creek, CA), a length of standard reinforced dialysate tubing (Part # 545327, Fresenius, Inc., Walnut Creek, CA), a Y-connector (Part # 674650, Fresenius, Inc., Walnut Creek, CA), two lengths of standard reinforced dialysate tubing with red and blue dialysate connectors (Parts # 641984, 641985, Fresenius, Inc., Walnut Creek, CA). As an alternative source for these products, Molded Products offers the following assembled dialysate lines (MPC-725 and MPC-750). The red and blue dialysate connectors were attached to the dialysate ports of the artificial kidney in the usual manner.

          Prior to this clinical study, an in-vitro experiment with blood line circuits containing normal saline demonstrated an equal distribution of flow to each dialyzer in the parallel configuration based on measurements over two hours on six circuit set-ups. Similar measurements demonstrated 400 ml/min of dialysate flow distributed to each dialyzer in the parallel configuration when the total dialysate flow was set at 800 ml/min. Blood flows were measured directly at the level of the dialyzers in six patients with doppler measurement of blood velocity (7-Mhz transducer, Acuson XP128; Acuson, Inc. Mountain View, CA). With knowledge of the internal diameter of the connectors to each dialyzer and blood velocity, the blood flow delivered to each dialyzer could be estimated. Differences between blood flow between the two dialyzers in the parallel configuration were in the range of 17 to 37 ml/min when the blood pump registered 400 ml/min. Blood flow rates were kept constant at 400 ml/min and dialysate flow rates were kept constant at 800 ml/min during the control and intervention periods. Therefore, in the parallel configuration, each dialyzer would receive approximately 200 ml/min of blood flow and 400 ml/min of dialysate flow when the dialysis machine blood flow and dialysate flow rates are set at 400 ml/min and 800 ml/min, respectively. During the clinical study, equal distribution of the blood flow to each dialyzer in the parallel configuration was confirmed by visual inspection. Uniform filling of the hollow fibers of each dialyzer was observed at the initiation of each dialysis session and inspection of the fibers during rinse-back was made for evidence of dialyzer clotting or underperfusion.

          Duration of the sessions (four hours) was the same in the control and intervention periods. Heparinization prescription did not change, with the double dialyzers in parallel receiving the same amount of heparin as previous sessions required, with adjustment by the nursing staff. The indication for the change in heparin infusion was an observation by the nursing staff of fiber clotting during the rinse back portion of the dialysis procedure. A second indicator is the observation by the reprocessing technician of clot-like strands flushed from the dialyzer during the cleaning of the used dialyzer performed routinely as part of the reprocessing program. Both indicators are factors which lead to an increase on heparin infusion in increments of 200 units per hour.

          In order to measure clearances of substances in the size range of middle molecules, the surrogate middle molecule iohexol, a freely dialyzable agent (MW 821), was injected (30 ml, 350 mg/ml) after the second to last dialysis session of the control and intervention periods allowing for equilibration of iohexol. [13] Pre- and post-dialysis blood samples from the last session of HDC and HDP were used to calculate an iohexol clearance (CIo). Since Kt/VIo = (CIoX t)/VdIo, then CIo= [(Kt/VIo ) X VdIo]/t where VdIo, the volume of distribution of iohexol, can be calculated as VdIo= (166 X weight in kilograms) + 2490 in men or (95 X weight in kilograms) + 6170 in women. [14] In this equation, the duration of dialysis session is in minutes, making the units of CIo ml/min. Kt/VIo was calculated with the Daugirdas formula above using RIo = post-dialysis iohexol concentration/ pre-dialysis iohexol concentration and t is the duration of the dialysis session in hours.[13,15] Iohexol reduction ratio (IoRR) was calculated as IoRR=1-RIo.

          During the cross over study, blood volume processed, amount of ultrafiltration and erythropoietin (Epogen, Amgen, Thousand Oaks, CA) doses were monitored for every session. Hematocrit level was monitored each week. Patients were monitored for adequacy of heparinization, episodes of clotted dialyzers, dialyzer reactions, blood leaks and signs of bacteremia during each HD session.

          As an assessment of long-term efficacy and safety, subjects who improved their urea clearance in the intervention period compared to the control period were continued on the double dialyzer in parallel configuration for one year. eKt/V, complete blood count, serum albumin were assessed monthly.

