Debate: Intermittent Hemodialysis versus Continuous Kidney Replacement Therapy in the Critically Ill Patient: Moderator Commentary : Clinical Journal of the American Society of Nephrology

Journal Logo

Critical Care Nephrology and Acute Kidney Injury

Debate: Intermittent Hemodialysis versus Continuous Kidney Replacement Therapy in the Critically Ill Patient: Moderator Commentary

Palevsky, Paul M.

Author Information
Clinical Journal of the American Society of Nephrology 18(5):p 644-646, May 2023. | DOI: 10.2215/CJN.0000000000000116


It has been more than four decades since the initial description of continuous kidney replacement therapy (CKRT). It was initially developed as continuous arteriovenous hemofiltration, a technologically simple therapy in which blood flow through an extracorporeal circuit was driven by the pressure gradient between the arterial and venous circulations with no pumps or pressure monitors, an ultrafiltrate was generated driven by the transmembrane pressure, and excess ultrafiltrate replaced intravenously.1 Prior to the introduction of this therapy, the options for acute dialysis had been limited to intermittent hemodialysis or acute peritoneal dialysis, with the former often difficult to perform in hemodynamically unstable patients. Thus, this novel modality allowed provision of therapy to patients who were often deemed too hemodynamically unstable to tolerate conventional hemodialysis.

Over the ensuing decades, both intermittent hemodialysis and CKRT have undergone substantial transformations. The bioincompatible cellulosic hemodialysis membranes that were standard have been replaced by much less bioreactive synthetic membranes; acetate has been replaced by bicarbonate as the standard buffer for dialysate; the biologic purity of dialysate has markedly increased; and ultrafiltration is now directly controlled rather than estimated on the basis of the ultrafiltration coefficient of the dialyzer and adjustments to the transmembrane pressure. Consequently, the ability to provide conventional hemodialysis to hemodynamically unstable patients has improved. At the same time, the continuous therapies have evolved from the initial technologically simple arteriovenous circuits, initially to rudimentary venovenous circuits cobbled together with separate blood pumps, air detectors, and pressure monitors, and ultimately to today's sophisticated microprocessor-driven integrated systems. In parallel, the fluids for continuous therapy have changed from admixtures of various off-the-shelf intravenous fluids or peritoneal dialysate with supraphysiologic dextrose concentrations used in the early decades of CKRT to the current panoply of commercially manufactured fit-for-purpose bicarbonate-buffered fluids. There has also been the development of hybrid therapies, either using the equipment for intermittent therapy, with blood and dialysate flow rates slowed down and treatment durations extended from 8 to 16 hours, or using the technology for continuous therapies, with augmented ultrafiltration and/or dialysate flow rates for treatments of 10–12 hours per day, all broadly included under the rubric prolonged intermittent KRT.

During these four decades, the management of critically ill patients has also undergone striking transformation, with improvement in outcomes for patients despite increased acuity of illness and increasing availability of other mechanical support for other organs, including more sophisticated mechanical ventilation, ventricular assist devices, and extracorporeal membrane oxygenation.

Not surprisingly, the availability of more options to provide KRT has led to inevitable debate over which, if any, is the best modality. The Kidney Disease Improving Global Outcomes (KDIGO) Clinical Practice Guideline for Acute Kidney Injury, written a decade ago, suggested that intermittent and continuous KRT should be used as complementary therapies with no single modality—conventional intermittent, prolonged intermittent, or continuous—as ideal for all patients.2 The guidelines go on to suggest that continuous therapy be used preferentially in patients with acute brain injury or other causes of increased intracranial pressure. Additionally, they recommend use of CKRT in patients with hemodynamic instability, noting that once hemodynamic stability is achieved, switching to standard intermittent hemodialysis is reasonable.2

Why then is there a debate if the choice of modality is not an either-or question? While the KDIGO guidelines carve a middle ground and are clear that these suggestions are generally based on limited data or expert opinion, there are some who vocally advocate for only using continuous therapy in the intensive care unit setting. This is exemplified by the design of several randomized clinical trials focused on timing and intensity of KRT.3–6 The Randomized Evaluation of Normal versus Augmented Level Replacement Therapy Study conducted by the Australian and New Zealand Intensive Care Society Clinical Trials Group is the largest clinical trial assessing intensity of KRT in AKI.4 In setting forth the rationale for the study, the investigators criticized the design of the Acute Renal Failure Trial Network Study, which allowed patients to switch between the modality of KRT on the basis of hemodynamic status as the justification for restricting KRT to continuous hemodiafiltration and prohibiting the use of intermittent hemodialysis.3 The Randomized Evaluation of Normal versus Augmented Level study investigators have continued this criticism, most recently publishing a comparison of recovery of kidney function on the basis of the initial modality of KRT across the two trials, despite the fact that all the patients who were initially treated with intermittent hemodialysis were hemodynamically stable Acute Renal Failure Trial Network Study subjects, claiming an association between initial use of intermittent hemodialysis and impaired recovery of kidney function in what is essentially a cross-study comparison.7 Similarly, trials of timing of KRT in AKI have varied regarding utilization of different modalities of KRT. In some of the commentaries after publication of the Early versus Late Initiation of Renal Replacement Therapy in Critically Ill Patients with Acute Kidney Injury and Artificial Kidney Initiation in Kidney Injury (AKIKI) trials,5,6 which differed in whether they exclusively used CKRT, the AKIKI trial was criticized for the use of intermittent hemodialysis in approximately half of patients who received KRT.8 Thus, it is apparent that a cadre of practitioners and investigators feel strongly that only CKRT should be used in the intensive care unit setting.

