Predicting High-Intensity Resuscitation Needs in Injured Patients Following Hemostasis

Abstract

Introduction: Best practices for the resuscitation of injured patients following hemostasis are poorly defined. The post-hemostasis phase of care is characterized by a wide range of physiologic derangements and multiple therapeutic modalities used to address them (e.g., blood products, intravenous crystalloids, and vasoactive medications). Using a cohort of injured patients from an academic level-one trauma center who required an immediate intervention in the operating room or angiography suite following arrival to the emergency department, we sought to define high-intensity resuscitation (HIR) in this post-hemostasis phase of care; we hypothesized that those who would go on to such a resuscitation could be identified, using only data commonly available at the time of ICU admission. Methods: Hemodynamic, laboratory, and procedure data were extracted for consecutive injured patients (2016-19) admitted to the trauma ICU following an emergent procedure in either the operating room or angiography suite. Significant resuscitation thresholds were defined as the approximate top decile of blood product (≥3 units) and crystalloid (≥4 Liters) use in the initial twelve hours of ICU care and persistent vasoactive medication use (between ICU hours 2-12). The primary outcome was a composite of any of these three modalities. Predictive modeling was performed using logistic regression with predictor variables selected using Least Absolute Shrinkage and Selection Operator (LASSO) estimation. Models were trained using 70% of the cohort and tested on the remaining 30%; their predictive ability was evaluated using area under receiver operator curves in the testing cohort. Continuous variables were depicted with medians and interquartile ranges and proportions as percentages. Results: Six-hundred-and-five (605) subjects were analyzed. A total of two-hundred-and-fifteen (36%) required at least one of the three HIR criteria (11% received ≥ 3 units of blood product, 15% ≥ 4L crystalloid, and 24% required persistent vasopressors). Predictor variables selected by LASSO included: shock index, lactate, base deficit, hematocrit, and INR. Area under receiver operator curves for HIR prediction achieved a value of 0.82. Conclusions: Data available at ICU admission following hemostasis can predict subsequent HIR. Following prospective validation, use of this model may facilitate triage, nursing ratio determination, and resource allocation.