Respiratory Syncytial Virus Infections (RVS) after Allogeneic Hematopoietic Stem Cell Transplantation (HSCT) in the Palivizumab (P) Era.

Respiratory syncytial virus infections after HSCT can lead to severe respiratory failure and is associated with a fatal issue in a substantial number of patients (pts). P has been used to treat RSV infections in HSCT recipients, however its impact on outcome is not well established. From January 1999 to March 2006, all pts with RSV infections after HSCT at Saint Louis Hospital were retrospectively reviewed in order to determine cumulative incidence of RSV-related death, risk factors for 4-month transplant-related-mortality post RSV diagnosis (early TRM), and eventual impact of P. Forty pts with RSV infections were identified, RSV occurred in a median of 177 days (d) (range: −15 d to 987 d ) after HSCT in 3 to 65 year-old recipients (median age: 16 years). 30 pts had received a graft from an unrelated donor. Characteristics at diagnosis were: pneumonia in 16, bronchitis or bronchiolitis in 8 and upper respiratory disease in 16 pts; 16 pts had hypoxemia. Among the 40 pts, 18 received P at diagnosis (15mg/kg - intravenous course - 1 to 3 monthly injections). The groups of pts with or without P were significantly unbalanced for age, source of stem cells, neutrophil and lymphocyte counts, median time from to transplantation to diagnosis (younger age, more cord blood transplantations, lower neutrophil and lymphocyte counts, and shorter duration between transplantation and infection in the P group). Only one patient (treated with P) died from RSV pneumonia alone, giving a 2.6% cumulative incidence of RSV-related death (95% confidence interval (CI): 0–7.7). Seven other pts died during the 4 months following RSV diagnosis. All of them died from respiratory failure of multiple causes (RSV +: n = 4; RSV−: n = 3). Early TRM incidence was 22% (95%CI: 17–27) and 3-year overall survival was 59% (95%CI: 43–83). Risk factors for early TRM in a multiple Cox model were stem cell source (unrelated cord blood, hazard ratio (HR): 4.9, 95%CI: 1.1–22, p=0.039) and hypoxemia at RSV diagnosis (HR: 7.31, 95%CI: 1.8–49). As P was predominantly used in pts with poorer prognosis, we cannot easily test its impact on early TRM: 23% (95%CI: 0–50) and 9% (95%CI:0–22) with or without P, respectively. Nevertheless, pts who received an unrelated graft and who were receiving ≥ 0.5 mg/kg/d prednisolone equivalent at RSV diagnosis were at higher risk of early death (HR: 8.15, 95%CI: 1.56–42). Interestingly, cumulative incidence of death in this group did not change with or without P (37% versus 40%). In contrast, there was no death in pts grafted from a related donor if RSV was diagnosed after engraftment. Overall survival of HSCT pts with RSV infection appeared better than previously reported, and RSV was rarely the direct cause of death after transplantation. The impact of P remains to be defined.