The New England Journal of Medicine
e-mail icon  FREE NEJM E-TOC    HOME   |   SUBSCRIBE   |   CURRENT ISSUE   |   PAST ISSUES   |   COLLECTIONS   |    Advanced Search
Sign in | Get NEJM's E-Mail Table of Contents — Free | Subscribe
 
Original Article
PreviousPrevious
Volume 347:2010-2019 December 19, 2002 Number 25
NextNext

Effect of Dialysis Dose and Membrane Flux in Maintenance Hemodialysis
Garabed Eknoyan, M.D., Gerald J. Beck, Ph.D., Alfred K. Cheung, M.D., John T. Daugirdas, M.D., Tom Greene, Ph.D., John W. Kusek, Ph.D., Michael Allon, M.D., James Bailey, M.D., James A. Delmez, M.D., Thomas A. Depner, M.D., Johanna T. Dwyer, D.Sc., R.D., Andrew S. Levey, M.D., Nathan W. Levin, M.D., Edgar Milford, M.D., Daniel B. Ornt, M.D., Michael V. Rocco, M.D., Gerald Schulman, M.D., Steve J. Schwab, M.D., Brendan P. Teehan, M.D., Robert Toto, M.D., for the Hemodialysis (HEMO) Study Group

 

This Article
-Abstract
- PDF
-PDA Full Text
-PowerPoint Slide Set

Commentary
-Editorial
by Himmelfarb, J.
-Letters

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited

More Information
-PubMed Citation
ABSTRACT

Background The effects of the dose of dialysis and the level of flux of the dialyzer membrane on mortality and morbidity among patients undergoing maintenance hemodialysis are uncertain.

Methods We undertook a randomized clinical trial in 1846 patients undergoing thrice-weekly dialysis, using a two-by-two factorial design to assign patients randomly to a standard or high dose of dialysis and to a low-flux or high-flux dialyzer.

Results In the standard-dose group, the mean (±SD) urea-reduction ratio was 66.3±2.5 percent, the single-pool Kt/V was 1.32±0.09, and the equilibrated Kt/V was 1.16±0.08; in the high-dose group, the values were 75.2±2.5 percent, 1.71±0.11, and 1.53±0.09, respectively. Flux, estimated on the basis of beta2-microglobulin clearance, was 3±7 ml per minute in the low-flux group and 34±11 ml per minute in the high-flux group. The primary outcome, death from any cause, was not significantly influenced by the dose or flux assignment: the relative risk of death in the high-dose group as compared with the standard-dose group was 0.96 (95 percent confidence interval, 0.84 to 1.10; P=0.53), and the relative risk of death in the high-flux group as compared with the low-flux group was 0.92 (95 percent confidence interval, 0.81 to 1.05; P=0.23). The main secondary outcomes (first hospitalization for cardiac causes or death from any cause, first hospitalization for infection or death from any cause, first 15 percent decrease in the serum albumin level or death from any cause, and all hospitalizations not related to vascular access) also did not differ significantly between either the dose groups or the flux groups. Possible benefits of the dose or flux interventions were suggested in two of seven prespecified subgroups of patients.

Conclusions Patients undergoing hemodialysis thrice weekly appear to have no major benefit from a higher dialysis dose than that recommended by current U.S. guidelines or from the use of a high-flux membrane.

Two treatment-related factors implicated in the substantial mortality and morbidity among patients undergoing maintenance hemodialysis1 are the dose of dialysis delivered and the size of molecules removed. An index of the dialysis dose is the fractional clearance of urea (molecular mass, 60 D) which is commonly expressed as the intradialytic urea-reduction ratio2 or as Kt/V, where K represents the rate of urea clearance by the dialyzer in milliliters per minute, t the duration in minutes of the treatment session, and V the volume of distribution of urea in the patient in milliliters.3 Current guidelines in the United States target a urea-reduction ratio of at least 65 percent or a single-pool Kt/V of at least 1.20.4 The urea-reduction ratio and single-pool Kt/V overestimate the delivered dose of dialysis, because they fail to account for blood urea rebound after dialysis. A more accurate measure of the dialysis dose, the equilibrated Kt/V, corrects for urea rebound and is usually 0.15 to 0.20 lower than the single-pool Kt/V5; by inference, current U.S. guidelines recommend a minimal equilibrated Kt/V of 1.00 to 1.05.

The National Cooperative Dialysis Study was to our knowledge the only randomized trial that evaluated the effect of the dose of dialysis on clinical outcomes.6 It established a beneficial effect of an increased dialysis dose on morbidity at dialysis doses well below the current standard and in patients with substantially fewer coexisting conditions than are found in the present population of patients undergoing hemodialysis. Some subsequent observational studies have reported continuing improvement in morbidity and mortality at dialysis doses well above those recommended in the current guidelines,7,8,9 whereas other studies have not found such improvement.10 Observational studies have also suggested that membranes with high porosity or flux, which clear larger solutes such as beta2-microglobulin (molecular mass, 11,900 D), are associated with improved outcomes.11,12 To our knowledge, no randomized studies of the effects of membrane flux on clinical outcomes have been performed. Our study, the Hemodialysis (HEMO) Study, was a randomized clinical trial designed to determine whether increasing the dose of dialysis or using a high-flux dialyzer membrane alters survival or morbidity among patients undergoing hemodialysis.

