Background: Extubation failure rates for critical patients in pediatric intensive care units (ICUs) range from 5% to 29%. Noninvasive (NIV) ventilation has been shown to decrease extubation failure. We compared reintubation rates and outcomes of patients supported with NIV neurally adjusted ventilation assist (NAVA) versus historical controls supported with high-flow nasal cannula (HFNC). Methods: Case–control study of infants less than three months of age who underwent cardiac surgery and received NIV support after extubation from January 2011 to May 2017. All patients supported with NIV NAVA after it became available in September 2013 were compared to matched patients extubated to HFNC from prior to September 2013. Results: Forty-two patients identified for the NIV NAVA group were matched with 42 historical controls supported with HFNC. Groups had similar baseline characteristics based on rate of acute kidney injury, number of single ventricle patients, Society of Thoracic Surgeons-European Association for Cardio-Thoracic Surgery (STAT) category, age, weight, bypass time, and duration of intubation. There was no significant difference in reintubation rates within 72 hours (14.3% in the HFNC group and 16.7% in the NIV NAVA group, P = 1.0). Median duration from extubation to coming off NIV support was longer in the NIV NAVA group (3.6 days vs 0.6 days, P < .001). Median time from extubation to ICU discharge was longer in the NIV NAVA group (10.5 vs 6.8 days, P = .02), as was total postoperative ICU length of stay (LOS; 17.6 vs 12.2, P = .01). Conclusions: Introduction of NIV NAVA for postextubation support did not reduce reintubation rates compared to HFNC. Further study is needed as adoption of NIV NAVA may prolong LOS.

4317 Introduction With a huge burden of hematological and oncological disorders for which hematopoietic stem cell transplant (HSCT) is a curative option, India saw the development of HSCT program from 1986 onwards. Since then, some 10 centres have performed over 1600 transplants until the year 2005, including one-third in thalassemic patients. However, current HSCT facilities are still limited and cannot address the treatment needs of the immense patient load. There is a need for new transplant centres in India. We are describing our experience of HSCT programme at Sir Ganga Ram Hospital, a tertiary hospital in North India, over a period of 3 years. Patients and methods A retrospective study of all patients who underwent HSCT over a period of 3 years, from January 2006 to August 2009 was done and follow-up data was analyzed. Results Thirty-nine transplants (16 allogenic and 23 autologous) were done during this period in 34 patients (26 males and 8 females). The median age of transplant patients was 34 years (11 months-68 years). Children comprised 41.2% of the patients. The indications for 16 allogenic transplants were thalassemia major-6, acute myeloid leukemia (AML)/myelodysplastic syndrome-6, severe aplastic anemia-3 and high-risk acute lymphoblastic leukemia-1. Donors were HLA-matched sibling in 13 cases, HLA-matched relative in 1 and unrelated umbilical cord blood in 2. The source of HSCT was peripheral blood in 8 patients, bone marrow in 6 and umbilical cord blood in 2. Fourteen patients underwent myeloablative transplants and two were given reduced intensity conditioning. Three Donor Lymphocyte infusions were given to two patients. Seven patients (44.8%) are alive and disease free at a median follow-up of 453 days (65-591 days). Thirteen patients showed neutrophil engraftment at a median duration of 14 days (range 11 to 44). Engraftment syndrome was seen after unrelated cord blood transplant in one child. Acute Graft versus host disease (GVHD) was seen in 6 patients, grade I-II was seen in 5 patients, which responded to steroids. Steroid-refractory grade IV GVHD was seen in one child; it did not respond to mycophenolate and infliximab. Massive bleeding (gastro-intestinal-2 and pulmonary-1) occurred in three children. Three patients developed sinusoidal obstruction syndrome. All were given Defibrotide and two responded. A total of 7 allogenic HSCT recipients have died, out of which 2 AML patients died of relapse at 56 and 90 days post-transplant. The main causes of death in 5 other patients were sepsis (3 bacterial and 2 fungal), sinusoidal obstruction syndrome (1 case), acute GVHD (1 case) and chronic GVHD (1 case). Two thalassemic children rejected graft but are alive and transfusion dependent. The main indications for 19 autologous HSCT were multiple myeloma-9, Non-Hodgkin's lymphoma-4, metastatic neuroblastoma-2, relapsed Hodgkin's lymphoma-1, relapsed rhabdomyosarcoma-1, relapsed primitive neuroectodermal tumor (PNET) of the kidney-1 and rejection post cord blood transplant in thalassemia-1. The source of HSCT was peripheral blood in 16 patients and bone marrow in 3. Ten patients (55.6%) are alive and disease-free at a median follow up of 114 days (range 21 - 617 days). Seventeen patients engrafted neutrophils at a median duration of 12 days (range 9 to 30). A child with metastatic neuroblastoma relapsed 8 months post transplant and is on palliative radiotherapy. An adult with indolent lymphoma relapsed 1-year post transplant and is in partial remission two years after relapse. Six autologous HSCT recipients have died. One adult with NHL died of progressive disease one-month post transplant and one with multiple myeloma died of relapse 3-years post transplant. One child with PNET received 4 autologous staged transplants and local radiotherapy with very good partial response but died of candida sepsis 92 days after his last transplant. Three other patients died of bacterial sepsis. At a median follow up of 200 days (range 21 to 1200 days), the estimated overall survival and event free survival for all the transplant population are 67.3% ± 8.6% and 63.5± 8.9% respectively. Overall transplant related mortality is 23.5%, with a decrease from 28.6% to 22.2% after a dedicated HSCT unit with HEPA filtered rooms became functional by mid 2007. Conclusion With our encouraging results we can conclude that it is possible to setup a HSCT unit in developing world and results get better with each passing year. Disclosures: No relevant conflicts of interest to declare.

Background: Antifibrinolytic agents are frequently used during pediatric heart surgery with cardiopulmonary bypass (CPB) to reduce transfusions. There are no studies comparing anti-inflammatory effects of antifibrinolytic agents, tranexamic acid (TXA) and Epsilon Aminocaproic acid (EACA). We compared the two agents in pediatric patients undergoing redo sternotomy with CPB. Aim: To compare anti-inflammatory effects of tranexamic acid versus aminocaproic acid in pediatric patients undergoing redo sternotomy and cardiopulmonary bypass. Methods: We conducted a randomized, double blind pilot study, comparing 10 subjects in each group receiving EACA and TXA. A cytokine panel was used to measure 13 inflammatory markers in pre, immediate post and 24 hours post-CPB period. Between group comparisons were tested with Mann-Whitney U tests and within group comparisons with Friedman tests. Results: Sample characteristics were comparable in both groups. Post CPB, plasma levels of 7 markers increased significantly (p<0.05) in both groups, including MCP-1; 3 increased significantly (p<0.03) in the EACA group alone, including GM-CSF; and 3 did not change over time (Table 1). No difference was found between groups for markers except for IL-10, which was significantly higher in EACA group post CPB. While absolute values of markers, chest tube output and volume of blood product needs were lower in TXA group, the differences were not statistically significant. Conclusion: There was no significant difference in anti-inflammatory profiles between EACA and TXA in this pilot study. GM-CSF and MCP-1 were increased in our study post CBP which has not been described in previous studies.