Contemporary impacts of a cancer diagnosis on survival following in-hospital cardiac arrest

Abstract

Aim: The objective of this study was to determine whether survival and post-arrest procedural utilization following in-hospital cardiac arrest (IHCA) differ in patients with and without comorbid cancer.

Methods: We retrospectively reviewed all adult (age ≥18 years old) hospital admissions complicated by IHCA from 2003 to 2014 using the National Inpatient Sample (NIS) dataset. Utilizing propensity score matching using age, gender, race, insurance, all hospital level variables, HCUP mortality score, diabetes, hypertension and cardiopulmonary resuscitation use, rates of survival to hospital discharge and post-arrest procedural utilization were compared.

Results: From 2003 to 2014, there were a total of 1,893,768 hospitalizations complicated by IHCA, of which 112,926 occurred in patients with history of cancer. In a propensity matched cohort from 2012 to 2014, those with cancer were less likely to survive the hospitalization (31% vs. 46%, p < 0.0001). Following an IHCA, rates of procedural utilization in patients with cancer were significantly less when compared to those without a concurrent malignancy: coronary angiography (4.0% vs. 13.0%), percutaneous coronary intervention (2.2% and 8.0%), targeted temperature management (0.8% vs. 6.0%); p < 0.0001 for all comparisons. This patient population was less likely to have acute coronary syndrome (12.6% vs. 27.0%) or congestive heart failure (24.5% vs. 38.2%); p < 0.0001 for both comparisons. Survival improved in both groups over the study period (p < 0.0001).

Conclusions: Patients with a history of cancer who sustain IHCA are less likely to receive post-arrest procedures and survive to hospital discharge. Given the expected rise in the rates of cancer survivorship, these findings highlight the need for broader application of potentially life-saving interventions to lower risk cancer patients who have sustained a cardiac arrest.

Keywords: Cancer; Cardio-oncology; Cardiovascular disease; In-hospital cardiac arrest.

Fig. 1 –

(A) Annual rate of IHCA among hospitalizations with and without co-diagnosis of cancer. (B) Propensity matched survival trends of IHCA among hospitalizations with and without cancer. [Hospital admissions from 2003 to 2014 were divided into individual year cohorts, after which cancer admissions with documented in-hospital cardiac arrest (IHCA) were propensity matched to non-cancer admissions using age, gender, race, insurance, hospital level variables, diabetes, hypertension, cardiopulmonary resuscitation (CPR) use, and Healthcare Cost and Utilization Project (HCUP) mortality score].

Fig. 2 –

Cardiopulmonary resuscitation utilization rates, by cancer status in a propensity matched cohort.

Fig. 3 –

(A) Procedural utilization among cancer hospitalizations within the propensity-matched 2012–2015 cohort. P values (all comparisons saw P-values < 0.001). (B) Procedural utilization in the high-survival cancer matched cohort. For this analysis, subjects with a history of cancer with a favorable five-year survival (thyroid, breast, prostate and testicular cancers or non-Hodgkin lymphoma) were matched to two controls without a history of cancer. Utilization of some procedures differed by co-diagnosis of cancer: P = 0.0498 for CPR use, P = 0.0337 for angiography use, P = 0.0151 for PCI use, P = 0.0804 for TTM use, P = 0.0543 for IABP use and P = 0.0080 for ICD implantation. (C) Procedural utilization among subjects with and without a cancer with presenting diagnosis of MI. *All comparisons were significant, expect for CPR use.

Fig. 4 –

Discharge disposition of hospitalizations following IHCA (p < 0.0001 for each pairwise comparison).