Methods

Study Design

The design of our study has been described previously.13,14 The study was approved by the institutional review board at each of 15 clinical centers associated with 72 participating dialysis units, and all patients gave written informed consent.

Eligible patients were randomly assigned in a 1:1 ratio with a two-by-two factorial design to either a standard-dose or a high-dose goal and to dialysis with either a low-flux or a high-flux dialyzer. Randomization was performed centrally with the use of random permuted blocks, with stratification according to clinical center, age, and diabetic status. The dose intervention was defined by the Kt/V. The single-pool Kt/V was determined by two-point urea modeling on the basis of the intradialytic reduction in blood urea and the intradialytic weight loss15 and was corrected to equilibrated Kt/V on the basis of the rate of dialysis (K/V).5,16 The standard-dose goal was an equilibrated Kt/V of 1.05, equivalent to a urea-reduction ratio of approximately 65 percent or a single-pool Kt/V of 1.25, depending on the amount of ultrafiltration. The high-dose goal was an equilibrated Kt/V of 1.45 (a urea-reduction ratio of about 75 percent or a single-pool Kt/V of approximately 1.65). The flux intervention was defined by the porosity of the dialysis membrane: the flux was classified as low if the mean beta2-microglobulin clearance was less than 10 ml per minute and as high if the ultrafiltration coefficient was more than 14 ml per hour per millimeter of mercury and the mean beta2-microglobulin clearance was more than 20 ml per minute. The beta2-microglobulin clearance was determined on the basis of the intradialytic changes in the beta2-microglobulin concentration and weight, under the assumption of a single-compartment model.17

A total of 8 types of low-flux dialyzers and 17 types of high-flux dialyzers were approved for use in the study. The most common dialyzers with low flux were the Fresenius F8 (used in 46 percent of sessions) and the Baxter CA210 (used in 43 percent); the most common dialyzers with high flux were the Fresenius F80 (used in 43 percent of sessions) and the Baxter CT190 (used in 48 percent). Use of dialyzers with unsubstituted cellulosic membranes was not permitted. Reuse of dialyzers was permitted; however, the number of uses allowed was limited by the clearance profile of the dialyzer.18 The most frequent reprocessing methods involved the use of Renalin (Minntech) without bleach (in 40 percent of sessions), formaldehyde or glutaraldehyde (Diacide, Gulfstream) with bleach (in 32 percent of sessions), or heated citric acid (in 10 percent of sessions). Established standards of general medical care were monitored by a quality-of-care committee.19

Study Patients

Patients 18 to 80 years of age who were undergoing in-center hemodialysis thrice weekly and had been undergoing hemodialysis for three or more months were enrolled between March 1995 and October 2000. The patients' base-line dialysis prescriptions were maintained for three weeks after enrollment while base-line characteristics were ascertained. Subsequently, dialysis prescriptions were adjusted to evaluate the ability to achieve the high-dose goal in each patient. Patients were excluded from randomization if their residual urea clearance exceeded 1.5 ml per minute per 35 liters of urea, if their serum albumin level was less than 2.6 g per deciliter, or if an equilibrated Kt/V of more than 1.30 was not achieved within 4.5 hours during two of three consecutive monitored dialysis sessions in which the high-dose goal was targeted. On the basis of the last criterion, very heavy patients were excluded; 97 percent of the patients who underwent randomization weighed less than 100 kg.

Interventions and Follow-up

Centrally determined dialysis prescriptions were transmitted to investigators at the time of randomization and updated monthly throughout follow-up. The target equilibrated Kt/V was achieved by manipulating the duration of the treatment session and the dialyzer clearance commensurate with ultrafiltration requirements. Dialysis was provided in as short a time as possible but not less than 2.5 hours. Adherence was monitored by monthly modeling of urea kinetics.

Beta2-microglobulin clearance was measured every two months in patients assigned to high-flux dialyzers and every six months in those assigned to low-flux dialyzers. Serum obtained monthly before dialysis was assayed for albumin by nephelometry. Blood urea nitrogen, serum beta2-microglobulin, and albumin levels were centrally determined (Spectra East, Rockleigh, N.J.). Standardized assessments of coexisting conditions (with the use of the Index of Coexisting Disease)20 were performed at base line and annually thereafter.19

Kinetic modeling, with multiple measurements of blood urea and beta2-microglobulin during and after dialysis, was performed at months 4 and 36 for validation of kinetic models. On the basis of these modeling sessions, a modified rate equation (Table 1) involving a smaller adjustment for urea rebound after dialysis and yielding a higher equilibrated Kt/V value was developed for reporting the study results. The original rate equation was retained in the definition of dose groups in the study design.

View this table:
[in this window]
[in a new window]
  		Table 1. Base-Line Characteristics.

