Incidence, predisposing factors, and outcome of engraftment syndrome in pediatric allogeneic stem cell transplant recipients.

Engraftment syndrome (ES) has been recognized as an inflammatory condition during neutrophil recovery after hematopoietic stem cell transplantation (HSCT) characterized by noninfectious fever and skin rash. It has been reported to occur frequently after autologous HSCT in children and adults, and has been shown to be an independent risk factor for increased transplant-related mortality (TRM). However, virtually no data exist on its occurrence after allogeneic HSCT in children. To determine incidence, predisposing factors for, and complications of ES in a pediatric transplant cohort, we analyzed 61 consecutive recipients of a myeloablative allogeneic HSCT for the occurrence of ES. Diagnosis of ES was established when children presented with > or =2 of the following symptoms within 7 days before engraftment: (1) fever >38.0 degrees C, (2) skin rash, (3) weight gain and albumin drop, or (4) dyspnea, hypoxia, and pulmonary infiltrates. Incidence of ES in this cohort was 48% (29 of 61). In a univariate analysis, posttransplant granulocyte-colony stimulating factor (G-CSF) administration (P = .02), and high mononuclear cell count (MNC) (P = .002) were identified as significant risk factors predisposing for the development of ES. In a multiple logistic regression analysis, amphotericin B therapy (P = .009) and high MNC (P = .004) were significant explanatory variables for ES risk. There was a slight trend toward a higher rate of chronic GVHD (cGVHD) in patients with ES (P = .11). However, after a median follow-up of 9.5 years overall survival (OS) (P = .53) and TRM (P = .65) did not differ between the 2 groups. ES presenting with fever, rash, weight gain, and pulmonary symptoms should be recognized as a frequent complication of allogeneic HSCT after myeloablative conditioning in children. Treatment with G-CSF, amphotericin B, and a high nucleated cell count of the graft predisposed for the development of ES in this study. OS and TRM in this cohort were not affected by the occurrence of ES.