Background

Daily short hemodialysis (HD) is often prescribed by simply doubling treatment frequency and halving treatment time; however, the effect of this prescription approach on the equilibrated HD dose (urea eKt/V) and whole body clearance for ₂-microglobulin has not been established.

Methods

We compared urea and ₂-microglobulin kinetics during and 60 minutes after a short HD treatment and a conventional HD treatment in a crossover study on 22 maintenance HD patients: 16 male and 6 female, 61 18 (mean standard deviation) years of age. One patient in each treatment modality was excluded from certain analyses because of missing data. Short and conventional HD treatments were essentially identical, except for treatment times, which were 116 14 and 241 27 minutes, respectively. Blood samples were collected at regular intervals during and after treatments, and additional blood and dialysate samples were collected at 60 minutes of treatment to evaluate dialyzer clearances.

Results

Plasma water urea clearances measured directly across the dialyzer during short and conventional HD treatments were not different (255 23 mL/min and 255 28 mL/min, respectively). The 60-minute postdialysis blood urea nitrogen concentration rebounded more (P < 0.01) after short HD than conventional HD (5.9 3.1 vs. 4.0 1.5 mg/dL, respectively). Calculated urea eKt/V values using the Daugirdas-Schneditz rate equation were not different from those measured during conventional HD using the 60-minute postdialysis concentration but significantly overestimated measured urea eKt/V values during short HD. Postdialysis rebound of ₂-microglobulin concentrations was variable but similar after short and conventional HD treatments (0.1 3.4 vs. 0.7 1.8 mg/L, respectively). Whole body clearances of ₂-microglobulin calculated from predialysis and immediate (10-second) postdialysis serum concentrations during short and conventional HD treatments were not different from each other (42.9 24.1 vs. 41.9 22.4 mL/min, respectively).

Conclusion

These observations show that the Daugirdas-Schneditz rate equation is accurate in predicting urea eKt/V during conventional, but not during short, HD. In contrast, whole body clearances of ₂-microglobulin during short and conventional HD treatments were similar. We conclude that calculation of accurate estimates of urea eKt/V, but not clearances of ₂-microglobulin, differ during short and conventional HD treatments.

METHODS

Patients

Twenty-two chronic hemodialysis patients treated three times per week by maintenance HD therapy were recruited from two separate dialysis units within the University of Utah Dialysis Program. Patients were excluded if they were medically unstable, had a hematocrit less than 28% from the most recent monthly determination, were hepatitis B–positive, hepatitis C–positive, or HIV-positive, were prisoners, were pregnant women, were minors below 18 years of age, or were mentally disabled. All study patients were informed of the purpose of the study and gave written, informed consent. The protocol for this study was approved by the Institutional Review Board at the University of Utah.

The age of the patients was 61 18 (standard deviation) years. Sixteen of the patients were male and six were female; 21 patients were Caucasian and one was Asian. Fourteen patients had native fistulas, six had synthetic grafts, and two used catheters for their vascular access.

Study design

This was a crossover study in which each patient was studied on two separate occasions (one week apart) during the patient's routine HD treatment schedule. During one study treatment, HD was performed according to the patients' routine prescription, with blood samples taken during and for one hour after this treatment. During the other study treatment, HD was performed for one half of the prescribed treatment time, after which the dialysis procedure was stopped for one hour while blood flow was continued at a slow rate (120 mL/min) through the extracorporeal circuit without dialysate flow. This latter study treatment simulated a short HD treatment followed by a one-hour postdialytic period. The remainder of the prescribed HD treatment was then completed after this one-hour pause period. The order of the study treatments was not randomly assigned because of practical concerns within the dialysis units. The first study treatment was conventional HD for seven patients and short HD for 15 patients.

