A37 - Lung cancer patients and unplanned presentations to hospital: insights from a single institution experience

Abstract

Background: New treatments for lung cancer (LC) has been introduced in clinical practice over the last years, however significant toxicities are still observed and may lead to unscheduled access to the oncology clinic as well as hospital admission. Furthermore, LC patients (pts) frequently require emergency medical care due to acute symptoms and life-threatening conditions. We report the magnitude of unplanned presentations and risk factors affecting their number and outcome in a University Hospital. Materials and methods: We retrospectively collected a consecutive cohort of 285 LC pts treated with chemotherapy and target therapy at the Department of Oncology of Udine, from 2012 to 2015 regardless of the stage. Our analysis focused on the subgroup of pts that experienced an unplanned accessed at both the Oncology and the Emergency Department. Demographical data, tumour characteristics, treatments information and blood tests have been recorded. Associations between clinico-pathological characteristics and hospitalization were investigated through logistic regression. The impact of patients' characteristics at first access on the number of presentations was explored through Mann-Whitney test. Results: In our study 46% of pts experienced an unplanned access to the Oncology or the Emergency Department: median access for patient is one (1-10). Hospitalization was the outcome of the first access for 16.1% of pts, while 21% of pts experienced one hospitalization overall and 5% of pts were admitted more than once. At first presentation, 74.6% pts had an ECOG PS 0-1; pain (37.7%) and fever (17.6%) were the main diagnoses performed during visit. Pulmonary embolism (PE) was found in 7.1% pts. Anemia and neutropenia was found in 28,2% and 9,2% of pts respectively; however no febrile neutropenia was diagnosed. Accesses to Emergency Department and ECOG PS 2-3 were closely related to hospitalization (respectively OR 2.62, 95%CI 1.05-6.52; p = 0.0385 and OR 6.27 95%CI 2.59-15.31; p < 0.0001). Notably, only diagnosis of PE significantly increased the number of unplanned accesses (P = 0.0387). Conclusion: To explore the impact of LC patients' characteristics on unplanned access pattern and hospitalization, could be crucial in order to better identify frail patients, minimize quality-of-life impairment and over-hospitalization.