Abstract 110: Lactate and Hypotension as Predictors of Mortality After In-Hospital Cardiac Arrest

Abstract

Background: In hospital cardiac arrest (IHCA) affects >290,000 people in USA annually. Although there are many differences between IHCA and out of hospital cardiac arrest (OHCA), the bulk of data used to guide management comes from OHCA studies. Prediction of mortality after IHCA could be useful in making decisions around post-arrest care. We hypothesized that elevated lactate and the need for vasopressor support after arrest would predict mortality in an IHCA population. Methods: Retrospective single-center observational study of all adult IHCA patients with sustained return of spontaneous circulation (ROSC), lactate within 2 hrs of ROSC, and intubated pre-arrest or within 1 hr after, from 2008 - 2018. Multivariable logistic regression was used to evaluate the association of post-ROSC lactate and need for vasopressors, as well as other covariates, with mortality. Backwards selection was used to determine the most parsimonious model. Results: Of 541 patients; 364 met criteria and were included. Overall mortality was 56%. The distributions of initial rhythm, pre-arrest vasopressor and pre-arrest mechanical ventilation were similar between groups. Patients who received vasopressors within 3 hrs of ROSC had higher mortality compared to patients who did not (58% vs. 43%, p-value 0.04). Elevated lactate level was also associated with mortality (44% if < 5 mmol/L, 58% if 5 - 10 mmol/L, and 73% if ≥10 mmol/L, p-value<0.01). Mortality in those with lactate <5 and no vasopressors was 33%, compared to 75% in those with lactate >10 and need for vasopressors (p<0.01). The most parsimonious predictive model included lactate, post-arrest vasopressor, age, arrest location, and pre-arrest diagnosis (AUC 0.68 [95 CI: 0.63-0.74]). Conclusion: Post-ROSC lactate and need for vasopressor were useful predictors of mortality, although AUC was lower than what has been reported in OHCA studies. Development of a more discriminating tool would be valuable to clinicians and in IHCA research.