Bio Shock Index: Proposal and Rationale for a New Predictive Tool for In-hospital Mortality in Traumatic Brain Injury Patients.

Abstract

BACKGROUND\nWe proposed a novel prognostic tool for the prediction of in-hospital mortality based on combination of hemodynamic parameters and biomarker in patients with TBI. We hypothesize that combination of shock index (SI) with high sensitive troponin T (HsTnT), the Bio-SI, has better prognostic power than its individual components.\n\n\nMETHODS\nA retrospective chart review was conducted (2011-2018) for TBI patients. Patients were categorized into two groups (low and high Bio-SI) based on the ROC curve.\n\n\nRESULTS\nA 2619 patients were admitted with TBI; of them 1471 fulfilled inclusion criteria and 73% had high Bio-SI (≥10). High Bio-SI were associated with more intraventricular hemorrhage (p=0.001), brain edema (p=0.001), and had lower mean arterial pressure (p=0.001), admission GCS (p=0.001) and higher SI (p=0.001), serum lactate (p=0.001), HsTnT values (p=0.001) and Rotterdam score (p=0.03). Patients with high Bio-SI had a prolonged hospital (p = 0.003) and ICU stay (p=0.001); longer ventilatory days (p=0.001) and had higher rates of pneumonia (p = 0.001), sepsis (p =0.001) and in-hospital mortality (p = 0.001). The Bio-SI showed high sensitivity and NPV (91.4% and 94.4%, respectively) as compared to elevated SI (50.2% and 87.6%, respectively) and positive troponin (79.7% and 93.7%, respectively).\n\n\nCONCLUSION\nthe Bio-SI is potentially a better tool than its individual components to predict in-hospital mortality among TBI patients; however, HsTnT alone outperforms SI. Prospective studies and multi center trials studying troponin levels and SI in all TBI patients, with the inclusion of outcome scores will prove or disprove the predictability of the new index.