Hemodialysis trends in time, 1989 to 1998, independent of dose and outcome

Hemodialysis Trends in Time, 1989 to 1998, Independent of Dose and Outcome Bernard Charra, MD, Guy Laurent, MD, Charles Chazot, MD, Guillaume Jean, MD, Jean-Claude Terrat, MD, and Thierry Vanel, MD ● The 1989 Morbidity, Mortality, and Prescription of Dialysis Symposium (Dallas, TX) deeply modified the way of prescribing and delivering hemodialysis (HD) in the United States, as elsewhere. Among strategies emerging for end-stage renal disease treatment improvement, increasing the delivered dose has been the central issue, whereas the session time effect has remained an unsettled question. The purpose of this article is to analyze the trends in HD session time over the last decade in Tassin, France, and elsewhere and their relationship to outcome, independent of dialysis dose. The published data indicate that in the United States, there has been an increase in session time. Outside the United States, where session time had been somewhat longer, there has been either no change or a decrease in time without apparent ill effects. The dialysis dose, as measured by the urea Kt/V, has increased everywhere because improved technology allows for efficient toxin removal within a very short dialysis session. In Tassin, a recent limited experience of high-dose shortened dialysis did not show a significant short-term survival difference, but an impaired control of blood pressure (BP) and nutrition. Thus far, the highest HD long-term survival rates have been reported by the groups that used the largest doses and the longer times of HD. Effects of dose and time are, at this point, impossible to disentangle. A longer dialysis time improves extracellular volume (ECV) and BP control and decreases cardiovascular mortality. Shortening the HD session leads to impaired control of BP and increasing cardiovascular morbidity (by far the first cause of mortality on HD). In the future, the length of dialysis session should be governed by the demand of BP control. r 1998 by the National Kidney Foundation, Inc. INDEX WORDS: Hemodialysis; time; dose; mortality; blood pressure.

N

INE YEARS have passed since the Morbidity, Mortality, and Prescription of Dialysis Symposium (MMPDS) was held in Dallas, TX.1 This milestone meeting has induced deep changes in the United States, as elsewhere, in the way dialysis is prescribed and delivered. Among questions unsettled by the MMPDS was the session time eventual effect. One purpose of this study was to collect and analyze the available information on hemodialysis (HD) time trends and its clinical effect during the last 10 years in Tassin, France, and elsewhere. Another purpose was to determine whether there is evidence that session time may influence outcome, and, if so, if it is possible to separate the effect of time from the effect of dose. Only patients dialyzed thrice weekly were considered in this study. PATIENTS AND METHODS We summarized the Tassin experience in the last decade, reviewed the literature, and directly questioned nephrologists involved in dialysis registries or organizations.

Changes in Tassin During the last 10 years, changes occurring in Tassin concerned the mode of treatment and the incident patient case mix. The traditional treatment delivered for the large majority of Tassin patients has remained almost unchanged for 30

years. It has been described in detail elsewhere2,3: three times weekly for 8 hours using cuprophane membrane dialyzers (mostly 1-m2 area Kiil up to 1990), with a blood flow of 200 mL/min or greater. The acetate buffer used until 1995 is now progressively replaced by bicarbonate in more than 70% of the patients. The mean urea Kt/V using secondgeneration Daugirdas method4 was 1.97 per session, and the mean protein catabolic rate (PCR) was 1.26. Mean protein and calorie intakes were 1.24 g/kg and 29 kg-cal/kg of body weight/d, respectively. The mean salt intake was 5 g/d and the mean interdialytic weight gain 1.7 kg/d. No antihypertensive medication was needed in 98% of the patients after 2 months of dialysis. The incidence of hypotensive episodes was 75 per 1,000 HD sessions. The mean predialysis blood pressure (BP) achieved in the overall population was normal (128/79 mm Hg). Furthermore, the ambulatory BP monitoring circadian value was also normal (119/71 mm Hg).5 A short (three times weekly for 5 hours) dialysis schedule has been progressively set up in the last 10 years using larger area size (1.5 to 2.2 m2 ) cellulosic dialyzers, bicarbonate buffer, and a 300-mL/min blood flow. The mean Kt/V and PCR achieved were slightly less than those on 8-hour dialysis (1.78 and 1.18, respectively). The mean protein and calorie intakes were slightly less (1.12 g/kg and 27 kg-cal/kg of body weight/d) than those on 8-hour HD, but neither salt intake (4.5 g/d) nor interdialytic weight gain (1.6 kg/d) were different. Antihypertensive medications needed to be used

From the Centre de Rein Artificiel, Tassin, France. Address reprint requests to Bernard Charra, MD, Centre de Rein Artificiel, 42 Avenue du 8-Mai-1945, 69160 Tassin, France. E-mail: [email protected]

r 1998 by the National Kidney Foundation, Inc. 0272-6386/98/3206-0411$3.00/0

American Journal of Kidney Diseases, Vol 32, No 6, Suppl 4 (December), 1998: pp S63-S70