          The SF-36 health status assessment instrument (The Medical Outcomes Trust, Boston, MA) was administered to each patient at the end of the control and intervention periods.[16] During the subsequent twelve months on the double dialyzer in parallel configuration, the SF-36 was administered at the end of months six and twelve.

         The patients provided informed consent and the study was approved by the University of Virginia Health Sciences Center Investigational Review Board. The patients were randomly assigned to begin the study in HDC (n = 7) or HDP (n = 7) period. Data are presented as means +/- standard error of the mean (SEM). Comparisons between the control (HDC) and intervention (HDP) measurements were made with paired two-tailed t tests; p values <0.05 were considered statistically significant. Correlation coefficient (R) was used to compare change in eKt/Vurea with body surface area and body weight. On the SF-36, a change in transformed score of two or more is considered significant.[16]


          Demographic data describing the study patients are listed in Table 1. The mean age of the patients was 56 years and average weight was 105 kg. Eleven of fourteen patients were men.

          The clearance results of HDC and HDP period are presented in Table 2. A significant increase in URR and eKt/Vurea occurred in all 14 patients during the intervention compared to the control period. The mean URR and eKt/Vurea for the fourteen patients during HDC were 0.68 +/- 0.006 and 1.16 +/- 0.021 respectively, compared to 0.73 +/- 0.006 and 1.34 +/- 0.021 during HDP (p<0.0001). Figure 1 shows the change in eKt/Vurea for each individual patient and the aggregate average of the group. The average eKt/Vurea increase was 0.18 (15%), while average URR increased 4.7 (7% ) using the double dialyzer in parallel. The values for spKt/Vurea were 1.33 +/- 0.023 during control and 1.53 +/- 0.026 with the double dialyzer in parallel (p <0.0001). Absolute improvement in eKt/Vurea was weakly correlated to body surface area (BSA)(correlation coefficient R = -0.257). We observed in this study that patients less than 2.2 m2 improved their eKt/Vureain the range of 0.08 to 0.39 while patients greater than 2.2 m2 improved in the range of 0.11 to 0.17 during the intervention versus the control periods. Correlations were even weaker when comparing absolute improvement in eKt/Vurea with body weight (R = -0.0475) and percentage improvement in eKt/Vurea with either BSA (R = -0.00708) or body weight (R = 0.0294).

          Iohexol clearance, CIo, obtained in twelve patients averaged 182 +/- 7.7 ml/min during HDP compared to 131 +/- 5.4 ml/min in the HDC sessions (p<0.0001). Kt/VIo and IoRR increased from 1.79 +/- 0.068 and .78 +/- 0.010 to 2.47 +/- 0.011 and .85 +/- 0.012, respectively (p < 0.00001). (Figure 2) One patient completed the intervention period but received a transplant prior to the measurement of iohexol clearance at the end of the control period and was omitted from the iohexol analysis. One patient had a allergy to iodine and iohexol was withheld.

          Data for measurement of eKt/Vurea and URR was obtained for 301 of the 336 total sessions of the cross-over study (153 in the control period and 148 in the intervention period, p = NS). The exceptions included the patient called for transplantation and one patient who developed a clotted vascular access while in the intervention period. This resulted in a need for a single HD session via a femoral catheter on a single dialyzer and no measurements are included from this episode. The remainder of the omissions was due to inadvertent failure to obtain the post-dialysis BUN.

          Adverse events during the study included one admission for volume overload and one blood leak during the control period. During the intervention period one patient experienced a peri-access hematoma, one developed a clotted vascular access and one patient was treated for pneumonia. None of these events required hospitalization and none could be attributed to the double dialyzer configuration. There were no dialyzer reactions, clotted dialyzers or bacteremias during HDC or HDP.

          Data from the dialysis sessions from the cross-over study of each patient are presented in Table 3. Dialysis duration, average volume of blood processed per session and the average amount of ultrafiltration required did not differ between the control and intervention sessions. The mean heparin dose, Epogen dose and hematocrit levels for the group as an aggregate did not differ between HDC and HDP, but there were some differences among individual patients.