 
Outcomes

The primary outcome was death from any cause. The main secondary outcomes were the rate of all hospitalizations not related to vascular access and three composite end points: the first hospitalization for cardiac causes or death from any cause, the first hospitalization for infection or death from any cause, and the first decline of more than 15 percent from base line in the serum albumin level or death from any cause. Four additional secondary outcomes were defined for the specific evaluation of cardiac or infectious events: death from cardiac causes, death from infectious causes, the composite of the first hospitalization for cardiac causes or death from cardiac causes, and the composite of the first hospitalization for infectious causes or death from infectious causes. Classifications of hospitalizations for cardiac or infectious causes and causes of death were determined locally and reviewed by an outcomes committee that was unaware of the treatment-group assignments.21

Statistical Analysis

The primary analysis was a Cox regression analysis22 of survival from the time of randomization. The effects of the dose and flux interventions were determined with stratification according to clinical center and were adjusted for the following prespecified base-line factors: age, sex, race, years of dialysis, presence or absence of diabetes, score for coexisting conditions excluding diabetes, albumin level, and the interaction of albumin level with time from randomization. For patients who received a kidney transplant, data were censored at the time of transplantation. However, in keeping with the intention-to-treat principle, data were not censored when patients were transferred to a nonparticipating center or switched to an alternative method of dialysis. Kaplan–Meier survival curves were constructed.23 All reported P values are two-sided, without adjustment for multiple comparisons.

Interactions of the dose and flux interventions with the prespecified base-line factors were tested individually to determine whether the interventions had different effects on mortality in subgroups defined according to these factors. Subgroups defined according to continuous variables (age, albumin level, and years of dialysis) were defined by their mean values.

Secondary composite outcomes were analyzed by Cox regressions similar to that used in the primary analysis, but with data on patients who were transferred to nonparticipating centers censored because hospitalizations and decreases in albumin levels could not be monitored. The rate of hospitalizations not related to vascular access was analyzed by overdispersed Poisson regression.24

We planned for randomization of 900 patients during an initial accrual period of 1.5 years, with 4 years of additional follow-up during which patients who died, received a kidney transplant, or were transferred to a nonparticipating dialysis unit would be replaced by newly randomized patients until the final year of the trial. Thus, the planned follow-up ranged from 1 to 6.5 years, depending on the time of randomization. Under conservative assumptions, including a 20 percent reduction in mortality as compared with that in the general population of patients undergoing hemodialysis, a six-month treatment lag time, and no carryover effects of the interventions after transfer to an alternative method of dialysis or to another dialysis unit, the study had 84 percent power to detect a 25 percent reduction in mortality for each intervention.

An external advisory committee monitored the study with the use of six annual interim analyses conducted with O'Brien–Fleming stopping boundaries.25 The nominal significance level for the final primary analysis of mortality was 0.042, in order to maintain an overall type I error rate of 5 percent for both interventions. The study was designed and conducted by a steering committee made up of academic investigators, who also held and analyzed the data.

Results

Base-Line Characteristics of the Patients

A total of 2677 patients were screened, of whom 1846 underwent randomization. Randomized patients had high rates of coexisting conditions: 96 percent had hypertension, 45 percent had either type 1 or type 2 diabetes, and 80 percent had a history of cardiac disease. Characteristics of the patients in the two dose groups were similar, as were the characteristics of those in the two flux groups (Table 1).

Characteristics of and Adherence to Treatment

Mean values for treatment variables (reported for 97 percent of monitored dialysis sessions) characterizing the treatment groups were attained by the first monthly measurement after randomization and remained stable throughout follow-up. The mean (±SD) equilibrated Kt/V during follow-up was 1.16±0.08 in the standard-dose group and 1.53±0.09 in the high-dose group (Table 2); the difference between the two was 0.37, or 92.5 percent of the targeted 0.40. The mean equilibrated Kt/V during follow-up was less than 1.25 for 93 percent of the patients in the standard-dose group and more than 1.35 for 92 percent of the patients in the high-dose group, indicating that there was little overlap between groups. The mean urea-reduction ratio was 66.3±2.5 percent in the standard-dose group and 75.2±2.5 percent in the high-dose group, and the mean single-pool Kt/V was 1.32±0.09 in the standard-dose group and 1.71±0.11 in the high-dose group.

View this table:
[in this window]
[in a new window]
  		Table 2. Mean Characteristics of Treatment during Follow-up.

 
The mean beta2-microglobulin clearance during follow-up was 30.4 ml per minute higher in the high-flux group (33.8±11.4 ml per minute) than in the low-flux group (3.4±7.2 ml per minute). The mean beta2-microglobulin clearance was less than 5 ml per minute for each low-flux dialyzer, regardless of reprocessing method, and exceeded 30 ml per minute for each combination of dialyzer and reprocessing method used in the high-flux group, except for the CT190 reused with the Renalin method (25 ml per minute).

