American Journal of Hospice and Palliative Medicine®, Ahead of Print.
BackgroundClinicians can appropriately terminate treatment or reduce treatment intensity by determining prognostic factors of end-of-life chemotherapy. In particular, it provides important information for patients with hematological malignancies who receive chemotherapy until near-the-end of life compared with patients with solid tumors. This study aimed to clarify whether existing prognostic tools are associated with the survival in patients with end-of-life hematological malignancies who received chemotherapy.MethodsWe retrospectively reviewed the records of 247 patients diagnosed with hematological malignancies and died at our university hospital hematology ward between May 2015 and May 2021. We performed multivariate analysis in 82 (33.2%) patients who received end-of-life chemotherapy using the Palliative Prognostic Index (PPI) and inflammation-based prognostic models, such as the Glasgow Prognostic Score (GPS), Prognostic Nutritional Index (PNI), and Controlling Nutrition Status (CONUT).ResultsOn comparing 82 patients who received end-of-life chemotherapy with 165 patients who did not, the proportion of patients with PPI group A, GPS score = 0, and CONUT normal/mild was significantly higher among patients who received chemotherapy. In multivariate analysis, we identified PPI groups B (2.0 < PPI ≤ 4.0) and C (PPI > 4.0) [hazard ratio (HR) 2.1290, 95% CI 1.1830-3.828, P = .01166, respectively] and age ≥ 65 years (HR 2.0170, 95% CI 1.1280-3.607, P = .01805) were associated with overall survival.ConclusionPPI use and age were independent associating factors for patients with hematological malignancies receiving end-of-life chemotherapy. PPI, a popular prognostic tool may be helpful for patients and hematologists to make decisions about end-of-life care.