Clinical study procedures

During the study treatment using the routine prescription, a blood sample (4 mL) was taken predialysis from the vascular access and at 30, 60, 120, 180 minutes from the arterial blood tubing of the extracorporeal dialysis circuit. At the end of the treatment, the blood flow rate was decreased to 120 mL/min and blood samples were taken after 10 seconds, 2, 10, 30, and 60 minutes. After 60 minutes of treatment, additional samples were taken from the venous blood tubing and the dialysate outflow to permit direct evaluation of dialyzer clearances. This protocol permitted evaluation of solute kinetics during and after a conventional HD treatment.

During the other study treatment, a blood sample (4 mL) was taken predialysis from the vascular access and at 30, 60, and 90 minutes from the arterial blood tubing. At the midpoint of the treatment (i.e., half of the usual treatment time), the ultrafiltration rate was adjusted to the minimum setting and the dialysate flow was stopped. The blood flow rate was decreased to 120 mL/min for one hour, and blood samples were taken after 10 seconds, 2, 10, 30, and 60 minutes. After this one-hour pause period, HD was restarted and continued to complete the usual treatment. Sixty minutes into this alternative treatment, additional samples were taken from the venous blood tubing and the dialysate outflow to permit the direct evaluation of dialyzer clearances. This protocol permitted the study of solute kinetics and postdialysis solute rebound for one hour after a short treatment without significantly interfering with the patient's scheduled HD treatment.

After the conventional HD treatment, a predialysis blood sample was taken at the next scheduled HD treatment (two days later) to evaluate the interdialytic urea generation rate. High flux dialyzers (Optiflux F160NR in 10 patients, and Optiflux F200NR in 12 patients; Fresenius North America, Ogden, UT, USA) were used at routinely prescribed blood, dialysate, and ultrafiltration flow rates. The actual blood flow rate was measured using the HD01 monitor (Transonic Systems, Ithaca, NY, USA) at 60 minutes during each study treatment.

Analytical assays

All blood and dialysate samples were collected in tubes without additional anticoagulant and allowed to stand at room temperature for 30 to 60 minutes. The samples were then centrifuged and the serum collected and stored at -70°C until they were assayed. All samples were assayed for the concentrations of urea and ₂-microglobulin. Urea nitrogen was measured using an automated analyzer (Beckman CX7; Beckman Coulter, Fullerton, CA, USA), and serum concentrations of ₂-microglobulin were measured using a radioimmunoassay (Immunos, Albany, CA, USA). Dialysate concentrations of ₂-microglobulin were frequently below 1 mg/L, and were considered to be too low for accurate determinations of ₂-microglobulin clearance (see below). Although the measured urea nitrogen concentration is that in serum, by convention, it will be referred to as the blood urea nitrogen (BUN) concentration. To assess plasma volume depletion during the study treatments, predialysis and immediate (10-second) postdialysis samples were also assayed for serum albumin concentrations using an automated analyzer (Beckman Coulter).

Data analysis

Blood-side and dialysate-side clearances for urea during conventional and short HD treatments were calculated using standard formulas and compared using paired Student t test. Blood-side blood water clearances were corrected by assuming that 89.4% of blood flow consists of blood water¹⁴. Postdialysis rebound was calculated both in concentration units and as a fraction of the immediate postdialysis serum concentration (10 seconds after slowing the blood flow rate to 120 mL/min); the extent of postdialysis rebound for short and conventional HD treatments was compared using repeated measures analysis of variance (ANOVA).

Single-pool urea Kt/V (spKt/V) values were calculated from the immediate (10-second) postdialysis concentration for both short and conventional HD treatments using the second generation Daugirdas formula¹⁵ or

(1)

where R denotes the ratio of the postdialysis to predialysis BUN concentration, T is treatment time in hours, UF is the volume of fluid removed during treatment in L, and BW is the postdialysis body weight in kg.

Equilibrated Kt/V (eKt/V) values were calculated using the rate equations described by Daugirdas and Schneditz¹⁰

(2)

and that recently derived from observations during the Hemodialysis (HEMO) Study^11,12

(3)

An alternative to these rate equations is that described by Tattersall et al¹⁶, or

(4)

(here, t indicates treatment time in minutes), and this equation was also used to calculate eKt/V. Equilibrated Kt/V values were also calculated from equation 1 using the 60-minute postdialysis concentrations (after correction for urea generation¹⁷); these values are referred to as the measured eKt/V values. The urea distribution volume was calculated from measured eKt/V, directly measured dialyzer clearance of urea, and treatment time. All calculated eKt/V values were compared with those measured using a paired Student t test.