Given this continued controversy, as part of our ongoing series in CJASN on Critical Care Nephrology and Acute Kidney Injury, we provide a virtual debate on modality of therapy. The proponents arguing for the exclusive use of continuous therapies are Sean Bagshaw, Javier Neyra, Ashita Tolwani, and Ron Wald,9 while Khalil Chaibi, Didier Dreyfuss, and Stéphane Gaudry make the case for using intermittent hemodialysis in the critical care setting.10

The key argument made in favor of CKRT is that it can be applied with greater flexibility and precision than intermittent hemodialysis. Essentially, that the continuous nature of CKRT allows for slow but steady clearance of solute and metabolic waste, gradual removal of excess volume, and gradual correction of electrolyte and acid-base derangements while minimizing hemodynamic stress. The counter argument is that, with the exception of specific settings (e.g., patients with brain injury or fulminant hepatic failure with intracranial hypertension, where the slow, continuous approach is preferred), intermittent and continuous therapies provide equivalent clinical outcomes while allowing earlier mobilization of patients out of bed and potentially providing greater operational efficiency by allowing a single dialysis machine to be used to treat multiple patients each day. Even Bagshaw and colleagues concede that in some settings, the speed of the hare is desirable, for example, that the rapidity of clearance during intermittent hemodialysis is preferred in managing AKI complicated by metformin-associated lactic acidosis or other drug intoxications.

Both sides agree that current data do not suggest that there is better survival with one modality or the other. Where they disagree is whether continuous therapy is associated with better recovery of kidney function than intermittent hemodialysis. Data from randomized clinical trials demonstrate similar survival and recovery of kidney function, regardless of the modality of KRT,11–13 although Bagshaw and colleagues discount these data on the basis of perceived flaws in study design. Most of their criticisms regarding generalizability are true of all clinical trials, including those of timing of therapy led by both sets of authors. Their criticism that the overall number of patients enrolled across all the randomized trials of modality is relatively limited is appropriate, as is the argument that the focus of these trials should be on the actual population of patients for which there is equipoise in practice. It is also important to recognize that protocols for provision of hemodialysis to patients with hemodynamic instability used in some of the studies modified the delivery of intermittent hemodialysis to minimize hemodynamic stress.11 Thus, merely stating that intermittent hemodialysis and continuous therapies provided equivalent outcomes is misleading if it is not acknowledged that similar outcomes may not be achievable with the short, rapid treatments that characterize most inpatient hemodialysis in the United States. Treatment duration will need to be extended from 3–4 hours to 5–5½ hours to achieve similar outcomes.

To make their argument that intermittent hemodialysis is associated with impaired recovery of kidney function as compared with continuous therapies, Bagshaw and colleagues focus on data from observational series.14 In these series, clinical characteristics among patients treated with intermittent hemodialysis as compared with continuous therapies were substantially different, and intermittent hemodialysis was associated with better survival rates but with lower rates of dialysis independence among surviving patients. However, as pointed out by Chaibi, Dreyfuss, and Gaudry, even after statistical adjustment, these observational data are subject to residual confounding. The two sides also disagree on relative costs because these are highly dependent on baseline assumptions, the specifics of hospital care-delivery models, and assumptions regarding outcomes. Unfortunately, considerations of cost are often of paramount importance, particularly in resource-constrained settings, and may dominate other factors in determining availability of therapies.

In the end, both sides of this debate agree that our goal must be to provide the best care possible to patients with AKI and that additional trials will be necessary to optimally inform our approaches to care. However, designing and conducting such trials will be challenging. What is the appropriate study population? Where is the clinical equipoise? Are there some patients too hemodynamically unstable for intermittent hemodialysis or some for whom continuous therapy would not be realistically considered? Does the timing of initiation of therapy alter the effect of modality on outcomes? What are the critical outcomes?