          Heparin requirements were different in four patients. In this study, no changes were made in the initial heparin load during the control or intervention periods and heparin re-bolus was not used. Three of these four patients required slightly more heparin infusion during HDP compared to HDC. The largest difference was 1500 units per treatment (approximately 400 units per hour) . One patient required more heparin during the control period.

          Mean Epogen doses differed in three patients; two of these three patients required less Epogen during HDP. During HDP one patient required an increase of approximately three thousand units of Epogen.

           The average of the weekly hematocrits differed significantly between the control and intervention periods in six patients; five of the six averaging a higher hematocrit during HDP. Two of these five patients had a significant difference in their Epogen dose; both required less Epogen during HDP compared to HDC.

          The average number of uses for each dialyzer was eight during this study and did not differ between the control and intervention period. Loss of 20% of total fiber bundle volume, an aesthetically displeasing artificial kidney or reaching the ten reprocesses limit were the reasons for the artificial kidneys to be discarded. Tracking of the fiber bundle volume, other than the 20% threshold, was not done.

          Following the cross-over study, eleven of the fourteen patients continued on the two dialyzers in parallel configuration for twelve months and sustained an average monthly eKt/V of 1.41 (± 0.023). Two patients did not complete twelve months on the double dialyzer because their access failed to maintain a blood flow rate of 400 ml/min. One patient did not continue for twelve months because he received a kidney transplant as noted above. The mean hematocrit level for this group was 32.9% (SEM ± 0.37) with a mean Epogen dose of 4290 units (SEM ± 347), not significantly different from the control period. Average serum albumin of the patients while on the double dialyzer in parallel configuration for twelve months was 3.80 grams/dL (SEM ± 0.06). When the values for the serum albumin from the twelve months prior to the study were retrospectively reviewed for these eleven patients it was 3.75 grams/dL (SEM ± 0.04)(p value = NS).

          The results of the SF-36 assessment of health status in the control period, intervention period, month six (HDP-6) and month twelve (HDP-12) on the double dialyzer configuration are listed in Table 4. A clinically significant increase was noted in six of eight domains when comparing the two week control and two week intervention periods. During the long-term period on the double dialyzer configuration, health status further improved in three domains with no deterioration in the other five. By month twelve, another significant increment occurred in five domains without change in three when compared to month six. At the six month and twelve month assessments, deterioration occurred in one domain (General Health) compared to the score during the two week intervention period but did not drop below the value of the control period. This overall improvement in health status compared to the control period corresponded to an average monthly eKt/V was 1.45 (± 0.032) at HDP-6 and 1.41 (± 0.023) at HDP-12.

          Cost analysis demonstrated an increased expense of approximately $14.27 per dialysis session using the double dialyzer in parallel configuration: the cost of a new F-80 dialyzer is approximately $32.00; the usual reuse number was eight thus making the extra dialyzer cost $4.00 per session; the cost of each reprocessing of a dialyzer is $5.10; the cost of the bloodline Y-connectors is $3.00 per session and they were not reused or reprocessed; additional nursing or technician time for set up was seven minutes at an estimated cost of $0.31 per minute or $2.17 per session. Estimated yearly cost increase would be $2,100 per patient.


          Use of the double dialyzer in a parallel configuration achieved an increment in delivered dialysis dose. Any increase may be particularly important for patients who have a high volume of distribution urea (Vdurea). The patients studied are at risk for receiving less than optimal dialysis despite maximizing blood and dialysate flow rates, surface area of artificial kidney and dialysis time. The percentage and absolute increase in clearance during the intervention period is clinically significant. Recent United States Renal Data System analysis has shown that a 0.1 increase in Kt/Vurea is associated with a substantially decreased adjusted relative risk of death from cardiac, cerebrovascular and infectious diseases.[17] The clinical effect of an increased Kt/Vurea may be even more beneficial for patients with a less than adequate Kt/Vurea (<1.2) who are increased to 1.2-1.4.[18] In patients who are greater than 80 kg body weight, a Kt/Vurea of >1.2 may be difficult to achieve because of their large Vdurea.[4] This method of two artificial kidneys used in parallel increases clearance after other important parameters such as blood flow rate, dialysate flow rate, artificial kidney size and time on dialysis have been optimized.

