Cumulative operating time and early transmembrane pressure surge in relation to mortality during continuous renal replacement therapy

Abstract During continuous renal replacement therapy (CRRT), transmembrane pressure (TMP) is monitored as a circuit-status indicator, but whether it carries prognostic meaning is unclear. This retrospective cohort study examined two circuit-derived measures. Cumulative operating time, the confirmatory analysis, was modeled as a time-varying exposure in MIMIC-IV and again in the multicenter eICU database, which stores no circuit pressures and so covers operating time only. An early surge, a rise of at least 100 mmHg above the baseline median within 12 h, was exploratory, fitted with adjusted Cox models. Cumulative operating time was not associated with mortality in 1493 patients (753 deaths; hazard ratio per 24 h 1.004; 95% CI, 0.999 to 1.009), consistently across settings and in eICU. Among the 862 patients with a baseline and a later TMP value, an early surge occurred in 46 (5.3%), of whom 30 died, and was associated with 28-day mortality (adjusted hazard ratio 1.449; 95% CI, 1.011 to 2.079). With only 46 exposed patients the estimate is imprecise, and the association did not survive false discovery rate correction or external validation, so it is hypothesis-generating. Whether TMP carries prognostic information is more likely to be settled by standardized recording than by larger retrospective cohorts.