376 Dynamic Changes in High Sensitivity Troponin T Are Associated With Increased Sepsis Mortality

Abstract

Elevated serum cardiac troponin is a marker of myocardial injury associated with increased mortality in sepsis. Increasing use of high sensitivity troponin T (hsTrop) in the emergency department (ED) has allowed earlier and more sensitive detection of myocardial injury in acute coronary syndromes but has yet to be studied in relation to sepsis. The objective of this study was to determine the association between hsTrop changes and mortality in sepsis. This was a retrospective cohort study performed at a single, tertiary academic center. A concurrent multicenter sepsis trial was being conducted during the study period (June 1st, 2018 and March 20th, 2020) and the cohort for this study was derived from the screening log of this unrelated study. Adult patients with a suspected infection and hypotension after 1 liter of crystalloid fluids were included. All data were collected via direct database query to the electronic health record. hsTrop was measured in the ED and again approximately 2 hours later. The difference (delta) between the second and first measurements was the primary exposure and treated as a 4- level categorical variable (delta >6 ng/L, - 6 > delta > 6, delta <- 6 and no delta obtained). The primary outcome was 28- day mortality and secondary outcomes included mechanical ventilation, use of vasopressors, and crystalloid fluid received. Inverse probability weighting (IPW) was used to weight covariates for the probability of obtaining a hsTrop delta measurement in the ED. Using IPW derived weights, a multivariable logistic regression model was used to determine the association between hsTrop delta and 28- day mortality. Descriptive statistics and univariate analyses were also performed. After exclusion of clinician suspected baseline hypotension or hypotension not due to sepsis cause, as well as non- sepsis causes of hsTrop delta, 763 patients met the inclusion and exclusion criteria. 48 had a delta > 6 ng/L, 54 had a delta < - 6 ng/L, 175 had a delta between - 6 and 6, and 486 had no hsTrop delta obtained in the ED. Patients who had a hsTrop testing skewed male, older and with history of renal, heart disease or with chief complaints of chest pain or shortness of breath. In the weighted multivariable analysis and after adjustment for sex, age, and Charlson comorbidity index, both an hsTrop delta > 6 (OR 3.4 [95% CI 1.7 - 7.2]) and delta < - 6 (OR 2.2 [95% CI 1.1 - 4.4]) were associated with increased mortality at 28 days. The hsTrop delta >6 group was also associated with an increased risk for early mechanical ventilation (HR 3.0 [95% CI 1.8 - 4.9]), use of vasopressors (OR 3.7 [95% CI 2.0 - 7.0]), and increased crystalloid fluids administered during the first 24 hours (mean 4060 mL vs. ∼ 3000 mL for all other groups, p = 0.01). Dynamic changes in hsTrop obtained at ED presentation are associated with increased mortality in sepsis. Both an increasing and decreasing hsTrop delta are associated with a higher risk for 28- day mortality. This study was limited by its retrospective and single site design. Future studies are required to determine how hsTrop can be used to modify treatment.