          In addition to increased urea clearance, we demonstrated increased clearance of iohexol with the double dialyzer in parallel configuration compared to the control period. Presumably this is due to the increase in surface area of the dialysis membrane. Iohexol's molecular weight approaches the size of middle molecules and can be reliably measured. Dialysis of middle molecules, substances with molecular weight in the range of 500-10,000 daltons, may have an additional clinical benefit apart from urea removal. If this hypotheses is confirmed in the current NIH sponsored multi-center hemodialysis study, then the double dialyzer in parallel configuration would be an additional strategy to increase middle molecule clearance by taking advantage of increased diffusive permeability or increased adsorption of these molecules by the dialyzer membrane.[19] The greater increase in Kt/VIo compared to eKt/Vurea is evidence of the relationship of clearance to Kt/V. Since the volume of distribution of iohexol is less than half that of urea, an increase in clearance of iohexol will result in a larger relative increase in Kt/VIo compared to the increase in urea clearance’s effect on eKt/Vurea.[13] Our present study did not try to determine clearances of molecules larger than iohexol such as beta-2-microglobulin.

          One might hypothesize that doubling the surface area of the dialysis membrane with two dialyzers in parallel should make a much greater impact on the increase of URR and eKt/Vurea than observed in this study. However, the surface area of the present generation of dialyzers, including the F-80A used in this study, is not the limiting factor in urea clearance with blood flow rates of 400 ml/min. Thus, doubling the dialyzer surface area would not be expected to double the urea clearance. This study does not support the usefulness of a new artificial kidney with twice the surface area of the dialyzers now available since such a product would no doubt be significantly more expensive than currently available dialyzers. Another limitation of this study is its applicability to large patients. We did not study average or small patients to see if a significant increase in URR or eKt/V could be achieved or if time could be shortened for patients with optimal URR or eKt/Vurea. The correlation between increasing patient size and decreasing improvement in eKt/Vurea in this study emphasizes the difficulty in optimizing clearances in large patients even when using techniques such as two dialyzers in parallel.

          Another important reason for the limited increase in eKt/Vurea and URR during HDP is that the dialyzer’s urea mass transfer-area coefficient (KoA) changes depending on dialysate flow rates.[20] When the dialyzers are in the parallel configuration and each is receiving 400 ml/min of dialysate, KoA is substantially less than it is with a 800 ml/min dialysate flow rate and a doubling of surface area does not result in a doubling of urea clearance. In this study we did not investigate the effect of varying dialysate flow rates on eKt/Vurea or URR with two dialyzers in the parallel configuration.

          Four HD characteristics which impact dialyzer clearance include heparinization, hematocrit levels, number of dialyzer reuses and vascular access recirculation. There were no differences in amount of heparin administered or hematocrit level between HDC and HDP for the entire group. Among individual patients there were four who required different heparin doses during HDC and HDP. The differences were small and there was no evidence that they affected clearances since there were no clotted dialyzers during either part of the study. Hematocrit levels were different in six patients but five out of six of these patients had higher hematocrits during HDP. Since clearance may decrease with increasing hematocrits, this difference would not account for the increased clearance obtained with HDP. The number of reuses for each dialyzer was the same in HDC and HDP. Access recirculation was not measured in this study and may have influenced the baseline values of eKt/Vurea and URR. The randomized, controlled structure of this study should have eliminated recirculation as a significant explanation for the improvement of clearances with the two dialyzers in parallel configuration.

          An important consideration in measuring Kt/Vurea and eKt/Vurea is the timing of the post-dialysis BUN sample. In this study we used a method of drawing blood from the arterial needle in the vascular access after hemodialysis was completely stopped. This method is not the slow flow method recommended in the Dialysis Outcome Quality Initiative (DOQI) practice guidelines.[21] By drawing the sample after the cessation of hemodialysis, some urea rebound may have occurred, slightly lowering the eKt/Vurea results.[22] The method was consistent in HDC and HDP and should not have affected the comparisons of the two periods. This same method was used to determine the monthly eKt/Vurea used to calculate HDP-6 and HDP-12. DOQI guidelines recommend calculating Kt/Vurea using the single pool method and await further confirmation on the usefulness of eKt/Vurea. Since most dialysis units use the single pool calculation, the absolute change in Kt/Vurea in a population similar to the patients in this study would be from 1.33 to 1.53. The percentage increase would remain the same at 15%.