Whole body clearance of ₂-microglobulin was calculated using the formula described previously¹³, assuming the distribution volume for ₂-microglobulin was one third of that for urea. Because the dialysate concentrations of ₂-microglobulin were low, and because ₂-microglobulin can adsorb to the membrane, the evaluation of its clearance directly across the dialyzer was only calculated using the blood-side, and not the dialysate-side, measurements. When evaluating this clearance for ₂-microglobulin, the arterial plasma flow rate was calculated as the measured blood flow rate times one minus the hematocrit. The venous plasma flow rate was calculated as the arterial plasma flow rate minus the ultrafiltration rate. A standard formula for calculating blood-side clearance directly across the dialyzer from arterial and venous serum ₂-microglobulin concentrations and plasma flow rates was used.

Fractional plasma volume depletion (PVD) as the result of HD treatment was calculated assuming no loss of albumin from the plasma space during the treatment as

(5)

where PV and A denote plasma volume and serum albumin concentration, and the subscript i and f denote predialysis and immediate (10-second) postdialysis values, respectively.

All empirical and calculated values are reported as mean standard deviation. All regression coefficients are reported as mean standard error.

RESULTS

Kinetics of urea

Table 1 shows treatment parameters for all 22 patients studied. By design, no differences were observed during short and conventional HD except treatment time and fluid removed because only half of the prescribed fluid loss was removed during the short HD treatment. Urea transfer characteristics for both Optiflux F160NR and F200NR dialyzers measured from cross dialyzer measurements are also reported in Table 1. No differences in dialyzer urea transfer characteristics were apparent during short and conventional HD treatments.

Table 1 - Treatment and dialyzer clearance characteristics during short and conventional HD.

Full table

Figure 1 shows the time-dependent changes in BUN concentrations during and 60 minutes after both short and conventional HD treatments. The decrease in BUN concentrations during both treatments was similar; however, the magnitude of the rebound in BUN, when measured in absolute concentrations was greater (P < 0.01) after short HD than after conventional HD treatments (5.9 3.1 vs. 4.0 1.5 mg/dL, respectively, see Figure 2. When expressed as a percentage of the immediate (10-second) postdialysis values, however, the rebound after 60 minutes was actually lower for short (21.7 7.6%) than for conventional (27.5 8.5%) HD treatments.

Figure 1.

Time dependence of blood urea nitrogen (BUN) concentrations during and 60 minutes after short (triangles, dashed lines) and conventional (squares, solid lines) hemodialysis (HD) treatments. The final times of short and conventional HD treatments were assumed to be 120 and 240 minutes, respectively. Mean values from 21 treatment sessions for both short and conventional HD are shown.

Full figure and legend (10K)

Figure 2.

Postdialysis rebound of blood urea nitrogen (BUN) concentrations calculated as the increase in concentration (in absolute magnitude) from its value 10 seconds after completion of conventional (squares, solid lines) and short (triangles, dashed lines) hemodialysis (HD) treatments. BUN concentrations were not corrected for interdialytic generation of urea. Mean values from 21 treatment sessions for both conventional and short HD are shown.

Full figure and legend (8K)

Table 2 compares estimates of urea kinetic parameters evaluated during short and conventional HD treatments. Calculated values of spKt/V for short HD were approximately one half of those for conventional HD treatments. Calculated values of eKt/V estimated using the Daugirdas-Schneditz rate equation (0.60 0.07 and 1.38 0.14, respectively) were approximately 8% less (P < 0.001) than those measured during short HD treatments (0.66 0.08), but were not different from those directly measured (1.37 0.13) during conventional HD treatments. In contrast, calculated values of eKt/V using the HEMO Study rate equation (0.67 0.08 and 1.43 0.15, respectively) were not significantly different from those directly measured during short HD treatments, but was approximately 5% greater (P < 0.001) than those directly measured during conventional HD treatments. Also shown in Table 2 are calculated values of eKt/V predicted using the equation suggested by Tattersall et al¹⁶. The predicted eKt/V values using this equation (0.64 0.08 and 1.38 0.15, respectively) were not different from those directly measured during both short (P = 0.16) and conventional (P = 0.20) HD treatments.