Death from Any Cause

For the 1846 randomized patients, the mean interval between randomization and the scheduled end of the study on December 31, 2001, was 4.48 years. However, attrition due to death and transplantation resulted in a mean follow-up time of 2.84 years, with 5237 total patient-years of follow-up. During follow-up, 392 patients left their dialysis center for reasons other than death, including 194 who received a transplant (95 in the standard-dose group and 99 in the high-dose group; 93 in the low-flux group and 101 in the high-flux group) and 198 others who switched methods of dialysis or transferred to nonparticipating facilities. Nonetheless, vital status was determined for all randomized patients. There were 871 deaths or 0.166 death per patient-year (Table 3 and Figure 1). Cardiac diseases were the leading cause of death (343 of 871 [39 percent]).

View this table:
[in this window]
[in a new window]
  		Table 3. Primary and Secondary Outcomes.

 

View larger version (16K):
[in this window]
[in a new window]
  		Figure 1. Survival Curves for the Treatment Groups.

After adjustment for the base-line factors, mortality in the high-dose group was 4 percent lower (95 percent confidence interval, –10 to 16; P=0.53) than that in the standard-dose group (Panel A), and mortality in the high-flux group was 8 percent lower (95 percent confidence interval, –5 to 19; P=0.23) than that in the low-flux group (Panel B).

 
All prespecified covariates were significant independent predictors of death (Table 4). The strongest predictors were age (a 41 percent increase in risk per 10-year increment), base-line serum albumin level (a 49 percent decrease in risk per increment of 0.5 g per deciliter), coexisting conditions (a 37 percent increase in risk per 1-unit increment in the score on the Index of Coexisting Disease), race (the risk of death was 23 percent lower for blacks), and years of dialysis (a 4 percent increase in risk per additional year of dialysis).

View this table:
[in this window]
[in a new window]
  		Table 4. Primary Cox Regression Analysis of Mortality.

 
Neither the differences between the two dose groups nor the differences between the two flux groups were significant, and the 95 percent confidence intervals for both interventions included zero benefit. After adjustment for base-line factors, the high-dose group had a risk of death that was 4 percentage points lower (95 percent confidence interval, –10 to 16; P=0.53) than that in the standard-dose group, and the high-flux group had a risk of death that was 8 percentage points lower (95 percent confidence interval, –5 to 19; P=0.23) than that in the low-flux group. The effects of the dose and flux interventions were similar at both levels of the other variable (P for interaction=0.30). Results were similar when the data were analyzed without adjustment for the predefined covariates.

Secondary Outcomes

Results for the main secondary outcomes are shown in Figure 2. Risk reductions for the high-dose group were 1 percentage point (95 percent confidence interval, –12 to 12; P=0.91) for the composite cardiac outcome, 3 percentage points (95 percent confidence interval, –9 to 14; P=0.60) for the composite infection-related outcome, and 4 percentage points (95 percent confidence interval, –6 to 13; P=0.38) for hospitalizations not related to vascular access. Effects of the high-flux as compared with the low-flux intervention ranged from an increase in risk of 1 percentage point (95 percent confidence interval, –9 to 11; P=0.87) for hospitalizations not related to vascular access to a decrease of 10 percentage points (95 percent confidence interval, –1 to 20; P=0.08) for the composite cardiac outcome, indicating that neither the dose intervention nor the flux intervention had statistically significant effects on any of the main secondary outcomes. Significant risk reductions (unadjusted P<0.05) in the risk of death from cardiac causes and in the combined outcome of first hospitalization for cardiac causes or death from cardiac causes were observed for the high-flux intervention.


View larger version (21K):
[in this window]
[in a new window]
  		Figure 2. Effects of the Treatment Interventions on Primary and Secondary Outcomes.

Values are shown as relative risks and 95 percent confidence intervals associated with assignment to the high-dose group as compared with assignment to the standard-dose group and assignment to the high-flux group as compared with assignment to the low-flux group. Analyses were stratified according to clinical center and adjusted for base-line age, sex, race, duration of dialysis, presence or absence of diabetes, score for coexisting conditions excluding diabetes, serum albumin level, and the interaction of albumin level with time from randomization. Percentage risk reductions for the high-dose and high-flux groups given in the text are obtained by subtraction of the relative risk from 1 followed by multiplication by 100.

 
Interactions of Treatment Interventions with Base-Line Characteristics

Among the seven prespecified base-line factors in the primary analysis, possible interactions were identified (unadjusted P<0.05) between the dose intervention and sex (P=0.01) and between the flux intervention and years of dialysis (<=3.7 years vs. >3.7 years, P=0.005). In the high-dose group, the risk of death among women was 19 percentage points lower than that in the standard-dose group, but the risk of death among men was 16 percentage points higher than that in the standard-dose group. The risk of death was 32 percentage points lower in the high-flux group than in the low-flux group among patients with more than 3.7 years of dialysis before randomization but was similar in the two flux groups among patients with fewer years of prior dialysis. The interaction between the flux group and years of dialysis, but not that between the dose group and sex, remained significant at the level of 0.0071 determined with the Bonferroni method for seven tests of interaction. However, the strength of the interaction between the flux group and years of dialysis was reduced if the number of years of dialysis was treated as a continuous variable (P=0.04).