          Recent reports of two F-80 dialyzers used in series have demonstrated increased urea clearance when used in a system for hemodialfiltration.[23-25] A major advantage of the two dialyzers in parallel used in the present study is the ability to utilize all the usual safety features that are standard on the Fresenius 2008E and H machines. The alarms do not have to be adjusted or bypassed and no clotting of the dialyzers or backfiltration was observed. This simple system of two dialyzers in parallel can be used in any dialysis unit without the need for costly special equipment.

          Following the cross over study, the subsequent twelve months of hemodialysis with two dialyzers in a parallel configuration demonstrated that the improved urea clearance can be expected to be sustained in patients with accesses that can deliver 400 ml/min of BFR. The SF-36 assessment also offers evidence that this increase in eKt/V is accompanied by improved health status.[26]

          Unit staff report that the set up time for the intervention period was approximately 5 to 7 minutes longer than the control period. Extra expense occurs for the connectors to allow for the distribution of blood flow and dialysate to each dialyzer. There is the added expense of reprocessing one additional dialyzer every session for the patients using this method. It should be noted that reuse of dialyzers more than eight times reported here would decrease cost of the double dialyzer configuration. Therefore, our estimation of additional cost of using two dialyzers in parallel of $2100 per patient per year is probably excessive since the number of re-uses is limited to ten in our study and the additional set up time may not lead to increased staff costs in actuality.

          In this study of stable hemodialysis patients with a large Vdurea, the doubling of the surface area of a biocompatible artificial kidney by using two dialyzers in a parallel configuration did not cause hemodynamic or hematologic adverse effects and increased eKt/Vurea and URR by an average of 22% and 9% respectively. The increase in eKt/V is sustainable in patients with vascular accesses that reliably deliver 400 ml/min of blood flow and improvement in patient perceived health status can be expected.


          The authors wish to thank C.V. Cook for secretarial assistance.


1. Carroll CE, Port FK, Daugirdas J, Agodoa LYC, Jones CA, Held PJ: Patient factors associated with delivered KT/V in the U.S. (abstract) J Am Soc Nephrol 6:594, 1995

2. HCFA. 1997 Annual Report ESRD Core Indicators Project. HCFA,DHHS, Baltimore, Maryland, December, 1997

3. HCFA. 1995 Core Indicators Project: Opportunities to Improve Care for Adult In-Center Hemodialysis Patients. HCFA,DHHS, Baltimore, Maryland, January, 1996

4. Ifudu O, Mayers JD, Matthew JJ, Fowler AM, Homel P, Friedman EA: Standardized hemodialysis prescriptions promote inadequate treatment in patients with large body mass. Ann Intern Med 128:451-454, 1998

5. Hakim RM, Held PJ, Stannard DC, Wolfe RA, Port FK, Daugirdas JT, Agodoa L: Effect of the dialysis membrane on mortality of chronic hemodialysis patients. Kidney Int 50:566-570, 1996

6. Levin NW, Parnell SL, Prince HN, Gotch F, Polaschegg HD, Levin R, Alto A, Kaufman AM: The use of heated citric acid for dialyzer reprocessing. J Am Soc Nephrol 6:1578-1585, 1995

7. Schoenfeld P, McLaughlin MD, Mendelson M: Heat disinfection of polysulfone hemodialyzers. Kidney Int 47:638-642, 1995

8. Ikizler TA, Schulman G: Adequacy of dialysis. Kidney Int 52(suppl 62):S96-S100, 1997

9. Priester-Coary A, Daugirdas JT: A recommended technique for obtaining the post-dialysis BUN. Semin Dial 10:23-25, 1997.