S63

S64

CHARRA ET AL

more often than on 8-hour HD (in 13% v 2% of the patients) and the incidence of hypotensive episodes was greater (120 per 1,000 sessions). Over the 10 years, the proportion of patients on the 5-hour schedule increased from 2% to nearly 20%. In the same time, Kiil dialyzer use decreased from 74% to 12% of patients. Reuse was banned in France in 1995. Altogether, although average dialysis session time decreased, because of the single use of higher performance dialyzers, the delivered dose, whether expressed in Kt/V or urea reduction ratio (1.70 to 2.0 and 71% to 79%, respectively) increased. The ratios of effectively delivered to prescribed HD time in the 5and 8-hour schedules were not different (both 0.99). Compliance was, therefore, equally good in both methods. The second important change in Tassin during the last 10 years concerns the incident population. The trend toward an increasingly higher risk population continued as in preceding years (Table 1): in 1997, more than 50% of the incident patients had diabetes or a vascular renal disease. The main change has been the rapidly growing incidence of diabetes; the mean age at start was up to 63 years. A significant cardiovascular history (angina, myocardial infarction, cerebrovascular accident, transient ischemic accident, peripheral ischemia) was present in almost 60% of the patients.

Changes in Methods and Patients Elsewhere Changes in HD method and population occurred in places other than Tassin. Registry reports and literature surveys provided information on trends on delivered dose of dialysis, as well as on patient demographics and comorbidity,6-9 but little on the time trends. The average session time has increased in the United States, remained stable in Australia and New Zealand, but decreased in Europe and Japan over the last 10 years. The only certitude is a negative one: HD session time is a blank spot, in that most registries collect very little or no information on it, as if dialysis session time was today an issue irrelevant to clinical outcome.

RESULTS

Tassin Long Dialysis Results The changes in the treatment method and incident population have had obvious effects on the unadjusted mortality. In the last decade, the crude mortality increased from 105 to 139 deaths per 1,000 patient years. It is of note that, in the same time, the cardiovascular mortality remained stable (37 deaths per 1,000 patient years).

A fair clinical outcome comparison must consider the changing case mix by stratifying the patients into risk groups. The standardized mortality ratio (SMR) adjusts for age, race, sex, and cause of renal failure and uses as reference the United States Renal Data System mortality table.10 Analysis using the SMR method shows (Table 2) that the stratified mortality has remained stable during these 10 years. The number of deaths observed using 8-hour dialysis in Tassin is approximately 50% of the expected value for similar-risk US patients using a lower dialysis dose within a shorter dialysis session. Recalculation of the Tassin SMR value according to dialysis session length (5 hours v 8 hours) for the years 1996 and 1997 results in a higher SMR value in the 5-hour than in the 8-hour group (0.65 v 0.45). Possibly because of the insufficient number of patients included in the 5-hour groups,11 their 0.65 SMR value is not significantly different from the expected US value (P ⫽ 0.16), whereas the 0.45 value achieved by the 8-hour group is very significantly different (P ⬍ 0.001). The analysis of the same data using the Cox model (Table 3) shows that age, cause of renal failure, and cardiovascular comorbidity are very strong independent predictors of mortality. Among treatment-related factors, urea Kt/V has no significant effect on mortality, whereas the more time-related middle molecule index has a significant effect. The strongest mortality predictors are serum albumin level and, even more, predialysis mean arterial pressure (MAP) calculated from all predialysis BP readings during the entire dialysis time. It is, therefore, of great importance that a long dialysis achieves better BP control, as shown by the effect of switching the same patient group from a long to a short dialysis or conversely. One hundred twenty-four unselected dialysis

Table 1. Tassin Incident Population Case Mix Evolution, 1989 to 1997

Diabetes mellitus (%) Nephrosclerosis (%) Other/unknown (%) Age at start (yr) Cardiovascular antecedent (%)

