Abstract
Background: Thrombocytopenia in patients treated with continuous renal replacement therapy (CRRT) in adults is associated with mortality. Pediatric data are limited. We evaluated the association between pre-CRRT thrombocytopenia and platelet decline at 24 hours of CRRT with outcomes. Methods: Secondary analysis of the Worldwide Exploration of Renal Replacement Outcomes Collaborative in Kidney Disease (WE-ROCK), including patients birth-25 years who underwent CRRT. Exclusions were end-stage kidney disease, non-acute kidney injury/fluid overload CRRT indication, concurrent extracorporeal membrane oxygenation, missing baseline platelets, platelet disorders, and hematologic malignancy. Primary exposures were (1) pre-CRRT thrombocytopenia, (≤100×103/μL) and (2) ≥30% decline at 24 hours of CRRT in those with pre-CRRT >100×103/μL. Primary outcome was survival to intensive care unit (ICU) discharge. Secondary outcomes included major adverse kidney events at 90 days (MAKE90) [death, dialysis dependence, creatinine >125% baseline]) from CRRT initiation. Results: 805 patients were included. 63.9% had baseline thrombocytopenia, median (IQR) platelets of 38 (20, 63) ×103/μL. Baseline thrombocytopenia occurred in younger septic patients with higher illness severity. A ≥30% decline occurred in 33% of patients. Those with a ≥30% platelet decline were more commonly younger patients and had smaller catheters. Pre-CRRT thrombocytopenia and platelet decline were associated with ICU mortality in univariate but not multivariate models. There was no association with MAKE90. Conclusions: Thrombocytopenia is common prior to CRRT initiation and is associated with greater illness severity. Mechanical factors (blood flow and line size) may contribute to platelet decline. Prospective studies are needed to delineate the clinical and mechanical factors associated with thrombocytopenia and associated outcomes.
