TY - JOUR
T1 - Predictors of intracranial hemorrhage during neonatal extracorporeal membrane oxygenation
AU - Hardart, George E.
AU - Fackler, James C.
PY - 1999
Y1 - 1999
N2 - Objective: To identify independent predictors of intracranial hemorrhage (ICH) during neonatal extracorporeal membrane oxygenation (EGMO). Study design: This retrospective cohort consisted of all neonates who did not have an ICH before treatment with ECMO identified in the Extracorporeal Life Support Organization Registry from 1992 to 1995 (n = 4550). Multiple logistic regression analysis was used to identify factors independently correlated with ICH and to develop a model that could be used to predict the risk of ICH in neonates treated with ECMO. Results: ICH was identified in 9.9% of patients. The factors associated with ICH remaining after adjusting for other significant variables (P < .01) were gestational age (GA) <34 weeks (odds ratio [OR] 12.1, 95% confidence intervals [CI] [6.6, 22]), GA 34 to <36 weeks (OR 4.1, CI [2.9, 5.8]), GA 36 to <38 weeks (OR 2.1, CI [1.6, 2.8]) primary diagnosis of sep- sis (OR 1.8, CI [1.4, 2.3]), epinephrine use (OR 1.9, CI [1.5, 2.5]), coagulopathy (OR 1.6, CI [1.1, 2.2]), arterial pH <7.0 (OR 2.5, CI [1.6, 3.9]), and arterial pH 7.0 to <7.2 (OR 1.8 CI [1.3, 2.5]). ICH rates for neonates receiving renovenous versus renoarterial ECMO and for those treated with or without cephalic jugular venous drainage were not significantly different. Conclusions: Gestational age, acidosis, sepsis, coagulopathy, and treatment with epinephrine are major independent factors associated with ICH in neonates treated with ECMO. In particular, GA <34 weeks remains a major barrier for use of current ECMO technologies.
AB - Objective: To identify independent predictors of intracranial hemorrhage (ICH) during neonatal extracorporeal membrane oxygenation (EGMO). Study design: This retrospective cohort consisted of all neonates who did not have an ICH before treatment with ECMO identified in the Extracorporeal Life Support Organization Registry from 1992 to 1995 (n = 4550). Multiple logistic regression analysis was used to identify factors independently correlated with ICH and to develop a model that could be used to predict the risk of ICH in neonates treated with ECMO. Results: ICH was identified in 9.9% of patients. The factors associated with ICH remaining after adjusting for other significant variables (P < .01) were gestational age (GA) <34 weeks (odds ratio [OR] 12.1, 95% confidence intervals [CI] [6.6, 22]), GA 34 to <36 weeks (OR 4.1, CI [2.9, 5.8]), GA 36 to <38 weeks (OR 2.1, CI [1.6, 2.8]) primary diagnosis of sep- sis (OR 1.8, CI [1.4, 2.3]), epinephrine use (OR 1.9, CI [1.5, 2.5]), coagulopathy (OR 1.6, CI [1.1, 2.2]), arterial pH <7.0 (OR 2.5, CI [1.6, 3.9]), and arterial pH 7.0 to <7.2 (OR 1.8 CI [1.3, 2.5]). ICH rates for neonates receiving renovenous versus renoarterial ECMO and for those treated with or without cephalic jugular venous drainage were not significantly different. Conclusions: Gestational age, acidosis, sepsis, coagulopathy, and treatment with epinephrine are major independent factors associated with ICH in neonates treated with ECMO. In particular, GA <34 weeks remains a major barrier for use of current ECMO technologies.
UR - http://www.scopus.com/inward/record.url?scp=0032986621&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0032986621&partnerID=8YFLogxK
U2 - 10.1016/S0022-3476(99)70408-7
DO - 10.1016/S0022-3476(99)70408-7
M3 - Article
C2 - 9931522
AN - SCOPUS:0032986621
SN - 0022-3476
VL - 134
SP - 156
EP - 159
JO - Journal of Pediatrics
JF - Journal of Pediatrics
IS - 2
ER -