Readmissions after Umbilical Cord Blood Transplantation and Impact on Overall Survival

Abstract

Patients treated with allogeneic hematopoietic stem cell transplantation (SCT) have high rates of readmission, but the incidence after umbilical cord blood transplantation (UCBT) is poorly described. The goal of this study was to identify the incidence and risk factors for readmission after UCBT and the impact of readmission on overall survival (OS). A retrospective review of patients receiving a UCBT at Dana-Farber/Brigham and Women's Hospital between January 1, 2004 and December 31, 2013 was performed. The readmission rates 30 days after discharge from the UCBT admission and at day +100 after the UCBT were examined. Reasons for readmission, as well as sociodemographic, disease-, and SCT-related variables were evaluated. Predictors of readmission and the impact of readmission on OS were identified using multivariate regression analysis. Of patients who received a UCBT, 42 of 126 patients (33.3%) were readmitted within 30 days of discharge and 57 of 123 patients (46.3%) were readmitted by day +100 after transplantation. The most common causes for readmission were infection (38.3%), fever without a source (14.8%), and graft-versus-host disease (8.6%). Infection during the index admission was the only significant risk factor for readmission at both time points in a univariate and multivariate regression analysis (OR, 11.66; 95% CI, 2.77 to 49.13; P < .01 and OR, 5.4; 96% CI, 1.87 to 15.58; P < .01). Prior radiation therapy was also associated with an increased risk of readmission at both time points in the multivariate regression model (OR, 20.6; 95% CI, 3.53 to 120.04; P ≤ .01 and OR, 5; 95% CI, 1.21 to 20.71; P = .03). The multivariate regression model also showed that black race and a median income of <60,000 in the patient's home zip code increased the risk of readmission by day +100 (OR, 30.17; 95% CI, 1.33 to 684.48; P = .03 and OR, 2.88; 95% CI, 1.04 to 7.8; P = .04, respectively). After adjusting for age, disease type, and the disease status at transplant, OS was reduced for the patients who were readmitted by day +100 (HR, 2.44; 95% CI, 1.46 to 4.06; P < .01). There was also a trend toward decreased survival in patients readmitted 30 days after discharge (HR, 1.58; 95% CI, .96 to 2.6; P = .07). Readmissions are common after UCBT. Infections and fever without a source are the most common causes of readmission. Being readmitted by day +100 resulted in a lower 5-year OS rate as compared with patients who were not readmitted. Prior radiation and infection during the transplant admission resulted in increased risk of readmission by 30 days and day +100. Similarly, race and socioeconomic status predicted readmission by day +100. Further understanding of the mechanisms leading to readmissions in these groups may allow for identification of interventions that could reduce readmissions and thus improve mortality.