Pediatric ECMO: unfavorable outcomes are associated with inflammation and endothelial activation.

BACKGROUND: Inflammatory and endothelial activation responses during extracorporeal membrane oxygenation (ECMO) support in children are poorly understood. In this study, we aimed to determine if circulating inflammatory, endothelial activation, and fibrinolytic markers are associated with mortality and with neurologic outcomes in children on ECMO. METHODS: We conducted a secondary analysis of a two-center prospective observational study of 99 neonatal and pediatric ECMO patients. Inflammatory (interferon gamma [IFNγ], interleukin-6 [IL-6], IL-1ß, tumor necrosis factor alpha [TNFα]), endothelial activation (E-selectin, P-selectin, intercellular adhesion molecule-3 [ICAM-3], thrombomodulin [TM]), and fibrinolytic markers (tissue plasminogen activator [tPA], plasminogen activator inhibitor-1 [PAI-1]) were measured in plasma on days 1, 2, 3, 5, 7, and every third day thereafter during the ECMO course. RESULTS: All ECMO day 1 inflammatory biomarkers were significantly elevated in children with abnormal vs. normal neuroimaging. ECMO day 1 and peak levels of IL-6 and PAI-1 were significantly elevated in children who died compared to those who survived to hospital discharge. Tested biomarkers showed no significant association with long-term neurobehavioral outcomes measured using the Vineland Adaptive Behavioral Scales, Second Edition. CONCLUSIONS: High levels of circulating inflammatory, endothelial activation, and fibrinolytic markers are associated with mortality and abnormal neuroimaging in children on ECMO. IMPACT: The inflammatory, endothelial activation, and fibrinolytic profile of children on ECMO differs by primary indication for extracorporeal support. Proinflammatory biomarkers on ECMO day 1 are associated with abnormal neurologic imaging in children on ECMO in univariable but not multivariable models. In multivariable models, a pronounced proinflammatory and prothrombotic biomarker profile on ECMO day 1 and longitudinally was significantly associated with mortality. Further studies are needed to identify inflammatory, endothelial, and fibrinolytic profiles associated with increased risk for neurologic injury and mortality through potential mediation of bleeding and thrombosis.

Genital Hiatus Size and the Development of Prolapse Among Parous Women.

OBJECTIVE: In cross-sectional studies, pelvic organ prolapse is strongly associated with genital hiatus (GH) size. The objective of this study was to estimate prolapse incidence by the size of the GH among parous women followed prospectively. METHODS: Data were derived from a longitudinal study of pelvic floor disorders. Participants were followed annually for 2-9 years. Genital hiatus size and prolapse beyond the hymen were assessed with annual pelvic organ prolapse quantification examinations. Kaplan-Meier methods described prolapse-free survival as a function of GH size. Accounting for changes over time in GH size, lognormal models were used to estimate prolapse-free survival by GH size. This analysis was repeated separately for women who gave birth exclusively by cesarean versus those with at least one vaginal birth. RESULTS: Among 1,492 participants, median age at enrollment was 38 years; 153 (10.3%) developed prolapse over 2-9 years. The cumulative probability of prolapse increased substantially as the size of the GH increased. Lognormal models predicted that the estimated median time to develop prolapse would be 33.4 years for women with a persistent GH of 3 cm; in contrast, the estimated median time to develop prolapse would be 5.8 years for a GH of 4.5 cm or greater. Considering separately women who gave birth by cesarean versus those with at least 1 vaginal birth, GH size drastically modified prolapse risk in both birth groups. CONCLUSIONS: Prolapse incidence is strongly associated with GH size, regardless of delivery mode. These findings suggest that a wider GH is an important predictor of future prolapse risk.

Outcomes of pediatric patients with oncologic disease or following hematopoietic stem cell transplant supported on extracorporeal membrane oxygenation: The PEDECOR experience.

BACKGROUND: Outcomes for patients with oncologic disease and/or after hematopoietic stem cell transplant (HSCT) requiring intensive care unit admission have improved, but indications for and outcomes after extracorporeal membrane oxygenation (ECMO) support in this population are poorly characterized. PROCEDURE: We analyzed data from consecutive patients < 18 years with oncologic disease and/or after HSCT reported to a pediatric ECMO registry by nine pediatric centers in the United States between 2011 and 2018. RESULTS: We identified 18 ECMO patients with oncologic disease and/or HSCT, and 415 ECMO controls matched with a propensity score algorithm based on age, gender, race, severity of illness at admission, and reason for ECMO. The primary indication for ECMO was respiratory failure in 66.7% in the oncologic disease and/or HSCT group, and in 70.7% in the matched ECMO control group. Eleven of 18 patients survived to hospital discharge (61.1%), similar to the matched control group (60.8%), P = 0.979. Children with oncologic disease and/or HSCT had lower mean platelet counts during ECMO and received higher volumes of platelets compared with the control group, mean 14.6 mL/kg/day (standard deviations [SD], 9.8) versus mean 9.3 mL/kg/day (SD, 10.4), P = 0.001. Of the 11 surviving children with oncologic disease and/or HSCT, five sustained new neurologic disorders (45.5%) versus 45 of 222 (20.3%) in the control group, P = 0.061. Bleeding complications were similar in the two groups. CONCLUSIONS: Outcomes of patients with oncologic disease and/or HSCT supported on ECMO in the current era are not significantly different compared with matched ECMO controls and are improved from previously published reports.