250: EPIDEMIOLOGY AND OUTCOMES OF IN-HOSPITAL CARDIAC ARREST IN SEPTIC PATIENTS WITH CANCER

Abstract

Introduction: Cancer is associated with increased risk of sepsis and sepsis is an underlying condition in up to 27% of in-hospital cardiac arrest (IHCA) patients, with worse short-term outcomes in this subgroup than among non-septic patients. However, the epidemiology and outcomes of IHCA in septic patients with cancer has not been characterized. Methods: We used a statewide data set to identify sepsis hospitalizations aged ≥18 years in Texas during 2014-2017. Sepsis was identified using “explicit” ICD-9 and ICD-10 codes for severe sepsis (995.92, R65.20) and for septic shock (785.52, R65.21). Cancer was defined by ICD-9 and ICD-10 codes 140.x-209.x, C00x-C96x, C7Ax, & C73x. Cardiopulmonary resuscitation was identified using ICD-9 and ICD-10 codes 99.60, 99.63, and 5A12012, excluding those with a principal diagnosis of cardiac arrest. A hierarchical logistic model was used to identify factors associated with hospital survival and survival trends. Logistic regression was used to examine trends of IHCA rates. Results: Among 45,737 sepsis hospitalizations with cancer, 1,862 (4.1%) had IHCA. Among hospitalizations with IHCA, 57.3% were aged ≥65 years, with 42.5% female, 51.7% racial/ethnic minority, and 6.7% had shockable rhythm. IHCA occurred in 9.3% (95% CI 8.8-9.8) of terminal sepsis hospitalizations, with the rate rising over time (odds ratio 1.09/year [95% CI 1.04-1.15]). Hospital survival among those with IHCA was 24.3% [95% CI 22.1-26.7]. On adjusted analyses, hospital survival was lower among Hispanics (adjusted odds ratio [aOR] 0.67 [95% CI 0.47-0.95]), & those with liver disease (aOR 0.49 [95% 0.26-0.91]) & decreased over time (aOR 0.57/year [95% CI 0.51-0.65]). Shockable rhythm was not associated with hospital survival (aOR 0.84 [95% CI 0.51-1.39]). Conclusions: Resuscitation of IHCA in septic patients with cancer was performed more selectively than that reported in the general population, but its rate rose over time. Hospital survival decreased substantially over time, contrasting opposite trends in the general population, with outcome disparities involving ethnicity and comorbidity domains. Further studies are needed to explore the factors underlying these observations, in order to inform efforts to improve outcomes of sepsis-associated IHCA in patients with cancer.