Table 2 - Urea kinetic parameters during short and conventional HD treatments.

Full table

The differences between measured and calculated eKt/V values are plotted versus treatment time in Figure 3. The slope of the regression of these data points was negative when predicting eKt/V using the Daugirdas-Schneditz rate equation (P < 0.001, Figure 3a and the HEMO Study rate equation (P = 0.01, Figure 3b. The slope was also negative in magnitude when predicting using the Tattersall et al equation, but this value was not statistically different from zero (P = 0.13).

Figure 3.

The difference between the equilibrated Kt/V (eKt/V) directly measured and that calculated by four separate rate equations plotted versus treatment time. Data from 42 treatment sessions are shown; those from patients treated with catheters are shown as open symbols. (A) Comparison with that calculated by the Daugirdas-Schneditz rate equation: the best fit linear regression equation was Y =-4.83 (1.30) 10^-4 X + 0.108 (0.025), P < 0.001. (B) Comparison with that calculated by the HEMO Study rate equation: the best fit linear regression equation was Y =-3.61 (1.40) 10^-4 X + 0.027 (0.027), P = 0.01. (C) Comparison with that calculated by the Tattersal et al equation: the best fit linear regression equation was Y =-2.04 (1.33) 10^-4 X + 0.035 (0.025), P = 0.13. (D) Comparison with that calculated by the rate equation derived from the data in this study (equation 5): the best fit linear regression equation was Y =-0.23 (1.24) 10^-4 X – 0.004 (0.024), P = 0.85.

Full figure and legend (10K)

Multiple linear regression analysis showed that both the rate of HD treatment (spKt/V/T or spK/V) and the dose (spKt/V) are significant predictors of the measured eKt/V. Thus, based on the current data, valid estimates of eKt/V depend on both the rate and dose of treatment; the equation that best predicts eKt/V from these factors was the following:

(6)

The difference between measured eKt/V values and those predicted using equation 6 are shown in Figure 3d versus treatment time for both short and conventional HD treatments combined. The slope of this equation is not significantly different from zero (P = 0.87). Figure 4 plots values of eKt/V predicted using equation 6 versus those measured. The high precision of the predicted eKt/V values in equation 6 is largely because it is a three-parameter equation, whereas equations 2 to 4 only contain one or two adjustable parameters.

Figure 4.

Direct comparison of eKt/V calculated using the equation derived in this study, equation 6, with measured values of eKt/V(N = 21).

Full figure and legend (9K)

Kinetics of ₂-microglobulin

Figure 5 shows the time dependence of serum ₂-microglobulin concentrations during and 60 minutes after both short and conventional HD treatments. The decrease in serum ₂-microglobulin concentration during both treatments was similar during the first two hours of treatment. Postdialysis rebound for serum ₂-microglobulin concentration 60 minutes after treatment was variable; the average magnitude of the rebound was less then 1.0 mg/L for both short and conventional HD treatments (0.1 3.4 vs. 0.7 1.8 mg/L, respectively). When expressed as percentages, postdialysis rebound of serum ₂-microglobulin concentrations was 3.5 23.0% and 6.5 13.6% for short and conventional HD treatments, respectively.

Figure 5.

Time dependence of ₂-microglobulin (Beta-2-M) concentrations during and 60 minutes after short (triangles, dashed lines) and conventional (squares, solid lines) hemodialysis (HD) treatments. The final times of short and conventional HD treatments were assumed as 120 and 240 minutes, respectively. Mean values from 21 treatment sessions for both short and conventional HD are shown.

