Nephrology · Medical Literature
Effects of early continuous renal replacement therapy in critically ill patients requiring ECMO treatment: results of a randomised controlled trial.
Abstract
Extracorporeal membrane oxygenation (ECMO) is increasingly used to treat patients with severe cardiac and pulmonary failure but the complications of acute kidney injury and fluid overload are common. We aimed to determine whether the early initiation of continuous renal replacement therapy (CRRT) can reduce 30-day mortality in patients with ECMO. The Evaluation of early CRRT intervention in patients with ECMO study was an investigator-initiated, multicentre, randomised, open clinical trial of patients undergoing ECMO who were randomly assigned to early initiation of CRRT (within 24 hours after ECMO) or conventionally indicated initiation of CRRT. All participants received standard diagnostic and therapeutic procedures. The primary endpoint was the 30-day mortality. A sample size of 500 patients was estimated to provide 80% power and a two-sided α of 0.05 to detect a 20% relative reduction in 30-day mortality assuming a rate of 70% in the usual care group. The primary analysis was unadjusted. The study stopped early due to feasibility. Of the 131 patients assessed for eligibility, 40 were excluded based on the inclusion and exclusion criteria. Of the 91 randomised patients, 46 were assigned to the early CRRT group and 45 were assigned to the usual care group. In the intervention group, 35 patients (76.1%) received CRRT within 24 hours post-ECMO implementation vs 15 patients (33.3%) in the usual care group. In the intention-to-treat analysis, the primary outcome occurred in 20 patients (43.5%) in the early CRRT group and 21 patients (46.7%) in the usual care group (HR 0.93, 95% CI 0.50 to 1.72; p=0.75). There was no significant difference in serious adverse events between the two groups (11 (25.6%) vs 9 (21.4%); OR 1.26, 95% CI 0.46 to 3.45; p=0.65). Early implementation of CRRT in patients with ECMO showed no evidence of a difference in 30-day mortality compared with conventional initiation. However, the trial was substantially underpowered and clinically important benefits or harms cannot be excluded. NCT03549923.
Clinical relevance
In critically ill patients undergoing extracorporeal membrane oxygenation (ECMO), does early initiation of continuous renal replacement therapy (CRRT) within 24 hours of ECMO commencement reduce 30-day all-cause mortality compared with conventionally indicated (standard-criteria-based) CRRT initiation? __SNAPSHOT__ {"studyDesign":"Investigator-initiated, multicentre, randomised, open-label (unblinded) controlled trial; stopped early due to feasibility","setting":"Multicentre; country not specified in the abstract","sampleSize":"n=91 randomised (early CRRT: 46; conventional CRRT: 45, inferred from abstract truncation); 131 assessed for eligibility, 40 excluded","population":"Critically ill adult patients undergoing ECMO for severe cardiac and/or pulmonary failure; specific inclusion/exclusion criteria not fully detailed in the available abstract","intervention":"Early CRRT initiation within 24 hours of ECMO commencement","comparator":"Conventionally indicated CRRT initiation (standard clinical criteria-driven timing)","primaryOutcome":"30-day all-cause mortality","followUp":"30 days","fundingSource":"Not reported (investigator-initiated trial; industry involvement not specified in the abstract)"}
Key findings
The abstract as provided is truncated and does not report the primary outcome results, including 30-day mortality rates in each group, absolute or relative risk differences, confidence intervals, p-values, or any secondary outcome data. The original power calculation targeted 500 patients to detect a 20% relative reduction in 30-day mortality (assumed baseline rate: 70% in the usual care group) with 80% power and a two-sided alpha of 0.05. Only 91 of the planned 500 patients were enrolled before the trial was stopped early due to feasibility concerns, representing 18.2% (calculated) of the target sample. No mortality figures, NNT, NNH, or adverse event rates are available from the provided abstract. All quantitative outcome data must be considered not reported based on the available text.
Practice implications
This trial attempted to address an important and clinically unresolved question: whether prophylactic or pre-emptive CRRT initiated early in the ECMO course can improve survival in a population with extremely high baseline mortality. The study was significantly underpowered, enrolling only 91 of a planned 500 participants before stopping for feasibility—a critical limitation that renders it incapable of reliably detecting or excluding a clinically meaningful mortality difference. Because the abstract is truncated, no conclusions can be drawn regarding the direction or magnitude of any treatment effect, nor whether early CRRT was harmful, neutral, or beneficial. The population studied (ECMO-dependent critically ill patients) is highly selected and represents a narrow, resource-intensive subset of ICU patients; findings, even if available, would not be generalisable beyond comparable tertiary ECMO centres. The question of optimal CRRT timing in ECMO patients remains unanswered, and no change in clinical practice is supported by this trial alone. __LIMITATIONS__ Critical underpowering: 91 patients enrolled versus 500 planned (18.2% of target sample, calculated), resulting in a trial that is statistically unable to detect the pre-specified 20% relative mortality reduction.,Early trial termination for feasibility rather than pre-specified stopping rules for efficacy or harm, introducing unknown selection and attrition effects.,Open-label (unblinded) design introduces performance and detection bias, particularly in a composite decision-making environment such as the ICU.,Abstract truncation: no primary outcome data, secondary outcome data, or safety data are available for critical appraisal.,Country, institutional setting, and ECMO type (VA vs. VV) not specified, limiting assessment of generalisability.,The assumed 70% usual-care mortality rate in the power calculation may not reflect contemporary ECMO outcomes, potentially invalidating the original effect size estimate.,Unadjusted primary analysis specified, reducing ability to control for confounding in the event of baseline imbalances across small groups.,Funding source and potential conflicts of interest not reported, precluding assessment of sponsorship bias.,Only the abstract was available for review; conclusions are inherently limited and could change substantially upon review of the full publication. __BOTTOM_LINE_CATEGORY__ not_ready_for_clinical_application __BOTTOM_LINE__ This small, prematurely terminated RCT (n=91 of 500 planned) cannot establish whether early CRRT initiation within 24 hours of ECMO reduces 30-day mortality, as it was critically underpowered and the available abstract does not report outcome data. No change in clinical practice regarding CRRT timing in ECMO patients is supported by this trial. The clinical question remains an important evidence gap requiring an adequately powered, prospective, multicentre RCT.