10. Daugidas JT: Second generation logarithmic estimates of single-pool variable volume of Kt/V: an analysis of error. J Am Soc Nephrol 4:1205-1213, 1993

11. Daugirdas JT, Schneditz D: Overestimation of hemodialysis dose depends on dialysis efficiency by regional blood flow but not by conventional two pool urea kinetic analysis. ASIO J 41:m719-m724, 1995

12. Owen WF, Lew NL, Liu Y, Lowrie EG, Lazarus JM: The urea reduction ratio and serum albumin concentration as predictor of mortality in patients undergoing hemodialysis. N Eng J Med 329:1001-1006, 1993

13. Sacamay TET, Bolton WK: Use of iohexol to quantify hemodialysis delivered and residual renal function. Kidney Int 54:986-991, 1998

14. Jacobsson L: A method for the calculation of renal clearance based on a single plasma sample. Clin Physiol 3:297-305, 1983

15. Brown SCW, O'Reilly PH: Iohexol clearance for the determination of glomerular filtration rate in clinical practice: evidence for a new gold standard. J Urol 146:675-679, 1991

16. Albert SM: Assessing health-related quality of life in chronic care populations. Journal of Mental Health and Aging 3:101-118, 1997.

17. Bloembergen WE, Stannard DC, Port FK, Wolfe RA, Pugh JA, Jones CA, Greer JW, Golper TA, Held PJ: Relationship of dose of hemodialysis and cause-specific mortality. Kidney Int 50:557-565, 1996

18. Gotch FA, Levin NW, Port FK, Wolfe RA, Uehlinger DE: Clinical outcome relative to the dose of dialysis is not what you think: the fallacy of the mean. Am J of Kidney Dis 30:1-15, 1997

19.  Postan S, Bailey J: Dialysis therapy. New Engl J Med 338(20):1428-1437, 1998

20.  Leypoldt JK, Cheung AK, Agodoa LY, Daugirdas JT, Greene T, Keshaviah PR: Hemodialyzer mass transfer-area coefficients for urea increase at high dialysate flow rates. Kidney Int 51:2013-2017, 1997

21.  NKF-DOQI Clinical Practice Guidelines for Hemodialysis Adequacy. New York, National Kidney Foundation, pp 25-40, 1997

22.  Blake PG: Practical guide to measuring adequacy of dialysis. Advances in Renal Replacement Therapy 6:80-84, 1999.

23.  Bosch JP, Mishkih GJ, Barlee V, von Albertini B: Hemodialysis high efficiency treatments: six year outcomes. (abstract) J Am Soc Nephrol 6:484, 1995

24. von Albertini B, Miller JH, Bosch JP, Gardner P, Shinaberger JH: Serial and parallel pairs of dialyzers for high efficiency treatment. (abstract) Kidney Int 31(1):247, 1987

25. Velasquez MT, von Albertini B, Lew SQ, Mishkin GJ, Bosch JP: Equal levels of blood pressure control in ESRD patients receiving high-efficiency hemodialysis and conventional hemodialysis. American Journal of Kidney Diseases 31:618-623, 1998

26. Sadler, JH: Health promotion for the end-stage renal disease patients. Advances in Renal Replacement Therapy 5:275-285, 1998


Figure 1. Schematic diagram of the two dialyzers in parallel configuration. Abbreviations: BFR = blood flow rate, DFR = dialysate flow rate.

Figure 2. Comparison of mean eKT/Vurea by patient for the control (HDC) and intervention (HDP) periods. The last entry represents the average of all measurements during HDC and HDP. Error bars = SEM. * = p value < 0.0001; † = p value < 0.02; ‡ = p value < 0.05.

Figure 1.

Figure 2.



Abbreviation: GN-glomerulonephritis


          Abbreviations: URR-urea reduction ratio; eKt/Vurea-equilibrated fractional clearance index of urea; spKt/Vurea-single pool calculation of fractional clearance index of urea; Kt/VIo-dialysis iohexol clearance normalized for the volume of distribution of iohexol; IoRR, iohexhol reduction ratio.


Abbreviation: EPO – Epogen; HDC – control HD; HDP – parallel HD; * p<0.0001; † p<0.02; ‡ p<0.05
SEM omitted to conserve space


Score = SF-36 transformed scale 0-100
D= change in score from HDC
*= difference of 2 or more considered significant

** Note : The bold text under "Methods"  was edited into this study by Molded Products, the bold text information, 
was not originally included in the University Of Virginia Study on "Improved Urea Reduction Ratio and KT/V in 
Large Hemodialysis Patients Using Two Dialyzers in Parallel