1989

1990

1991

1992

1993

1994

1995

1996

1997

12.8 25.6 61.5 54.1 41.0

18.4 21.1 60.5 55.4 46.0

17.1 24.4 58.5 56.0 47.0

26.4 22.6 50.9 56.5 50.0

20.5 25.6 53.8 58.4 54.0

26.7 20.0 53.3 59.0 57.0

21.1 28.1 50.9 62.0 55.0

25.5 26.0 48.5 61.5 58.0

31.3 21.9 46.9 63.5 61.0

TRENDS IN DIALYSIS TIME

S65

Table 2. Tassin Annual Standardized Mortality Ratio, 1989 to 1997 Calendar Year

O/E* Deaths

SMR†

P

1989 1990 1991 1992 1993 1994 1995 1996 1997

23/43.7 14/42.4 18/44.7 15/46.1 23/47.7 20/50.3 23/57.0 27/56.4 25/48.5

0.53 0.33 0.40 0.33 0.48 0.40 0.40 0.51 0.52

⬍0.005 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001

*O, observed number; E, expected number. †SMR, standardized mortality ratio compared with USRDS mortality tables.10

patients started on dialysis elsewhere were dialyzed temporarily in Tassin, usually while awaiting kidney transplantation. All had been treated for at least 6 months on a 5-hour (or less) schedule, and 65 patients (52%) received a regular antihypertensive treatment. All were switched to the 8-hour dialysis schedule. After 3 months of long dialysis, the mean postdialysis weight (Fig 1) was slightly reduced, predialysis BP was back to almost normal, and antihypertensive treatment was discontinued in all but one patient. Thereafter, BP continued to decrease but body weight increased because of anabolism. Conversely, 49 long-HD Tassin patients were switched to a 5-hour schedule. All had been dialyzed for 8 hours for 6 months or longer. They were all selected, normotensive without any antihypertensive medication, had moderate interdialytic weight gains, and had a good dialysis tolerance. Their blood access allowed for a 300-mL/ min flow. Dialyzer area and blood flow were

Fig 1. One-year evolution of postdialysis weight (M) and predialysis mean arterial pressure (䉱) in 124 patients switched from three times 5- to three times 8-hour dialysis.

increased to maintain an unchanged urea Kt/V. After 1 year (Fig 2), the mean Kt/V remained grossly unchanged (from 1.89 initially to 1.79). Mean arterial pressure was slightly but significantly increased (⫹10 mm Hg) despite a 2.5-kg average dry weight reduction, and antihypertensive medications were introduced in four cases. Changes in Results Using Shorter/Longer Dialysis Outside Tassin The most obvious and positive effect of the MMPDS has been to sensitize the US nephrological community so that the delivered dose of

Table 3. Tassin Patients Mortality Using Proportional Hazards Model

Age at start Diabetes CV antecedent* Middle molecule index Kt/V urea Mean arterial pressure Initial mean arterial pressure Serum albumin

Regression Coefficient

95% Confidence Interval

Relative Risk

95% Confidence Interval

0.049 0.606 0.631 ⫺0.404 0.153 0.033 ⫺0.022 ⫺0.029

0.033-0.067 0.131-1.081 0.204-1.057 ⫺0.681-⫺0.128 ⫺0.348-0.675 0.011-0.056 ⫺0.032-⫺0.011 ⫺0.064-0.007

1.050 1.833 1.879 0.668 1.165 1.034 0.978 0.971

1.033-1.069 1.139-2.947 1.226-2.878 0.506-0.880 0.706-1.964 1.011-1.057 0.968-0.989 0.938-0.993

*CV, cardiovascular antecedent (myocardial infarction, angina, cerebrovascular accident, transient ischemic attack, peripheral ischemia).

S66

CHARRA ET AL

Outside the United States in the past decade, the dialysis session time has tended to decrease in Europe19 and Japan.9,20 Interestingly, in both cases, data analysis shows that gross mortality is lower in patients receiving more than 12 hours of dialysis weekly than in those receiving less.9,19,20 One of these studies20 provides evidence that patient survival does not level off before delivered Kt/V dose reaches 1.8 and dialysis session time reaches 5 hours. According to this study, increasing the session time from 3 to more than 4.5 hours decreases the mortality by a factor of more than 3.20 DISCUSSION

Fig 2. One-year evolution of postdialysis weight (M) and predialysis mean arterial pressure (䉱) in 49 patients switched from three times 8- to three times 5-hour dialysis.

dialysis has been increased since 1990. This was achieved, at least in part, by increasing the dialysis session time.6 The adjusted annual mortality rate, which had reached a 25.8% peak in 1988, improved over the following years, reaching a nadir of 21.7 in 1994.7 This survival improvement cannot be explained by a change in patient case mix, which continued to worsen in the meantime. Observational cross-sectional analysis of populations12 suggesting a lower gross mortality when delivering a higher Kt/V dose has been confirmed while adjusting for an extensive list of risk and comorbidity factors.13 Since 1992, a number of studies in the United States have reported the effect of increasing the delivered dose on clinical outcome.14-17 Whether crosssectional14,17 or prospective and longitudinal,15,16 all these studies showed that when delivered dose increases, mortality decreases. In all cases, the dose was increased by a combination of increased dialyzer clearance and dialysis session time. It is, therefore, impossible to conclude from these studies if the causative factor of the clinical improvement is longer time, higher dose, or both. It is of note that increasing the delivered dose reduces all causes of mortality.18

