Outcomes of Hospitalization for Stem Cell Transplant in Sickle Cell Disease: Are We There Yet?

Background: Sickle cell disease (SCD) is a common hemoglobinopathy, characterized by vaso-occlusive crises and affects over 100,000 people in United States. It afflicts long-lasting organ damage with a spectrum of clinical severity. Median survival for SCD is shortened to the 6th decade of life despite advances in medical care (Elmariah et al Am J Hematol 2014). Allogeneic stem cell transplant (SCT) is a potentially curative option and is increasingly considered in patients with severe symptomatic SCD. The use of SCT is limited by donor availability and treatment related complications. Several advancements in conditioning regimen and use of alternate donor source have favorably impacted the feasibility of this approach. We identified SCT performed for sickle cell disease in the Healthcare Cost and Utilization Project National Inpatient Sample (HCUP-NIS) from 2002-2014 in an attempt to identify hospitalization outcomes, factors affecting length of stay and healthcare utilization. Methods: HCUP-NIS is a 20% stratified sample of all discharges from hospitals across 46 states in the United States and incorporates weighting algorithms to predict nationwide estimates. We used International Classification of Diseases, Ninth Revision (ICD-9) procedure codes to identify HCT hospitalizations {Bone marrow (BM):41.02, 41.03, Peripheral blood (PB): 41.05, 41.08 and Cord blood (CB): 41.06} for sickle cell disease (282.5,282.6X -282.6X). We excluded patients who underwent SCT for indications other than SCD. Surveyfreq was used to calculate proportions and surveymeans was used to calculate median length of stay and hospital charges. Kruskal-Wallis test was used for non-parametric data. Chi-square for categorical data frequency, P value of < 0.05 was statistically significant. All analysis was performed using SAS 9.4. Results: Outcomes were analyzed from a total of 742 hospitalizations for SCT from 2002-2014 (table 1). Median age for stem cell transplant was 9 years. GVHD occurred in 14% of stem cell transplants. Overall, in-hospital mortality was low at 2.6% while mortality in patients who developed GVHD was 14%. Bacterial infections (including C.difficile) occurred more commonly than viral or fungal infections (table 2). Patients who developed graft vs host disease (GVHD) were also more likely to have bacterial, viral and fungal infections than those without. Pain crisis was noted in 9% of total admissions while stroke occurred in 6%. Median length of stay (LOS) was 35 days and median charges were $359,646. If GVHD developed, median LOS increased to 54 days while median charges increased to $712,324. Similarly, bacterial sepsis was associated with a longer median LOS of 63 days while median charges increased to $626,986 (table 3). Conclusions: The rate of inpatient mortality with SCT in sickle cell disease is lower than the overall inpatient mortality rate for allogeneic SCT (7%; Godara et al bbmt 2018), indicating a favorable outcome for these patients. Infections do occur commonly during the course of hospitalization, especially in association with GVHD. Length of stay is adversely impacted by occurrence of GVHD, bacterial sepsis, C.difficile infection and viral infections. While we are limited by duration of follow up in our study, these patterns suggest some essential modifiers for inpatient morbidity and mortality, therefore require validation in a large prospective study. Disclosures No relevant conflicts of interest to declare.