SCCM Resource Library-So When Do I Start RRT in the Critically Ill Adult Patient?

Video Transcription

Hello, my name is Sean Bagshaw. I'm a professor of critical care medicine at the University of Alberta in Edmonton, Canada. It's my pleasure today to talk to you about  when I start re-replacement therapy in critically ill patients. I'd like to thank the SCCM for the invitation and opportunity to participate in this year's meeting.  Unfortunately, it has now gone virtual. I have these disclosures for 2021-22.  Let's first begin with when to start re-replacement therapy. Existing clinical practice guidelines, of which the three most prominent are listed, are now more than 10 years old  and do not make strong recommendations on when to start re-replacement therapy, likely due to a relatively weak and uncertain evidence base at the time of their publication.  The statements included, though, are all clinically sensible. And there's general agreement for re-replacement therapy initiation when confronted with urgent indications.  However, new trial evidence has been published and it's likely that these clinical practice guidelines deserve to be updated.  So let's review the rationale for re-replacement therapy in ICU settings. There are certainly benefits and risks. Re-replacement therapy will effectively address  the first three benefits noted here. It is edema control, metabolic and pH homeostasis, fluid balance homeostasis. So part of the question is whether addressing  any or all of these in a given patient, plus the purported systemic effects, will translate into improved patient outcomes and efficient resources.  What do we know about clinicians and their decision-making around starting re-replacement therapy? Well, there have been several surveys  and observational studies of clinical practice trying to address this. And I would argue that the decision remains largely subjective, with clinicians doing their best  to integrate available bedside information, so patient's physiologic reserve, the trajectory of illness, the evolving indications,  with the existing evidence base and resource availability. I would, however, submit that the decision about escalation to providing re-replacement therapy,  like all decisions to expand organ support in ICU settings, is susceptible to clinician bias.  Some general statements to this would include that clinicians report starting when confronted with life-threatening complications, but there's wide variation in the minimum severity  of indications that would prompt the initiation of re-replacement therapy, and there's many factors that appear to modify that decision. Patient age, baseline comorbidity,  the responsiveness to a diuretic challenge, illness acuity, the consultant service, and re-replacement therapy logistics, such as the time of day and the day of the week.  What about the existing evidence base? Well, this individual patient meta-analysis that was led by Stephane Gaudry and colleagues, pooled the available data from 10 RCTs.  This was before the publication of the main START-AKI trial, and included 2,143 patients, and found no differences in short-term mortality  between early and delayed strategies for RRT initiation. However, also showed that fewer patients in the delayed strategy received uterine replacement therapy.  However, I would argue there's some residual uncertainty around these trials, driven by differences in design, settings, case mix, eligibility criteria,  and in particular, definitions for timing of the intervention, early versus delayed. As such, this has continued to contribute to some clinical uncertainty and practice variations.  We hope that the multinational START-AKI trial, which randomized 2,927 patients to strategies of accelerated and standard RRT initiation,  would help resolve some of the lingering uncertainty for clinicians, and work towards greater harmonization of practice. As you may know, the key findings from START-AKI were,  there was no difference in 90-day all-cause mortality, the primary endpoint, and this was a robust finding in pre-specified adjusted, subgroup,  and sensitivity analyses that explored as well for heterogeneity and treatment effect. Number two, there was greater RRT dependence at 90 days  amongst those in the accelerated strategy, and again, this was robust in pre-specified inverse probability and multinomial regression analyses.  Number three, there was a greater incidence of adverse events in the accelerated strategy, and this appeared to be driven by more episodes of hemodynamic instability.  And finally, number four, 38% of patients in the standard strategy did not receive any renal replacement therapy. As such, START-AKI appeared to reinforce  the acceptability of a watch-and-wait strategy, whereby RRT would be initiated in response to conventional indications and clinician judgment.  However, there arose uncertainty about how long could renal replacement therapy be safely delayed, and could a strategy of even more delayed RRT be acceptable, improve outcomes,  and reduce RRT utilization?  So this was the primary question asked by the multinational French, or sort of multicenter French, AKIKI-2 trial, where 278 patients were randomized  to either delayed or more delayed RRT initiation strategies. The primary endpoint was RRT-free days, which was interestingly not found  to be significantly different between the two groups. In a pre-planned adjusted analysis, the more delayed strategy was associated with an absolute 11% increase in mortality,  arguably a clinically important difference, and a greater adjusted hazards of death at 60 days, along with SAPS-2 score and basic mechanical ventilation.  So these observations provide insights into the notion that excessive delay to renal replacement therapy among patients with non-resolving AKI may in fact be harmful,  potentially due to the prolonged exposure to azotemia, fluid accumulation, or interventions used to mitigate the complications of AKI.  As such, I would suggest there remains some uncertainty in when to commence renal replacement therapy. This was certainly evident in the START-AKI trial.  We designed START-AKI to be pragmatic and real-world. As such, we sought clinician equipoise as part of the eligibility criteria,  recognizing and acknowledging that clinical uncertainty many clinicians are confronted with when deciding to initiate renal replacement therapy.  However, we also noted that a significant number of patients who were eligible for START-AKI were excluded due to the lack of individual clinician equipoise.  Indeed, this represented about 89% of all exclusions of patients who had fulfilled provisional eligibility.  Of these 7,886 patients, 72% were excluded due to clinicians mandating deferral of renal replacement therapy. Data were captured on a subset of these patients,  about 50%, and we showed that relatively few went on to receive renal replacement therapy, only 11%. Alternatively, the remaining 2,196 patients were excluded  due to clinicians mandating immediate renal replacement therapy. Data were captured on about a third of these patients, and it showed that 91% went on  to receive renal replacement therapy. So these observations would imply relatively wide uncertainty for predicting the clinical course of AKI  and the likelihood of either recovery and or worsening AKI that would prompt renal replacement therapy initiation.  So do we have new tools on the horizon that may aid in decision-making for clinicians on when to initiate renal replacement therapy?  Well, recently, this prospective observational cohort study of 364 critically ill patients, called the RUBY study, enrolled within 36 hours of developing stage 2-3 AKI,  discovered and described a novel biomarker, urinary CC motif chemokine ligand 14, or CCL14, to predict the occurrence of persistent severe AKI,  defined as stage 3 AKI occurring for more than 72 hours.  Urinary CCL14 was found to have the best prediction for persistent severe AKI amongst all biomarkers assessed with an ROC of 0.83. Urinary CCL14 was also found to enhance  the predictive ability over a baseline clinical model in the logistic regression that included variables like BMI, diabetes, non-renal APACHE2 score, AKI stage,  and the trajectory of serum creatinine to an ROC of 0.89. Finally, the occurrence of secondary endpoints, such as renal replacement therapy or death,  was greater with increasing tertiles of urinary CCL14. As such, the RUBY study provides preliminary evidence at least that urinary CCL14 may provide novel diagnostic  and prognostic information to clinicians to aid in clinical decision-making, i.e., timing of initiation. However, these findings necessitated external validation.  And we recently reported this in a secondary analysis of 195 patients with stage 2, 3 AKI enrolled in the multinational SAFIRE study. Again, the primary endpoint here  was persistent severe AKI, similarly defined, and the secondary endpoints were a composite of RT initiation and or death.  We found that the ROC for urinary CCL14 to predict persistent severe AKI in the SAFIRE cohort was 0.81, which was pretty similar to that found in the RUBY study,  as shown in the figure on the left. You can also see in the figure on the right that we further confirmed that higher tertiles of urinary CCL14 portended greater risk  for the secondary composite endpoint of RT initiation or death at 90 days.  So we recently published this algorithm to provide clinicians with some guidance around decision-making for when to initiate renal placement therapy  for patients who develop moderate to severe AKI while in the ICU. Naturally, in the presence of potentially life-threatening or medically refractory complications of AKI,  initiation of renal placement therapy may seem prudent. But in the absence of those, given the totality of evidence from the individual patient meta-analysis and start AKI,  we would suggest a watch-and-wait strategy may be acceptable, where you optimize systemic hemodynamics and you avoid secondary kidney insults, et cetera.  Of course, at present, we're not certain of the precise role of clinical prediction models, novel biomarkers like CCL14, or other dynamic tests for informing decisions  about the probability of greater benefit relative to risk of starting renal placement therapy in a given patient. Certainly, clinician judgment on this will always play a role.  This will likely be the focus of further clinical trials in the future.  So some final thoughts. The decision to initiate renal placement therapy is complex and represents a significant escalation in support, can be influenced by a host of factors.  Some are patient-specific, but many are clinician and health system related. There has been wide variation in clinical practice driven by further variation  in individual clinician echopoies, certainly as shown in start AKI. Recent large randomized trials have suggested no benefit to an early or accelerated approach.  However, that would reinforce that a watch-and-wait approach would be clinically acceptable. And future clinical trials should integrate additional measures  of clinical decision support, such as novel biomarkers like CCL14.  And with that, I'll thank you for your attention, and I hope we have some opportunities for dialogue and question.

