Incidence and Outcomes for Veno-Occlusive Disease in a Pediatric Transplant Cohort on Retrospective Application of Modified Seattle and EBMT Criteria

Background Veno-occlusive disease (VOD) is a serious complication of hematopoietic stem cell transplant (HSCT) and is associated with significant mortality. There are published clinical criteria applied for diagnosis, including Baltimore, modified Seattle, and European Society of Bone Marrow Transplantation (EBMT). The lack of consensus on what defines VOD and the subjective nature of clinical criteria used in its definition likely account for the broad range of VOD incidence (10-35%). Given improved outcomes with earlier treatment with defibrotide, there exists a need to be more precise and prompt in diagnosing VOD. The primary goal of this study was to assess the incidence of VOD, as designated in the electronic medical record (EMR) and compare this to when the actual clinical definitions were met. The secondary goal was to identify additional clinical or laboratory trends that could be predictive of VOD risk and outcome. Methods A retrospective analysis was performed on pediatric patients that underwent HSCT at a single institution from 2009-2018. Various patient and transplant data were collected and analyzed in a univariate and multivariate manner to assess potential risk factors. Outcomes included VOD, 100-day survival, and length of hospitalization post-HSCT. FisherOs exact test was used for categorical variables; t test or ANOVA were used for continuous variables. Results A total of 215 transplants in 184 pediatric patients met inclusion for the analysis. Twenty-seven patients were diagnosed with VOD per the EMR (15%), 22 were treated with defibrotide. There were 63 transplants that met the modified Seattle criteria; however, these patients were not clinically diagnosed with VOD by the treating physician (30%). On application of the EBMT criteria, the estimated incidence was 50%. In the entire cohort, 16 patients died within day 100 from the HSCT (100-day survival =91%). Interestingly, percent survival at 100 days for those who met either modified Seattle or EBMT criteria but were not clinically diagnosed with VOD was similar to those who did not meet criteria (92% vs 95%) yet differed from those given a clinical diagnosis of VOD (78%, p=0.013, Figure 1). Only those with a clinical diagnosis of VOD had a longer hospitalization post-HSCT (Figure 2). An increase in VOD was observed for patients receiving myeloablative conditioning (p=0.033) while age, sex, stem cell source, transaminitis at the time of transplant, number of blood transfusions pre-HSCT ( 10), and rapidity of weight gain were not risk factors for VOD. Late diagnosis (u003e14 days post-HSCT) and rapid rise in total bilirubin were associated with a poor outcome (Figure 3). In conclusion, our study demonstrates the challenges in using defined clinical criteria to diagnose VOD and verifies earlier studies noting the importance of prompt diagnosis and treatment for optimal outcomes.