The session time has decreased everywhere but in the United States, where the average dialysis time was initially the shortest. The session time tends to be 4 to 5 hours almost everywhere. In our unit, we have somewhat followed this trend because of the increasing proportion of patients receiving 5-hour HD. This universal time reduction contrasts with an universal increase in delivered dialysis dose. The mechanistic analysis of the National Cooperative Dialysis Study (NCDS) data21 has brought on the concept of urea Kt/V. Although short HD time had been recognized as a significant morbidity factor in the initial NCDS analysis, it was considered in this later development. Besides, the interpretation of Gotch and Sargent21 of the NCDS data was that clinical failure was a step function of delivered urea Kt/V, and that no clinical benefit could be expected from delivering a Kt/V dose superior to 0.9. In the following years, reduction of dialysis session time became a major target and adequate dialysis was often assimilated to a minimal acceptable dose.22 Before the Dallas 1989 meeting, some investigators already questioned the short dialysis policy.23-25 This was confirmed some years later by reports showing that mortality risk ratio increases when reducing the session time.26,27 Opinions diverged then, and still do, on the cause of this higher mortality: insufficient time effect28 or insufficient dose because of a time reduction not adequately compensated by a dialysis process efficiency increase.29 Our own experience with a 5-hour dialysis, if 5 hours can be today considered a short dialysis

TRENDS IN DIALYSIS TIME

time, is too limited to allow for a firm statement about the effect of time per se on mortality. However, reducing session time has led to an increased morbidity: hypertension became more difficult to control, nutrition was impaired, and intradialytic morbidity was increased. A more prolonged observation time may disclose a survival difference between 5- and 8-hour dialysis. The duration of the dialysis session is a powerful determinant of dialysis dose, as measured by urea Kt/V. For a given patient (and V), K is limited by many factors, especially blood flow and the different recirculation components so that within operational range, it is extremely difficult to substantially increase Kt/V without increasing time (t). To achieve a Kt/V of 1.3, as suggested by the Dialysis Outcome Quality Initiative,30 it has recently been shown using a on-line urea monitor that the minimum session time is 5 hours.31 The dialysis dose has tended to be confounded with urea (or small solutes) removal alone, leaving apart the larger solutes. The effect of time on solute removal is not homogeneous; it becomes increasingly important as solute size increases. The middle-molecule hypothesis was initially proposed32 to explain the lack of correlation between clinical outcome and level of uremic toxicity in the narrow sense of the term. Although the theory was and still is speculative, the fact remains that small-solute accumulation alone is not able to explain the full span of problems observed in chronic renal failure. In uremia, a whole spectrum of middle molecular size solutes (300 to 15,000 da) have been identified33 and implicated in infection,34 hypertension35 and hypotension,36 nutritional problems,37 bone disease,38 and late osteoarticular complications.39 Although they do not have a middle molecular size, some solutes behave as middle molecules because they have a low intercompartmental diffusibility. For the same urea clearance, a shorter HD provides a lower phosphate clearance.40 Conversely, long daily HD41 allows for easy control of hyperphosphoremia. Middle-molecule removal increases patient survival more significantly than small-solute removal according to our (Table 3) and others’ experience.42 The same is true for morbidity.43 A longer session time reduces dialysis side effects. The lower rate of intradialytic events that

S67

we observed using longer sessions, as did others,44,45 enhances compliance,30 a factor of demonstrated importance in dialysis dose delivery. Conversely, the more the session is shortened, the more problems it develops, the more the patient wants to cut it short, and the less prone he is to get his prescribed dialysis dose. A recent multivariate regression study on the barriers to adequate HD delivery46 showed that the two major barriers are patient noncompliance and insufficient prescription, and that increasing the session time to more than 4 hours is the best way to increase dialysis prescription. Nutrition is strongly correlated with dialysis patient outcome.47,48 Increasing the urea Kt/V improves protein and calorie nutrition.49 Conversely, appetite appears to depend more on middle- than small-molecular size solutes.37 Indeed, nutrition is better on 8-hour than short dialysis in our experience, as in that of others.41 A long dialysis time allows for an almost unrestricted food intake. Except for sodium and, to a minor degree, potassium, 8-hour dialysis patients are encouraged to eat a large amount of protein and calories; they do not need to use resins and receive no vitamin supplementation. This large diet explains the important anabolic response after the onset of long dialysis.50 The essential effect of increasing dialysis session time is to improve extracellular volume (ECV) and BP control. The fact that a low BP at the initiation of dialysis, often indicating heart failure, an independent cause of death at that stage,51 correlates with a high short-term mortality52,53 does not contradict the fact that long-term mortality increases with hypertension.51,54-59 The effect of BP on target organs and mortality depends not on a single value, especially not the value at the start of treatment when almost 90% of the patients are hypertensive, but of its integrated values over time.60 In uremic, as in nonuremic patients, the benefit of controlling hypertension takes some years to appear, and it persists for years.61 Treated BP, rather than pretreatment value, predicts survival in hypertensive patients.62 The wide development of shortened dialysis has rapidly led to increased hypertension,25 cardiovascular morbidity, and mortality.23 Shortened session time, with higher ultrafiltration rates and increased use of antihypertensive medica-