Discussion

Among patients undergoing maintenance hemodialysis who were receiving thrice-weekly treatments lasting 2.5 to 4.5 hours each, neither a higher dose of dialysis nor the use of high-flux membranes significantly improved survival or reduced morbidity. This lack of effect was found despite a clear separation in fractional urea and beta2-microglobulin clearances between the treatment groups. Study participants had a substantial prevalence of coexisting conditions similar to that in the overall population of patients in the United States undergoing hemodialysis. The study included a higher percentage of blacks (63 percent) than this overall population does, because of the predominance of urban centers, but patients with diabetes, elderly patients, and patients with cardiac disease were well represented. After the age-related entry criteria and race had been accounted for, mortality in our study was similar to that in the general population of patients in the United States undergoing hemodialysis.26 These findings support the continued use of the current U.S. practice guidelines, which recommend a single-pool Kt/V of at least 1.24 but make no recommendation for or against the routine use of high-flux membranes.

The mean achieved equilibrated Kt/V, single-pool Kt/V, and urea-reduction ratio in our standard-dose group were 1.16, 1.32, and 66.3 percent, respectively. Several observational studies, but to our knowledge no other randomized trials, have examined the relation between mortality and dialysis doses above these levels. Some,9,27 but not all,7 of these studies have reported reductions in mortality of approximately 20 percentage points with an increase in dose similar to that used in our study. The overall reduction in mortality in our high-dose group was 4.3 percentage points, with a 95 percent confidence interval of –10 percentage points to 16 percentage points. The upper confidence limits for risk reductions in secondary composite outcomes were lower, ranging from 9 percentage points for the composite albumin end point to 14 percentage points for the composite infection end point. Because randomization prevents many biases that can occur in observational studies,28,29 our study appears to rule out an average beneficial effect of a higher dose on survival similar to the larger estimates reported in some observational studies.

The high-flux group in our study had a risk of death from any cause that was 8 percentage points lower than that in the low-flux group, with an upper confidence limit of 19 percentage points. This finding contrasts with those of observational studies12,30 that have reported greater reductions in mortality of 20 to 24 percentage points associated with high-flux dialysis. Although, in our study, total mortality was not significantly reduced in the high-flux group, possible reductions in the rate of death and hospitalizations from cardiac causes were suggested (Figure 2 and Table 3).

The primary intention-to-treat analysis pertains to the overall effect for all patients and does not rule out the possibility of different effects in specific subgroups. Several observational studies have proposed that an increased dose of dialysis may be of particular benefit in whites as compared with blacks,31 in white women as compared with black men,32 in smaller patients,32 and possibly in patients with diabetes as compared with nondiabetics.33 Subgroup analysis in our study suggests that the high dose had a greater benefit in women but not in whites or patients with diabetes. Because of the strong association between sex and body size, the issue of possible dependence of the dose effect on size is complex and is not considered here. Our results also suggest a benefit of high-flux membranes for patients who had been undergoing dialysis for more than 3.7 years. Sex and years of dialysis were among seven factors that were prespecified for investigation of subgroup effects. However, the risk of false positive results from multiple subgroup analyses must be considered, and the results of such analyses should be interpreted cautiously.

The difference between the doses achieved in the high-dose and standard-dose groups most likely represents the maximal practical difference under conditions of thrice-weekly dialysis, as currently practiced in the United States. The higher small-solute clearances that are achievable with much longer or more frequent treatments might result in a clinical benefit that was not attained by the high dose we used.34,35 By the same token, the higher beta2-microglobulin clearances achievable with hemodiafiltration11,36 or sorbent techniques37 might also improve outcomes.

In summary, although the effect of the dose and level of membrane flux may vary among selected subgroups of patients, the primary results of our study indicate that, with a schedule of thrice-weekly dialysis, neither an increased dose of dialysis nor use of a high-flux membrane substantially improves survival, reduces the rate of hospitalization, or maintains serum albumin levels as compared with a standard dose and use of low-flux membranes.

Supported by the National Institute of Diabetes and Digestive and Kidney Diseases. Additional support was provided by Baxter Healthcare, Fresenius Medical Care, R&D Laboratories, and Ross Laboratories.

Drs. Eknoyan, Levin, Schwab, and Toto have reported consulting for Amgen; Dr. Eknoyan has reported consulting for Sigma Tau and Fresenius and receiving lecture fees from Genzyme. Dr. Delmez has reported consulting for Abbott Laboratories. Drs. Cheung and Schwab have reported consulting for Gambro. Dr. Toto has reported consulting for Merck and Bristol-Myers Squibb and receiving lecture fees from Amgen, Merck, Bristol-Myers Squibb, and Ortho Biotech. Dr. Schwab has reported receiving lecture fees from Amgen. Dr. Beck has reported having equity ownership or stock options in Amgen, Pfizer, and Eli Lilly; Dr. Delmez has reported having equity ownership or stock options in 3M and Gelrex Pharm; and Dr. Levin has reported having equity ownership or stock options in Fresenius. Dr. Levey has reported receiving grant support from Amgen; Dr. Schwab has reported receiving grant support from the Vasca LifeSite Trial; and Dr. Toto has reported receiving grant support from Eli Lilly.

