Our findings support the feasibility of exercise training interventions during hospitalization, including immunocompromised children.

Background: Advances in perinatal care can influence morbidity and mortality in newborns at the limit of viability. Knowledge of these changes over time may help improve clinical decision making, optimize resource allocation and increase quality of care. Objectives: To evaluate the influence on morbidity and mortality of changes introduced in the perinatal care of preterm infants (22-26 weeks' gestational age, GA) in Spain between two consecutive periods (2002-2006 and 2007-2011). Methods: An analysis of prospectively collected data in a national database network (SEN1500) was performed. All live newborn infants of 22-26 weeks' GA born in or transferred to referral centers of the SEN1500 network in the first 28 days of life were included. Perinatal interventions, clinical management, neonatal morbidity, and survival until hospital discharge were retrieved. Results: A total of 5,470 newborns were included (2,533 and 2,937 in each period, respectively). The major changes introduced during the second period were as follows: (1) lower proportion of extramural births (11.0 vs. 8.9%, p = 0.01), (2) increase in antenatal steroids (69.5 vs. 80.8%, p < 0.001), (3) delivery by C-section (41.8 vs. 48.3%, p < 0.001) and (4) use of CPAP during resuscitation (7.8 vs. 20.7%, p < 0.001). Death in the delivery room decreased from 5.1 to 3.2% (p < 0.001). Survival increased from 49.9 to 57.9% (p < 0.001), and survival without major morbidity increased from 18.1 to 21.2% (p = 0.006). Conclusions: During the second period, a greater attachment to practices proven to have a beneficial impact on survival and reduction of morbidity in the extremely preterm infant was noted, and survival and survival without major morbidity increased. A more conservative approach was detected for newborns of 22 weeks' GA.

Summary:We prospectively analyzed the incidence, risk factors and outcome of engraftment syndrome (ES) in 112 patients undergoing autologous peripheral blood progenitor cell transplantation with different malignancies between January 1999 and December 2003. The median age was 8 years (range 1-18). There were 73 males. There were 37 hematological neoplasias and 75 solid tumors. Disease status at transplantation was early in 49, intermediate in 15 and 48 in advanced phase. The median CD34 þ cells infused was 4.6 Â 10 6 /kg. With a median follow-up of 23 months (1-116 months), 38 patients developed ES. The cumulative incidence of ES was 34.574.5% and the event-free survival was 58.3712%. There were no differences in the causes of death between patients with or without ES. A high number of CD34 þ cells/kg infused, patients transplanted in early phase, the type of malignancy (solid tumor) and conditioning regimens other than busulfan based were significantly associated with ES in a multivariate analysis.

Transplant-related problems have been partially overcome by using reduced-intensity conditioning (RIC), graft engineering, and alternative donors. In all, 21 leukemia patients with no suitable donor received a hematopoietic stem cell transplantation from a mismatched/haploidentical related (n=16) or unrelated donor (n=5). Fludarabine-RIC and PBSC graft were used. Manipulation was done by CD34+ selection (n=9) or CD3/CD19 depletion (n=12). Results were compared with patients (n=26) conditioned with the same regimen and grafted with a CD34+-selected PBSC from identical related donors. Median time to neutrophil recovery was 12 days (range, 10-19 d). Platelet engraftment was faster with a CD3/CD19-depleted graft (median, 11 d; range, 9-21) than with a CD34+ graft (median, 14 d; range, 9-53; P=0.003). Full donor chimerism in bone marrow CD34+ cells was higher in CD3/CD19-depleted graft group compared with CD34+-selected group (P=0.02). CD3/CD19 depletion showed higher natural killer cell counts even after 1 year. Nonrelapse mortality (7% for matched CD34+-selected grafts and 11% for mismatched/haplo-CD3/CD19-depleted grafts), relapse probability (27% for related CD34+-selected patients and 33% for related CD3/CD19-depleted patients), and disease-free survival were similar for both the groups. In conclusion, using graft engineering procedures after RIC for hematopoietic stem cell transplantation offers a high probability of engraftment, fast immune recovery, and very low mortality even with mismatched donors.

We retrospectively analyzed outcomes in 67 children with acute leukemia who received hematopoietic stem cell transplantation from alternative allogeneic donors: 29 received a haploidentical family donor and 38, an unrelated cord blood donor. All transplantations were performed from 1996 through 2010 in our center. Neutrophil and platelet engraftment were significantly delayed after cord blood transplantation. The median times to neutrophil and platelet recovery were 13 d (7-34) and 11 d (5-70) after haploidentical transplant and 20 d (9-125) and 56 d (12-200) after cord blood (P < 0.001). All supportive care measures included red blood cell, and platelet transfusions were significantly increased in cord blood transplantation group.Transplant-related mortality rates was lower with haplo donors (25 ± 9%) than with cord blood donors (47 ± 9%) (P< 0.05). Acute graft-versus-host disease (GVHD) more than grade II was also lower in haploidentical transplants (19 ± 7%) than in cord blood transplants (44 ± 10%) (P < 0.03). Relapse and chronic GVHD incidence were not significantly different in the two groups. Leukemia-free survival was higher after haploidentical transplants (44 ± 10%) than after cord blood transplants (33 ± 7%) (P < 0.03). Main differences were observed in patients diagnosed with acute lymphoblastic leukemia: haplo, 41 ± 13%; cord blood, 26 ± 9% (P < 0.03) and in advanced phase of disease: haplo, 37 ± 14%; cord blood, 21 ± 8% (P < 0.05). In conclusion, haploidentical transplants are a good and promising alternative option for patients with childhood leukemia who lack an human leukocyte antigen-matched donor (sibling or unrelated donor).