S68

tions, leads to a vicious circle63-65 amplifying BP variations and leading to both intradialytic hypotension and interdialytic hypertension. According to the literature, a persistent strict control of volume and BP is extremely difficult to achieve on conventional (4- to 5-hour) HD. Up to now (in small series of patients), using a strict lowsalt diet, lower dialysate sodium, and intensified ultrafiltration, could this goal be achieved on short dialysis.66-68 By contrast, a prolonged session time allows for better control of ECV (expressed by dry weight achievement) and BP with less use of antihypertensive medications. Although there may be a center effect (combining diet, doctor’s dose, strict dry weight, and BP follow-up policy), the good ECV and BP control achieved in Tassin cannot be reduced to it. In the Tassin 5-hour versus 8-hour HD experience, hypertension control was more difficult to achieve in the shorter schedule and forced the use of medications. The experience of modifying the session time within the same group of patients confirms the key role of session time. Other groups using 8-hour dialysis in the past69-71 and more recently41,72,73 also regularly achieved and still achieve long-standing normotension without antihypertensive medications. Thus far, the only proven effective way to control BP in the dialysis patient is the drugfree dry weight method.74 Is dose or time more important? Available data do not provide a real answer, and an answer is not to be expected for some time. Even the current HEMO study75 will not be informative on this point. It uses very limited time ranges, and accordingly even if the higher Kt/V subgroup turns out to have a better outcome, it will be impossible to say if this is caused by a higher dose or a longer time. Although there is no single clear evidence for it, a cluster of facts indicate that session time does influence outcome. Optimal dialysis needs several conditions, not just one. Time is the common pathway that allows not only the provision of a good dose of dialysis in terms of small and, even more, bigger solutes, but also the provision of satisfactory nutrition and ECV and BP control. This last point is essential; cardiovascular morbidity is by far the first cause of death on dialysis. Cardiovascular mortality is also the only domain in which the long dialysis makes the difference with the short,

CHARRA ET AL

now conventional, dialysis. Whereas increasing the dose reduces all causes of mortality,18 our experience shows that the better ECV and BP control allowed by long hemodialysis reduces specifically the cardiovascular mortality. Increased HD time correlates with better control of ECV and BP, less cardiovascular morbidity and mortality, and longer survival. A physician’s first ethical duty is to give patients the best survival chance, even if the method used is not yet evidence based. Knowledge must not kill wisdom. Therefore, until some way is eventually found to achieve normal ECV and BP within a short dialysis time, it sounds reasonable that dialysis session time be governed by the demand of BP control. REFERENCES 1. Hull AR, Parker TF: Proceedings from the Morbidity, Mortality and Prescription of Dialysis Symposium, Dallas, Texas, September 1989: Introduction and summary. Am J Kidney Dis 15:375-383, 1990 2. Laurent G, Calemard G, Charra B: Long dialysis: A review of fifteen years experience in one center: 1968-1983. Proc Eur Dial Transplant Assoc 20:122-129, 1983 3. Charra B, Calemard E, Laurent G: Importance of treatment time and blood pressure control in achieving long-term survival on dialysis. Am J Nephrol 16:35-44, 1996 4. Daugirdas JT: Second-generation logarithmic estimates of single-pool variable volume Kt/V: An analysis of error. J Am Soc Nephrol 4:1205-1213, 1993 5. Chazot C, Charra B, Laurent G, Didier C, Vo Van C, Terrat JC, Calemard E, Vanel T, Ruffet M: Interdialysis blood pressure control by long hemodialysis sessions. Nephrol Dial Transplant 10:831-837, 1995 6. US Renal Data System: USRDS 1995 Annual Data Report. Bethesda, MD, Department of Health and Human Services, The National Institute of Health, National Institute of Diabetes and Digestive and Kidney Diseases, 1995 7. US Renal Data System: USRDS 1997 Annual Data Report. Bethesda, MD, Department of Health and Human Services, The National Institute of Health, National Institute of Diabetes and Digestive and Kidney Diseases, 1997 8. Australia and New Zealand Dialysis and Transplant Registry (ANZDATA): Nineteenth Report of the Australia and New Zealand Dialysis and Transplant Registry (ANZDATA). Adelaide, Australia, The Queen Elizabeth Hospital, 1996 9. Teraoka S, Toma H, Nihei H, Ota K, Babazono T, Ishikawa I, Shinoda A, Maeda K, Koshikawa S, Takahashi T, Sonoda T: Current status of renal replacement therapy in Japan. Am J Kidney Dis 25:151-164, 1995 10. Wolfe RA, Gaylin DS, Port FK, Held PJ, Wood CL: Using USRDS-generated mortality tables to compare local ESRD mortality rates to national rates. Kidney Int 42:991996, 1992 11. Wolfe RA: The standardized mortality ratio revisited:

TRENDS IN DIALYSIS TIME

Improvements, innovations and limitations. Am J Kidney Dis 24:290-297, 1994 12. Acchiardo SR, Runyan K, Hatten KW, Dyson B, Fuller J, Moore L: Impact of Kt/V and protein catabolic rate on hemodialysis mortality. J Am Soc Nephrol 3:351, 1992 (abstr) 13. Held PJ, Port FK, Wolfe RA, Stannard D, Carroll CE, Daugirdas JT, Bloembergen WE, Greer JW, Hakim RM: The dose of hemodialysis and patient mortality. Kidney Int 50:550-556, 1996 14. Owen WF, Lew NL, Yan Lu SM, Lowrie EG, Lazarus JM: The urea reduction ratio and serum albumin concentration as predictors of mortality in patients undergoing hemodialysis. N Engl J Med 329:1001-1006, 1993 15. Parker TF, Husni L, Huang W, Lew M, Lowrie EG, and the Dallas Nephrology Associates: Survival of hemodialysis patients in United States is improved with a greater quantity of dialysis. Am J Kidney Dis 23:670-680, 1994 16. Hakim RM, Breyer J, Ismail N, Schulman G: Effects of dose of dialysis on morbidity and mortality. Am J Kidney Dis 23:661-669, 1994 17. Collins AJ, Ma J, Umen A, Keshaviah PR: Urea index and other predictors of hemodialysis patient survival. Am J Kidney Dis 23:272-282, 1994 18. Bloembergen WE, Stannard D, 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 19. Valderrabano F: Weekly duration of dialysis treatment—Does it matter for survival? Nephrol Dial Transplant 11:569-572, 1996 20. Shinzato T, Nakai S, Akiba T, Yamasaki C, Sasaki R, Kitaoka T, Kubo K, Shinoda T, Kurokawa K, Marumo F, Sato T, Maeda K: Survival in long-term hemodialysis patients: Results from the annual survey of the Japanese Society for Dialysis Therapy. Nephrol Dial Transplant 11: 2139-2142, 1996 21. Gotch FA, Sargent JA: A mechanistic analysis of the National Cooperative Study (NCDS). Kidney Int 28:526534, 1985 22. Gotch FA, Sargent JA: A theoretical definition of minimal acceptable dialysis therapy. Kidney Int 13:S108S111, 1978 (suppl 8) 23. Kramer P, Broyer M, Brunner FP, Brynger H: Combined report on regular dialysis and transplantation in Europe. Proc Eur Dial Transplant Assoc 19:4-59, 1982 24. Kjellstrand CM: Short dialysis increases morbidity and mortality, in Lysaght MJ, Wetzels E, Gurland HJ (eds): Disputed Issues in Renal Failure Therapy. Basel, Switzerland, Karger, 1985, pp 65-77 25. Wizemann V, Kramer W: Short-term dialysis, longterm complications. Blood Purif 5:193-201, 1987 26. Lowrie EG, Lew NL: Death risk in hemodialysis patients: The predictive value of commonly measured variables and an evaluation of death rate differences between facilities. Am J Kidney Dis 15:458-482, 1990 27. Held PJ, Levin NW, Bovbjerg RR, Pauly MV, Diamond LH: Mortality and duration of hemodialysis treatment. JAMA 265:871-875, 1991 28. Shaldon S: The influence of dialysis time and dialyzer