Full figure and legend (9K)

Calculated values of whole body ₂-microglobulin clearance from predialysis and postdialysis (10-second) concentrations^11,13 and those measured directly across the dialyzer are shown in Table 3. Whole body ₂-microglobulin clearances yielded values of 42.9 24.1 and 41.9 22.4 mL/min for short HD and conventional HD treatments, respectively. When both short and conventional HD treatments were combined for Optiflux F160NR and Optiflux F200NR dialyzers, the calculated whole body ₂-microglobulin clearances were 42.9 24.6 and 41.9 21.9 mL/min, respectively. None of the differences in any calculated whole body ₂-microglobulin clearances were statistically significant. Calculated clearances of ₂-microglobulin across the dialyzer were approximately half of those calculated from predialysis and postdialysis concentrations, but were similar for both short and conventional HD treatments (Table 3).

Table 3 - Clearances of ₂-microglobulin during short and conventional HD treatments calculated from predialysis and postdialysis concentrations (whole body) or directly across the dialyzer.

Full table

Predialysis and immediate (10-second) postdialysis serum albumin concentrations were 3.5 0.3 and 3.4 0.4 g/dL for short HD and 3.4 0.4 and 3.5 0.5 g/dL for conventional HD treatments. Calculated plasma volume depletion was -2.2% 7.1% for short HD and 1.9% 8.7% for conventional HD treatments, respectively. These data suggest that plasma volume depletion was small and quite variable in these studies.

DISCUSSION

A more accurate measure of the true dose of hemodialysis than spKt/V is the parameter urea eKt/V, which can be most accurately determined using the equilibrated BUN concentration obtained at 60 minutes' postdialysis⁹. Assessment of conventional hemodialysis dose using the concept of urea eKt/V can be performed using an immediate (10- or 20-second) postdialysis blood sample only if it is corrected for postdialysis rebound of urea. Several equations have been previously proposed to correct for postdialysis urea rebound^{10,11,12,16,18,19}, and some of these formulas have been shown to be approximately equivalent to each other during conventional HD treatments²⁰. The formulas by both Daugirdas-Schneditz and Tattersall et al have been shown to accurately correct for postdialysis rebound of urea during conventional HD treatments^17,19,21,22; however, the adaptability of these equations to short HD treatments has not been studied extensively.

The results from the current study show that neither the Daugirdas-Schneditz nor the HEMO Study rate equation is accurate in predicting postdialysis urea rebound after both short and conventional HD treatments. As in several previous studies^17,19,21,22, the current observations show that the Daugirdas-Schneditz rate equation is accurate in predicting postdialysis urea rebound after conventional HD treatments. The results from this study show, however, that this equation is not accurate in predicting postdialysis urea rebound after short HD treatments. Paradoxically, the current observations suggest that the HEMO Study rate equation cannot be used to correct for postdialysis urea rebound during conventional HD treatments. One possible reason for this discrepancy is that the immediate postdialysis sample in the current study was collected after 10 seconds (using a blood flow rate of 120 mL/min) instead of the 20 seconds preferred in the HEMO Study (using a blood flow rate of approximately 80 mL/min). We chose to sample postdialysis blood as proposed by Kapoian et al²³ because of concerns of potential thrombosis in the extracorporeal circuit during our unique study protocol. It is possible that sampling blood during this shorter time period after ending the HD treatments more accurately assessed postdialysis urea rebound due to access recirculation, uncontaminated by that caused by cardiopulmonary recirculation.

Our results further suggest that the formula proposed by Tattersall et al¹⁶ may be more useful for predicting postdialysis urea rebound during short HD treatments because its predictions of urea eKt/V were not statistically different from those values directly measured Figure 3c. It should be noted, however, that the accuracy of the Tattersall et al equation in predicting the measured eKt/V was not considerably greater than those using the other equations. Further studies using large numbers of HD patients will be necessary to determine the accuracy of the Tattersall et al equation for predicting postdialysis urea rebound during both short and conventional HD treatments. The prediction of postdialysis urea rebound and eKt/V is more precise if equation 6 is used instead.

