To determine if cancer patients experience variability in incidence or management of emergency general surgery (EGS) conditions compared to non-cancer patients. The true frequency, and natural history of EGS conditions among cancer patients has not been characterized. We utilized SEER-Medicare data from January 2006-December 2015 to compare patients with breast, prostate, and lung cancer to a non-cancer cohort. Patients were followed from date of cancer diagnosis, or an index date for non-cancer patients, to the development of an EGS condition, death or last follow up. We assessed the cumulative incidence of EGS conditions over time, and fit multivariable Cox proportional hazards models to evaluate the impact of time-dependent surgical intervention on mortality. We identified 322,756 patients with breast (N=82,147), lung (N=128,618), and prostate cancer (N=111,991) and 210,429 non-cancer patients.. Cancer patients had a higher incidence of an EGS condition within the first year after diagnosis (4.8% vs. 3.2%), with lung (6.8%) and breast cancer (4.0%) showing consistent rends. Cancer patients were less likely to undergo surgery for (13% vs. 14%, P=0.005), though this varied by cancer type and EGS conditions. Patients with breast (HR 1.27, 95%CI 1.17-1.39) and lung cancer (HR 3.27, 95%CI 3.07-3.48) were more likely to die within 30-days of an EGS diagnosis. Cancer patients experience a higher incidence of EGS conditions within the first year following diagnosis, but are less likely to undergo surgery. Future research is needed to explore the interplay between EGS conditions, their management, and receipt of intended oncologic therapy, and resulting outcomes.
Read full abstract