S69

reuse on survival. Nephrol Dial Transplant 10:57-62, 1995 (suppl 3) 29. Gotch FA, Yarian S, Keen M: A kinetic survey of US hemodialysis prescriptions. Am J Kidney Dis 15:511-515, 1990 30. Letteri JM: Behind the scenes of the NKF-DOQI guidelines. Hemodialysis adequacy. Dial Transplant 26:827831, 1997 31. Basham LM, MacDonald Ager JW: Time spent on dialysis: The paramount consideration in dialysis adequacy? Dial Transplant 26:739-749, 1997 32. Babb AL, Popovich RP, Christopher TG, Scribner BH: The genesis of the square meter-hour hypothesis. Trans Am Soc Artif Int Organs 18:81-86, 1971 33. Cheung AK: Quantitation of dialysis: The importance of membrane and middle molecules. Blood Purif 12:42-53, 1994 34. Abiko T, Onodera T, Sekino HA: A peptide isolated from the hemodialysate of uremic patients with immunodeficiency inhibits lymphocyte stimulation. J Appl Biochem 3:562-569, 1981 35. Shichiri M, Hirata Y, Ando K, Emori T, Ohta K, Kimoto S, Ogura M, Inoue A, Marumo F: Plasma endothelin levels in hypertension and chronic renal failure. Hypertension 15:493-496, 1990 36. Henderson I: Hemodynamic instability during different forms of dialysis therapy: Do we really know why? Blood Purif 14:395-404, 1996 37. Bergstro¨m J, Mamoun H, Anderstam B: Middle molecules isolated from uremic ultrafiltrate and normal urine induce dose-dependent inhibition of appetite in rat. J Am Soc Nephrol 5:488, 1994 (abstr) 38. Massry SG: Parathyroid hormone: A uremic toxin, in Ringoir S, Vanholder R, Massry SG (eds): Uremic Toxins. New York, NY, Plenum, 1987, p 1-17 39. Gejyo F, Yamada T, Odani S, Kunimoto T, Katoka H, Suzuki M, Hirasawa Y, Shirahama T, Cohen AS, Schmid K: A new form of amyloid protein associated with chronic hemodialysis was identified as B2-microglobulin. Biochem Biophys Res Commun 129:701-706, 1985 40. Mactier RA, Madi AM, Allam BF: Comparison of high-efficiency and standard haemodialysis providing equal urea clearance by partial and total dialysate quantification. Nephrol Dial Transplant 12:1182-1186, 1997 41. Pierratos A, Ouwendyk M, Francoeur R, Vas S, Raj DSC, Ecclestone AM, Langos V, Uldall PR: Nocturnal hemodialysis: Three-year experience. J Am Soc Nephrol 9:859-868, 1998 42. Leypoldt JK, Cheung AK: Removal of high-molecular-weight solutes during high-efficiency and high-flux haemodialysis. Nephrol Dial Transplant 11:329-335, 1996 43. Vanherweghem JL, Genevois PA, Dhaene M: Blood urea nitrogen, middle molecules, and uremic morbidity in hemodialysis patients. Dial Transplant 15:86-91, 1986 44. Pierratos A, Uldall PR, Ouwendyk M, Francoeur R, Vas S: Two-year experience with slow nocturnal hemodialysis. J Am Soc Nephrol 7:1417, 1996 (abstr) 45. Berland Y, Brunet P: Morbidity and mortality in hemodialysis: The role of schedules. Artif Organs 19:827831, 1995 46. Seghal AR, Snow RJ, Singer ME, Amini SB, De Oreo

S70

PB, Silver MR, Cebul RD: Barriers to adequate delivery of hemodialysis. Am J Kidney Dis 31:593-601, 1998 47. Kopple JD, Henry DA, Roberts CE, Goodman WG, Blumenkrantz MJ: Relationship between nutritional status of patients undergoing maintenance hemodialysis and duration of dialysis therapy, in Giordano C, Friedman EA (eds): Uremia. Pathobiology of Patients Treated for 10 Years or More. Milano, Wichtig, 1981, pp 25-31 48. Acchiardo SR, Moore LW, Latow PA: Malnutrition as the main factor in morbidity and mortality of dialysis patients. Kidney Int 6:199-203, 1983 (suppl 16) 49. Lindsay RM, Spanner E: A hypothesis: The protein rate is dependent upon the type and amount of treatment in dialysed uremic patients. Am J Kidney Dis 13:382-389, 1989 50. Chazot C, Vo Van C, Charra B, Jean G, Vanel T, Laurent G: Anabolic response after the onset of hemodialysis treatment. J Am Soc Nephrol 7:1476, 1996 (abstr) 51. Foley RN, Parfrey PS, Harnett JD, Kent GM, Murray DC, Barre PE: Impact of hypertension on cardiomyopathy, morbidity and mortality in end-stage renal disease. Kidney Int 49:1379-1385, 1996 52. Soucie JM, Mac Clellan W: Early death in dialysis patients: Risk factors and impact on incidence and mortality rates. J Am Soc Nephrol 6:563, 1995 (abstr) 53. Iseki I, Miyasato F, Tokuyama K, Nishime K, Uehara H, Shiohira Y, Sunagawa H, Yoshihara K, Yoshi S, Toma S, Kowatari T, Wake T, Oura T, Fukiyama K: Low diastolic blood pressure, hypoalbuminemia, and risk of death in a cohort of chronic hemodialysis patients. Kidney Int 51:12121217, 1997 54. Lazarus JM, Lowrie EG, Hampers CL, Hampers JP: Cardiovascular disease in uremic patients on hemodialysis. Kidney Int 7:S165-S175, 1975, (suppl 2) 55. Haire HM, Sherrard DJ, Scardapane D, Curtis FK, Brunzell JD: Smoking, hypertension and mortality in a maintenance dialysis population. Cardiovasc Med 3:11631168, 1978 56. Degoulet P, Legrain M, Reach I, Aime´ F, Devrie`s C, Rojas P, Jacobs C: Mortality risk factors in patients treated by chronic hemodialysis. Nephron 31:103-110, 1982 57. Charra B, Calemard E, Cuche M, Laurent G: Control of hypertension and prolonged survival on maintenance hemodialysis. Nephron 33:96-99, 1983 58. Mailloux LU, Bellucci A, Napolitano B, Mossey RT, Wilkes BM, Bluestone PA: Survival estimates for 683 patients from 1970 through 1989: Identification of risk factors for survival. Clin Nephrol 42:127-135, 1994 59. Tomita J, Kimura G, Inoue T, Takashi I, Sanai T, Nakamura S, Baba S, Matsuoka H, Omae T: Role of systolic blood pressure in determining prognosis of hemodialysis patients. Am J Kidney Dis 25:405-412, 1995 60. Flanigan MJ: Dialysis and hypertension: Worthy of investigation? Semin Dial 11:109-112, 1998 61. Daugherty SA, Ford CE, Curb JD, Langford H,