It should be noted that two patients in this study had a catheter as their vascular access. The kinetics of postdialysis urea rebound for these patients may be different from those with an arteriovenous vascular access¹⁰. In this study we combined all patients for analysis because it was not obvious that the patients with central accesses were substantially different from the other patients (see Figure 3. Furthermore, reanalysis of these data excluding those patients did not alter the conclusions from these data analyses.

Based on the results summarized in Table 2, we can estimate the sum total of urea eKt/V delivered per week during short and conventional HD as 3.96 (6 0.66/treatment) and 4.11 (3 1.37/treatment), respectively, when short HD treatments are prescribed by doubling treatment frequency and halving treatment time per session. This calculation indicates a small decrease (approximately 4%) in weekly eKt/V per week during daily short HD compared with that for conventional HD therapy. Using standard urea Kt/V as described by Gotch^24,25 as the dose measure, however, the hemodialysis dose during daily short HD will considerably exceed that during conventional HD. Whether the adequate treatment dose is better evaluated using the weekly standard urea Kt/V than weekly eKt/V will require data on clinical outcomes using these parameters as guides of therapy.

In contrast to urea, clearances of ₂-microglobulin (calculated either using predialysis and postdialysis serum concentrations or directly across the dialyzer) during short and conventional HD treatments were indistinguishable. Such agreement may be partly due to the lack of significant postdialysis rebound of ₂-microglobulin in this study compared with that previously reported²⁶. The magnitude of postdialysis rebound reported in the current study was closer to that reported in preliminary data from the HEMO Study of approximately 4%[abstract; Leypoldt et al, J Am Soc Nephrol 9:299A, 1998]. The lack of postdialysis rebound of ₂-microglobulin in this study may have been due to substantial plasma volume depletion during these treatments; however, the method used for calculating this parameter from changes in serum albumin concentrations may not have been accurate enough to demonstrate this. The current results suggest that single compartment models of ₂-microglobulin kinetics may have only limited usefulness for evaluating removal of middle molecules during daily short HD therapy.

CONCLUSION

The results from this study demonstrate that calculation of accurate estimates of urea eKt/V, but not of clearances of ₂-microglobulin, differ during short and conventional HD treatments. It is important to note that these studies were performed during only a single isolated short treatment, and may not apply to patients treated regularly using short HD treatments. Further studies in the latter patients are necessary to validate these findings.

