Predictive Factors for Failure of Non-Invasive Ventilation in Hospitalized Patients with Cancer

Abstract

Introduction: The non-invasive ventilation (NIV) presents confirmed benefits in various clinical conditions, however, the results in patients with cancer are controversial. Objective: To analyze the predicting factors for failure of the NIV in cancer patients; To describe hospital mortality and overall survival after admission. Method: Study of retrospective cohort including patients with solid tumors and hematological neoplasm who have been admitted to the hospital stay at Hospital of Cancer I of the National Cancer Institute (HCI/INCA) between Jan 1 st and Dec 31 2017 and were submitted to NIV. The association between the exposure (clinical and socio-demographic variables) and the outcome (NIV failure) was performed by gross and adjusted logistic regression. The Kaplan-Meier method was used to analyze the overall survival. Results: Sixty-six patients with mean age of 62.3 years (± 15.0 years) were included. The average lasting time of the first session was 49.8 min (±30.9), the average number of sessions was 2.1 (±1.4). The patients who showed failure had longer time hospital stay (11.8 days vs 6.0 days) and higher hospital mortality (90.9 vs 43.6%). The patients with lung infection showed a higher risk of 4.71 times of failure in NIV related to those patients who showed succeeding (OR 4.71; IC 95%, 1.14-19.47; p=0.032). Conclusion: Patients who showed lung infection were more likely to failure in NIV. Was observed a worst overall survival between those patients who failed in NIV.