Abstract Body: Background
The epidemiology and prognostic factors of out-of-hospital cardiac arrest (OHCA) due to cancer remain poorly described. A better understanding of these aspects is essential for clinical decision-making in this patient population.
Hypothesis
We hypothesize that cancer-induced OHCA is associated with extremely poor survival, and that only a few factors are significantly associated with survival.
Aims
To characterize the clinical features and identify factors associated with one-month survival following cancer-induced OHCA in Japan.
Methods
We conducted a retrospective observational study using data from OHCA cases in Japan between 2005 and 2021. Cancer-induced OHCA was defined as cardiac arrest with a presumed cause of “malignant tumor” recorded in the registry. The primary outcome was one-month survival. Patients aged ≤18 years and those who did not receive resuscitation were excluded. Characteristics of cancer-induced OHCA were compared with those of other non-cardiac OHCA. Then, we analyzed demographic and prehospital factors of cancer-induced OHCA using multivariable logistic regression. To address potential time-dependent bias regarding epinephrine use, landmark analyses were conducted at 10, 20, and 30 minutes after CPR initiation.
Results
Among 2,081,013 OHCA cases, 34,467 (1.7%) were attributed to cancer. One-month survival rate in this group was 1.8%, significantly lower than that of other non-cardiac causes (7.9%). Compared to other non-cardiac OHCA, cancer-induced OHCA cases were more often witnessed (48% vs. 43%) and more likely to present with asystole (72% vs. 67%). Intubation (5% vs. 10%) and epinephrine administration (23% vs. 28%) were less often performed in the cancer group. Survivors tended to have witnessed arrests (adjusted OR [aOR] 2.01), shockable rhythm (aOR 2.39), physician-provided advanced life support in the field (aOR 2.68), palpable pulse upon emergency medical service arrival (aOR 4.75), and ROSC at hospital arrival (aOR 8.76). While epinephrine use was negatively associated with one-month survival (aOR 0.37), landmark analysis at 10 minutes revealed a positive association (aOR 1.72), suggesting time-dependent bias.
Conclusions
Cancer-induced OHCA was associated with extremely poor survival. One-month survival was linked to identifiable early features, such as shockable rhythm and early ROSC. These features may help clinicians in tailoring resuscitation strategies and post-arrest care for patients with cancer-induced OHCA.