References

1. TRAEGER, J, GALLAND, R, DELAWARI, E, et al: Time needed to improve clinical parameters by daily hemodialysis. Home Hemodial Int 1999 3: 29–32,
2. GALLAND, R, TRAEGER, J, DELAWARI, E, et al: Daily hemodialysis versus standard hemodialysis: TAC, TAD, Weekly eKt/V, std(Kt/V), and PCRn. Home Hemodial Int 1999 3: 33–36,
3. TING, G, CARRIE, B, FREITAS, T, et al: Global ESRD costs associated with a short daily hemodialysis program in the United States. Home Hemodial Int 1999 3: 41–44,
4. TING, G: The strategic role of daily hemodialysis in managed care in the United States. Semin Dial 2000 13: 385–388,
5. GALLAND, R, TRAEGER, J, ARKOUCHE, W, et al: Short daily hemodialysis rapidly improves nutritional status in hemodialysis patients. Kidney Int 2002 60: 1555–1560,
6. KOOISTRA, MP, VOS, J, KOOMANS, HA, et al: Daily home haemodialysis in The Netherlands: Effects on metabolic control, haemodynamics, and quality of life. Nephrol Dial Transplant 1998 13: 2853–2860,
7. PICCOLI, GB, CALDERINI, M, BECHIS, F, et al: Daily dialysis and flexible schedules: How to assess Kt/V and EKRc? Hemodial Int 2001 5: 13–18,
8. SURI, R, DEPNER, TA, BLAKE, PG, et al: Adequacy of quotidian hemodialysis. Am J Kidney Dis 2003 42(Suppl 1):S42–S48,
9. DEPNER, T, BECK, G, DAUGIRDAS, J, et al: Lessons from the Hemodialysis (HEMO) Study: An improved measure of the actual hemodialysis dose. Am J Kidney Dis 1999 33: 142–149,
10. 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. ASAIO J 1995 41: M719–M724,
11. EKNOYAN, G, BECK, GJ, CHEUNG, AK, et al: Effect of dialysis dose and membrane flux in maintenance hemodialysis. N Engl J Med 2002 347: 2010–2019,
12. DAUGIRDAS, JT, GREENE, T, DEPNER, TA, et al: Factors that affect postdialysis rebound in serum urea concentration, including the rate of dialysis: Results from the HEMO Study. J Am Soc Nephrol 2004 15: 194–203,
13. LEYPOLDT, JK, CHEUNG, AK, DEETER, RB: Single compartment models for evaluating beta 2-microglobulin clearance during hemodialysis. ASAIO J 1997 43: 904–909,
14. DAUGIRDAS, JT, VAN STONE, JC: Physiologic principles and urea kinetic modeling, in Handbook of Dialysis, 2001 edited by Daugirdas JT, Blake PG, Ing TS, Philadelphia, Lippincott Williams & Wilkins, pp 15–45
15. DAUGIRDAS, JT: Second generation logarithmic estimates of single-pool variable volume Kt/V: An analysis of error. J Am Soc Nephrol 1993 4: 1205–1213,
16. TATTERSALL, JE, DETAKATS, D, CHAMNEY, P, et al: The post-hemodialysis rebound: Predicting and quantifying its effect on Kt/V. Kidney Int 1996 50: 2094–2102,
17. BANKHEAD, MM, TOTO, RD, STAR, RA: Accuracy of urea removal estimated by kinetic models. Kidney Int 1995 48: 785–793,
18. SMYE, SW, EVANS, JH, WILL, E, et al: Paediatric haemodialysis: Estimation of treatment efficiency in the presence of urea rebound. Clin Phys Physiol Meas 1992 13: 51–62,
19. MADUELL, F, GARCIA-VALDECASAS, J, GARCIA, H, et al: Validation of different methods to calculate Kt/V considering postdialysis rebound. Nephrol Dial Transplant 1997 12: 1928–1933,
20. SMYE, SW, TATTERSALL, JE, WILL, EJ: Modeling the postdialysis rebound: The reconciliation of current formulas. ASAIO J 1999 45: 562–567,
21. HEMO STUDY GROUP, DAUGIRDAS, JT, DEPNER, TA, et al: Comparison of methods to predict equilibrated Kt/V in the HEMO Pilot Study. Kidney Int 1997 52: 1395–1405,
22. DEPNER, TA, GREENE, T, GOTCH, FA, et al: Imprecision of the hemodialysis dose when measured directly from urea removal. Kidney Int 1999 55: 635–647,
23. KAPOIAN, T, STEWARD, CA, SHERMAN, RA: Validation of a revised slow-stop flow recirculation method. Kidney Int 1997 52: 839–842,
24. GOTCH, FA: Modeling the dose of home dialysis. Home Hemodial Int 1999 3: 37–40,
25. GOTCH, FA: Normalization of dialysis dose for daily dialysis. Hemodial Int 2001 5: 102–105,
26. LEYPOLDT, JK, CHEUNG, AK, DEETER, RB: Rebound kinetics of beta2-microglobulin after hemodialysis. Kidney Int 1999 56: 1571–1577,

Acknowledgments

The technical expertise of Janice F. Gilson and Craig D. Kamerath is gratefully acknowledged. Fresenius Medical Care-North America kindly donated the dialyzers for this study. This work was presented at the American Society of Nephrology Meeting, October 30 to November 4, 2002, in Philadelphia, PA, and has been previously published in abstract form (abstract; Leypoldt et al, J Am Soc Nephrol 13:413A, 2002). This work was supported by the National Institute of Diabetes, Digestive and Kidney Diseases.