Impact of Renal Replacement Therapy on Severe Lactic Acidosis in ICU Patients: A MIMIC-IV retrospective replication analysis

INTRODUCTION: Severe lactic acidosis (SLA), defined as lactate ≥5 mmol/L, is a critical condition in intensive care unit (ICU) patients and is associated with high mortality. Despite the broad use of renal replacement therapy (RRT) to manage metabolic derangements, its impact on outcomes in SLA remains uncertain. Implementing a replication analysis, we aimed to validate the previous findings by Van de Ginste et al. regarding the association between RRT initiation and patient outcomes for patients with SLA in ICU. METHODS: We conducted a retrospective cohort analysis using the MIMIC-IV database (2008-2019). Adult ICU patients with severe lactic acidosis (lactate ≥5 mmol/L) during their first ICU stay were included. Patients on maintenance dialysis, with prior RRT, or Do-Not-Resuscitate status were excluded. RRT was considered if initiated within 24 h of reaching the lactate threshold. Logistic regression was used to assess the association between RRT and 90-day mortality. RESULTS: Of 5317 screened patients, 4450 met inclusion criteria; 314 (7.1%) received early RRT. Overall 90-day mortality was 51.3%. The propensity to receive RRT was related to the lactate level and to the SOFA renal score. Higher age, serum lactate, lower pH, and greater organ dysfunction were associated with mortality. After adjustment, RRT remained associated with increased 90-day mortality (OR 1.51; 95% CI 1.05 - 2.20). DISCUSSION: Replicating previous work using a large, independent database (MIMIC-IV), our findings confirm that early initiation of RRT in patients with severe lactic acidosis is associated with poor prognosis, even after adjustment for illness severity. These results reinforce the conclusions that RRT may not mitigate the adverse outcomes of severe lactic acidosis. The results thus highlight the necessity for prospective studies to clarify the role of RRT in this critical patient population.