Video Summary

In this video, Dr. Sean Bagshaw discusses the timing of renal replacement therapy (RRT) initiation in critically ill patients. He mentions that existing clinical practice guidelines are outdated and do not provide strong recommendations on when to start RRT. The decision to initiate RRT is subjective and based on a clinician's integration of available bedside information, evidence base, and resource availability. Dr. Bagshaw discusses the results of recent trials, including the START-AKI trial, which found no difference in mortality between early and delayed strategies for RRT initiation. He also mentions the AKIKI-2 trial, which showed that excessive delay in RRT initiation may be harmful. Dr. Bagshaw suggests that there is still uncertainty regarding the timing of RRT initiation and that new tools, such as biomarkers like urinary CCL14, may aid clinicians in decision-making. He concludes by emphasizing the complexity of the decision and the need for further research and clinical trials.

Asset Subtitle

Renal, 2022

Asset Caption

Extensive effort has been expended to determine the association between the timing of renal replacement therapy (RRT) initiation and patient outcome. Landmark trials have been published during the past few years, studying thousands of patients from all over the world. What have we learned? What will change? What lessons are applicable and what require more attention? This session will discuss the headlines, nuances, and lessons that are appropriate to extrapolate, as well as those that may not be, as the use of RRT expands in all areas of critical care.

Meta Tag

Content Type Presentation

Knowledge Area Renal

Knowledge Level Intermediate

Knowledge Level Advanced

Membership Level Select

Tag Renal Replacement Therapy

Year 2022

Keywords

renal replacement therapy

RRT initiation

clinical practice guidelines

timing of RRT initiation

biomarkers

	Society of Critical Care Medicine 500 Midway Drive Mount Prospect, IL 60056 USA Phone: +1 847 827-6888 Fax: +1 847 439-7226 Email: support@sccm.org	Contact Us About SCCM Newsroom Advertising & Sponsorship DONATE	MySCCM LearnICU Patients & Families Surviving Sepsis Campaign Critical Care Societies Collaborative	GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. \| Privacy Statement \| Terms & Conditions The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.