We are indebted to the patients who participated in the study; to all our consultants, especially Frank Gotch; to Allan Collins and Shu Chen of the U.S. Renal Data System for providing additional hospitalization data; and to Isabel McPhillips for assistance in the preparation of the manuscript.

* The institutions and investigators in the study group are listed in the Appendix.


Source Information

From the Baylor College of Medicine, Houston (G.E.); the Cleveland Clinic Foundation, Cleveland (G.J.B., T.G.); the University of Utah and the Veterans Affairs Salt Lake City Health Care System, Salt Lake City (A.K.C.); the University of Illinois and the Veterans Affairs Chicago Health Care System, Chicago (J.T. Daugirdas); the National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, Md. (J.W.K.); the University of Alabama at Birmingham, Birmingham (M.A.); Emory University Hospital, Atlanta (J.B.); Washington University, St. Louis (J.A.D.); the University of California at Davis, Sacramento (T.A.D.); New England Medical Center, Boston (J.T. Dwyer, A.S.L.); Beth Israel Medical Center, New York (N.W.L.); Brigham and Women's Hospital, Boston (E.M.); the University of Rochester, Rochester, N.Y. (D.B.O.); Wake Forest University, Winston-Salem, N.C. (M.V.R.); Vanderbilt University, Nashville (G.S.); Duke University, Durham, N.C. (S.J.S.); Lankenau Hospital and Medical Research Center, Wynnewood, Pa. (B.P.T.); and the University of Texas Southwestern Medical Center, Dallas (R.T.).

Address reprint requests to Dr. Beck at the HEMO Study Data Coordinating Center, Dept. of Biostatistics and Epidemiology, Wb4, Cleveland Clinic Foundation, 9500 Euclid Ave., Cleveland, OH 44195, or at beckg{at}ccf.org.