As stated by Bagshaw and colleagues, “positioning this debate as ‘for or against’ and ‘all or nothing’ is overly simplistic.”9 Both sides in this debate would certainly agree that for each patient, we should be asking how to best provide support for their acutely failing kidneys. Rather than blanket dictums regarding the use of one or the other modality, we should focus on whether and when KRT should be started and, when it is to be started, incorporating both patient and health system considerations into the decision making of which modality to use. As we have learned from the clinical trials of timing and intensity of therapy, we need less dogma and more pragmatism.


P.M. Palevsky reports employment with the US Department of Veterans Affairs, VA Pittsburgh Healthcare System, and University of Pittsburgh; and consultancy agreements with Janssen Research & Development, LLC. P.M. Palevsky reports advisory or leadership roles as National Kidney Foundation: Immediate Past President, Member, Scientific Advisory Board; Renal Physicians Association: Member, Quality, Safety and Accountability Committee; Quality Insights Renal Network 4: Chair, Medical Review Board; UpToDate: Section Editor, Acute Kidney Injury; CJASN: Member, Editorial Board; and Journal of Intensive Care Medicine: Member, Editorial Board.



Author Contributions

Writing – original draft: Paul M. Palevsky.

Writing – review & editing: Paul M. Palevsky.


1. Kramer P, Wigger W, Rieger J, Matthaei D, Scheler F. Arteriovenous haemofiltration: a new and simple method for treatment of over-hydrated patients resistant to diuretics [in German]. Klin Wochenschr. 1977;55(22):1121–1122. doi:10.1007/bf01477940
2. Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group. KDIGO clinical practice guideline for acute kidney injury. Kidney Int Suppl. 2012;2(1):3. doi:10.1038/kisup.2012.3
3. Palevsky PM, Zhang JH, O'Connor TZ, et al. Intensity of renal support in critically ill patients with acute kidney injury. N Engl J Med. 2008;359(1):7–20. doi:10.1056/NEJMoa0802639
4. Bellomo R, Cass A, Cole L, et al. Intensity of continuous renal-replacement therapy in critically ill patients. N Engl J Med. 2009;361(17):1627–1638. doi:10.1056/NEJMoa0902413
5. Zarbock A, Kellum JA, Schmidt C, et al. Effect of early vs delayed initiation of renal replacement therapy on mortality in critically ill patients with acute kidney injury: the ELAIN randomized clinical trial. JAMA. 2016;315(20):2190–2199. doi:10.1001/jama.2016.5828
6. Gaudry S, Hajage D, Schortgen F, et al. Initiation strategies for renal-replacement therapy in the intensive care unit. N Engl J Med. 2016;375(2):122–133. doi:10.1056/nejmoa1603017
7. Naorungroj T, Neto AS, Wang A, Gallagher M, Bellomo R. Renal outcomes according to renal replacement therapy modality and treatment protocol in the ATN and RENAL trials. Crit Care. 2022;26(1):269. doi:10.1186/s13054-022-04151-5
8. Romagnoli S, Ricci Z. When to start a renal replacement therapy in acute kidney injury (AKI) patients: many irons in the fire. Ann Transl Med. 2016;4(18):355. doi:10.21037/atm.2016.08.55
9. Bagshaw SM, Neyra JA, Tolwani AJ, Wald R. Debate: intermittent hemodialysis versus continuous kidney replacement therapy in the critically ill patient: the argument for CKRT. Clin J Am Soc Nephrol. 2023;18(5):647–660. doi:10.2215/CJN.0000000000000056
10. Chaibi K, Dreyfuss D, Gaudry S. Intermittent hemodialysis versus continuous kidney replacement therapy in the critically ill patient: the choice should be evidence based. Clin J Am Soc Nephrol. 2023;18(5):661–667. doi:10.2215/CJN.0000000000000104
11. Vinsonneau C, Camus C, Combes A, et al. Continuous venovenous haemodiafiltration versus intermittent haemodialysis for acute renal failure in patients with multiple-organ dysfunction syndrome: a multicentre randomised trial. Lancet. 2006;368(9533):379–385. doi:10.1016/s0140-6736(06)69111-3
12. Bagshaw SM, Berthiaume LR, Delaney A, Bellomo R. Continuous versus intermittent renal replacement therapy for critically ill patients with acute kidney injury: a meta-analysis. Crit Care Med. 2008;36(2):610–617. doi:10.1097/01.ccm.0b013e3181611f552
13. Nash DM, Przech S, Wald R, O'Reilly D. Systematic review and meta-analysis of renal replacement therapy modalities for acute kidney injury in the intensive care unit. J Crit Care. 2017;41:138–144. doi:10.1016/j.jcrc.2017.05.002
14. Schneider AG, Bagshaw SM. Effects of renal replacement therapy on renal recovery after acute kidney injury. Nephron Clin Pract. 2014;127(1-4):35–41. doi:10.1159/000363671

acute kidney injury; dialysis; hemodialysis; Critical Care Nephrology and Acute Kidney Injury Series

Copyright © 2023 by the American Society of Nephrology