CHARRA ET AL

Taylor JO, Payne GH, Barhani NO, Pressel S, Hawkins CM: Persistance of reduction of blood pressure and mortality of participants in the Hypertension Detection and Follow-Up Program. JAMA 259:2113-2122, 1988 62. Bulpitt CJ, Beevers DG, Butler A, Coles EC, Fletcher AE, Hunt D, Munro-Faure AD, Newson R, O’Riordan PW, Petrie JC, Rajagopalan B, Rylance PB, Twallin G, Webster J, Dollery CT: Treated blood pressure, rather than pretreatment predicts survival in hypertensive patients. A report from the DHSS hypertension care computing project (DHCCP). J Hypertens 6:627-632, 1988 63. Kjellstrand KM, Evans RL, Petersen RJ, Shideman JR, von Hartitzsch B, Buselmeier TJ: The ‘‘unphysiology’’ of dialysis: A major cause of dialysis side effects? Kidney Int 10:S30-S34, 1975, (suppl 7) 64. Mailloux LU, Bellucci AG, Napolitano B, Mossey RT: The contribution of hypertension to dialysis patients outcomes, a point of view. Am Soc Artif Int Organs J 40:130-137, 1994 65. Charra B: Does empirical long slow dialysis result in better survival? If so, how and why? Am Soc Artif Int Organs J 47:819-822, 1993 66. Cohen EP: Dialysis hypertension: Dry weight and dialysis time. Nephrol Dial Transplant 13:554-555, 1998 ¨ zkahya M, Ok E, Cirit M, Akc¸ic¸ek E, Basc¸i A, 67. O Dorhout Mees EJ: Regression of left ventricular hypertrophy in hemodialysis patients by ultrafiltration and reduced salt intake without antihypertensive drugs. Nephrol Dial Transplant 13:1489-1493, 1998 68. Krautzig S, Kielstein JT, Granoleras C, Shaldon S, Lonneman G, Koch KM: Control of hypertension in HD patients by reduction of dialysate sodium and low salt diet. J Am Soc Nephrol 8:243A, 1997 (abstr) 69. Hegstro¨m RM, Murray JS, Pendras JP, Burnell JM, Scribner BH: Hemodialysis in the treatment of chronic uremia. Trans Am Soc Artif Int Organs 7:136-152, 1961 70. Comty C, Baillod RM, Crockett R, Shaldon S: Forty months experience with a nurse-patient operated chronic dialysis unit. Proc Eur Dial Transplant Assoc 3:98-101, 1966 71. Vertes V, Cangiano JL, Berman LB, Gould A: Hypertension in end-stage renal disease. N Engl J Med 280:978981, 1969 72. Goldsmith DJA, Covic AA, Venning MC, Ackrill P: Ambulatory blood pressure monitoring in renal dialysis and transplant patients. Am J Kidney Dis 29:593-600, 1997 73. Buttimore AL, Lynn KL, Bailey RR, Robson RA, Little P: 25 years of universal home dialysis at Christchuch Hospital. Nephrology 3:S559, 1997 (abstr; suppl 1) 74. Charra B: Control of blood pressure in long slow hemodialysis. Blood Purif 12:252-258, 1994 75. Eknoyan G, Levey AS, Beck GJ, Agodoa LY, Daugirdas JT, Kusek JW, Levin NW, Schulman G: The hemodialysis (HEMO) study: Rationale for selection of interventions. Semin Dial 9:24-33, 1996