References

  1. Renal Data System. USRDS 2001 annual data report: atlas of end-stage renal disease in the United States. Bethesda, Md.: National Institute of Diabetes and Digestive and Kidney Diseases, 2001. 
  2. Owen WF Jr, Lew NL, Liu Y, Lowrie EG, Lazarus JM. The urea reduction ratio and serum albumin concentration as predictors of mortality in patients undergoing hemodialysis. N Engl J Med 1993;329:1001-1006. [Free Full Text]
  3. Gotch FA, Sargent JA. A mechanistic analysis of the National Cooperative Dialysis Study (NCDS). Kidney Int 1985;28:526-534. [ISI][Medline]
  4. Eknoyan G, Levin N. NKF-K/DOQI clinical practice guidelines: update 2000. Am J Kidney Dis 2001;37:Suppl 1:S5-S6. [Erratum, Am J Kidney Dis 2001;38:917.] [Medline]
  5. Daugirdas JT, Depner TA, Gotch FA, et al. Comparison of methods to predict equilibrated Kt/V in the HEMO Pilot Study. Kidney Int 1997;52:1395-1405. [Medline]
  6. Lowrie EG, Laird NM, Parker TF, Sargent JA. Effect of the hemodialysis prescription on patient morbidity: report from the National Cooperative Dialysis Study. N Engl J Med 1981;305:1176-1181. [Abstract]
  7. Shinzato T, Nakai S, Akiba T, et al. Survival in long-term haemodialysis patients: results from the annual survey of the Japanese Society for Dialysis Therapy. Nephrol Dial Transplant 1997;12:884-888. [Free Full Text]
  8. Wolfe RA, Ashby VB, Daugirdas JT, Agodoa LY, Jones CA, Port FK. Body size, dose of hemodialysis, and mortality. Am J Kidney Dis 2000;35:80-88. [Medline]
  9. Port FK, Ashby VB, Dhingra RK, Roys EC, Wolfe RA. Dialysis dose and body mass index are strongly associated with survival in hemodialysis patients. J Am Soc Nephrol 2002;13:1061-1066. [Free Full Text]
  10. Held PJ, Port FK, Wolfe RA, et al. The dose of hemodialysis and patient mortality. Kidney Int 1996;50:550-556. [ISI][Medline]
  11. Locatelli F, Marcelli D, Conte F, Limido A, Malberti F, Spotti D. Comparison of mortality in ESRD patients on convective and diffusive extracorporeal treatments: the Registro Lombardo Dialisi E Trapianto. Kidney Int 1999;55:286-293. [CrossRef][ISI][Medline]
  12. Port FK, Wolfe RA, Hulbert-Shearon TE, et al. Mortality risk by hemodialyzer reuse practice and dialyzer membrane characteristics: results from the USRDS dialysis morbidity and mortality study. Am J Kidney Dis 2001;37:276-286. [ISI][Medline]
  13. Eknoyan G, Levey AS, Beck GJ, et al. The Hemodialysis (HEMO) Study: rationale for selection of interventions. Semin Dial 1996;9:24-33. 
  14. Greene T, Beck GJ, Gassman JJ, et al. Design and statistical issues of the Hemodialysis (HEMO) Study. Control Clin Trials 2000;21:502-525. [CrossRef][ISI][Medline]
  15. Depner TA. Prescribing hemodialysis: a guide to urea modeling. Dordrecht, the Netherlands: Kluwer Academic, 1991.
  16. 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. [Medline]
  17. Leypoldt JK, Cheung AK, Deeter RB. Single compartment models for evaluating beta 2-microglobulin clearance during hemodialysis. ASAIO J 1997;43:904-909. [Medline]
  18. Cheung AK, Agodoa LY, Daugirdas JT, et al. Effects of hemodialyzer reuse on clearances of urea and beta2-microglobulin. J Am Soc Nephrol 1999;10:117-127. [Free Full Text]
  19. HEMO Study protocol. Bethesda, Md.: National Institute of Diabetes and Digestive and Kidney Diseases, June 1999.
  20. Miskulin DC, Athienites NV, Yan G, et al. Comorbidity assessment using the Index of Coexistent Diseases in a multicenter clinical trial. Kidney Int 2001;60:1498-1510. [CrossRef][Medline]
  21. Rocco MV, Yan G, Gassman J. Comparison of causes of death using HEMO Study and HCFA end-stage renal disease death notification classification systems. Am J Kidney Dis 2002;39:146-153. [ISI][Medline]
  22. Cox DR. Regression models and life-tables. J R Stat Soc [B] 1972;34: 187-220.
  23. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457-81.
  24. McCullagh P, Nelder JA. Generalized linear models. 2nd ed. London: Chapman & Hall, 1989.
  25. O'Brien PC, Fleming TR. A multiple testing procedure for clinical trials. Biometrics 1979;35:549-556. [CrossRef][ISI][Medline]
  26. Renal Data System. USRDS 2000 annual data report: atlas of end-stage renal disease in the United States. Bethesda, Md.: National Institute of Diabetes and Digestive and Kidney Diseases, 2000 (table H.5).
  27. Li Z, Lew NL, Lazarus JM, Lowrie EG. Comparing the urea reduction ratio and the urea product as outcome-based measures of hemodialysis dose. Am J Kidney Dis 2000;35:598-605. [Medline]
  28. The justification for randomized controlled trials. In: Pocock SJ. Clinical trials: a practical approach. New York: John Wiley, 1983:50-65.
  29. Dunn D, Babiker A, Hooker M, Darbyshire J. The dangers of inferring treatment effects from observational data: a case study in HIV infection. Control Clin Trials 2002;23:106-110. [CrossRef][Medline]
  30. Hornberger JC, Chernew M, Petersen J, Garber AM. A multivariate analysis of mortality and hospital admissions with high-flux dialysis. J Am Soc Nephrol 1992;3:1227-1237. [Abstract]
  31. Owen WF Jr, Chertow GM, Lazarus JM, Lowrie EG. Dose of hemodialysis and survival: differences by race and sex. JAMA 1998;280:1764-1768. [Free Full Text]
  32. Lowrie EG, Chertow GM, Lew NL, Lazarus JM, Owen WF. The urea [clearance x dialysis time] product (Kt) as an outcome-based measure of hemodialysis dose. Kidney Int 1999;56:729-737. [CrossRef][ISI][Medline]
  33. Collins AJ, Ma JZ, Umen A, Keshaviah P. Urea index and other predictors of hemodialysis patient survival. Am J Kidney Dis 1994;23:272-282. [Erratum, Am J Kidney Dis 1994;24:157.] [ISI][Medline]
  34. Mohr PE, Neumann PJ, Franco SJ, Marainen J, Lockridge R, Ting G. The case for daily dialysis: its impact on costs and quality of life. Am J Kidney Dis 2001;37:777-789. [Medline]
  35. Vos PF, Zilch O, Kooistra MP. Clinical outcome of daily dialysis. Am J Kidney Dis 2001;37:Suppl 2:S99-S102. [Medline]
  36. Ward RA, Schmidt B, Hullin J, Hillebrand GF, Samtleben W. A comparison of on-line hemodiafiltration and high-flux hemodialysis: a prospective clinical study. J Am Soc Nephrol 2000;11:2344-2350. [Free Full Text]
  37. Winchester JF. Sorbent hemoperfusion in end-stage renal disease: an in-depth review. Adv Ren Replace Ther 2002;9:19-25. [CrossRef][Medline]
Appendix

The following institutions and investigators participated in the HEMO Study: Beth Israel Medical Center — N. Levin, A. Kaufman, J. Burrowes, S. Gibbons, D. Schneditz, M. DeVita, R. Marcus, I. Meisels, J. Uribarri, C. Williams, S. Patel, C. Campasano, S. Cevallos, F. Coles, M. Damon, J. Dumont, B. Ferris, K. Gans, S. Golub, J. Leung; Brigham and Women's Hospital — E. Milford, G. Chertow, T. Steinman, M. Williams, A. Bullard, J. Rehm-McGillicuddy, P. Hertello, P. Hayden, L. Campbell, W. Owen, P. Keaney, D. Kines, L. Lumadue, G. Interbartolo, E. Melville, C. Spencer, D. Bission, M. Lazarus; Duke University Medical Center — S. Schwab, D. Butterly, M. Berkoben, W. Owen, S. Minda, D. Bartholomay, L. Poe, A. Brandenburg, M. Hunsburger; Emory University Hospital — J. Bailey, E. Macon, B. Maroni, I. Brumfield, G. Carmichael, J. Miller, S. Marjoram, A. Yabrow, L. Akpele, A. McGhee, N. Dandrea; Lankenau Hospital–Medical Research Center — B. Teehan, R. Benz, J. Brown, C. Bergen, J. Butler; New England Medical Center — A. Levey, K. Meyer, R. Chawla, J. Strom, G. Narayan, N. Aurigemma, A. Martin, C. Moleske, L. Uhlin, H. Han, L. Lambert, N. Athienites, D. Miskulin, M. Sarnak, M. Unruh; University of Alabama at Birmingham — M. Allon, E. Rutsky, J. Lewis, J. Forehand, J. Dockery, L. Leith, W. Perdue, S. Wade Patrick, N. Miller, J. Jordan, T. Smith, D. Dunston, D. Gregg; University of California at Davis — T. Depner, G. Kaysen, M. Powers, A. Reasons, L. Eder, S. Garcia, N. Scott, T. Nguyen, Z. Ali, J. Bowman, K. Van Sickle; University of Illinois — J. Daugirdas, P. Balter, A. Priester-Coary, C. Carey, A. Frydrych, S. Haynes, B. McQuiston; University of Rochester — D. Ornt, J. Holley, W. Choudhry, M. Schiff, S. Silver, N. Ferris, C. Fantauzzo, D. Pomerantz, A. Erenstone, E. Fennelley, S. Moser, L. Palm-Montalbano, D. Novak, J. Pata, L. Kirk; University of Texas Southwestern Medical Center — R. Toto, R. Star, T. Parker, R. Hootkins, S. Glowacki, R. Kunau, J. Krivacic, C. Wright, L. Villemarette, J. Odom, L. Sturdivant, S. Bergeron, D. Dews, R. Santos, L. Colon, S. Black, A. Reyes, R. Paul, E. Hatfield; University of Utah — A. Cheung, J. Leypoldt, S. Beddhu, C. Kablitz, K. Allen-Brady, J. Davis, R. Deeter, J. Gilson, L. Sala, S. Ware, O. Brumbaugh, A. Atkinson, K. Krill; Vanderbilt University Medical Center — G. Schulman, J. Lewis, M. Faulkner, S. McLeroy, M. Sika, P. Hopkins, S. Powers, M. Deere, A. Gung, D. Nunn, F. Hendricks, L. Kitchen, L. Watkins, J. Sturgis, A. Taheri, P. Rickard, M. Byrd, E. Leavell, R. Waller, J. Bigelow, L. Busby, A. Fowler, K. Channell; Wake Forest University — M. Rocco, J. Burkart, S. Crawford, P. Green, D. Poole, L. Hagan, T. Young; Washington University — J. Delmaz, D. Coyne, D. Windus, R. Creaghan, K. Giles, K. Norwood, J. Audrain; National Institute of Diabetes and Digestive and Kidney Diseases — J. Kusek, L. Agodoa, J. Briggs, P. Kimmel, R. Star. Steering Committee — G. Eknoyan, Chair. Data Coordinating Center (Cleveland Clinic Foundation) — G. Beck, J. Gassman, T. Greene, R. Heyka, M. Drabik, B. Larive, H. Litowitz, I. McPhillips, L. Paranandi, M. Radeva, C. Rasmussen, L. Tuason, B. Weiss, K. Wiggins, G. Yan. Nutrition Coordinating Center (New England Medical Center) — J. Dwyer, J. Leung, P. Cunniff, R. Henry, D. Bell. Spectra East — J. Zazra, O. Gindi-Takla, F. Wawra. Center for Medicare and Medicaid Services — J. Greer. Consultants — W. Chumlea, F. Gotch, R. Hays, M. Keen. Industrial Representatives — M. Arzac, W. Clark, D. Cockram, R. Makoff, B. Rogers. External Advisory Committee — R. Blantz, W. Harmon, L. Hunsicker, J. Kopple, W. Mitch, A. Nissenson, J. Stokes, W. Vollmer, R. Wolfe.


 

This Article
-Abstract
- PDF
-PDA Full Text
-PowerPoint Slide Set

Commentary
-Editorial
by Himmelfarb, J.
-Letters

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited

More Information
-PubMed Citation

Related Letters:

Effect of Dialysis Dose and Membrane Flux in Maintenance Hemodialysis
Scribner B. H., Blagg C. R., Friedman E. A., Hoenich N. A., Locatelli F., Greene T., Cheung A. K., Eknoyan G., the Hemodialysis (HEMO) Study Group , Himmelfarb J.
Extract | Full Text | PDF  
N Engl J Med 2003; 348:1491-1494, Apr 10, 2003. Correspondence

This article has been cited by other articles:


HOME  |  SUBSCRIBE  |  SEARCH  |  CURRENT ISSUE  |  PAST ISSUES  |  COLLECTIONS  |  PRIVACY  |  HELP  |  beta.nejm.org

Comments and questions? Please contact us.

The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved.