N. Ssemakula et al., SURVIVAL OF PATIENTS WITH CONGENITAL DIAPHRAGMATIC-HERNIA DURING THE ECMO ERA - AN 11-YEAR EXPERIENCE, Journal of pediatric surgery, 32(12), 1997, pp. 1683-1689
Background/Purpose: Congenital diaphragmatic hernia (CDH) is associate
d with significant mortality and morbidity. To evaluate the impact of
extracorporeal membrane oxygenation (ECMO) on survival, a review of ou
r experience with CDH patients was initiated. Methods: The authors per
formed a retrospective non randomized analysis of 98 consecutive CDH p
atients who were ECMO candidates, and were symptomatic within the fi r
st day of life, and underwent repair between May 1985 and May 1996. Th
e patients were divided into three groups: Group 1 (n = 38) refers to
patients who were clinically stable and underwent repair before 48 hou
rs of age and did not need ECMO rescue; Group 2 (n = 29) consists of p
atients who underwent repair but required ECMO rescue; and Group 3 (n
= 31) refers to patients who met ECMO criteria preoperatively and requ
ired ECMO for stabilization and later underwent repair on ECMO. The Ka
plan-Meier survival graph was used for survival analysis. Results: Dur
ing the Ii-year span, the overall survival rate of all CDH patients wa
s 72% (71 of 98). The survival rate of patients who did not require EC
MO support was 92% (35 of 38), whereas patients who required ECMO afte
r repair had a 72% (21 of 29) survival rate. These were com pa red wit
h a 48% ,(15 of 31) survival rate for those undergoing repair on ECMO.
The differences in survival among the three groups were statistically
significant using the log-rank test (P=.0018). Conclusions: Survival
was significantly better for infants who underwent successful repair w
ithout ECMO than those who required ECMO rescue pre-or postrepair. The
overall improved survival of CDH patients to 72% compared with histor
ical controls of 38% to 58% may be attributed to ECMO, but the require
ment of ECMO before repair, as well as the presence of congenital anom
alies (P<.01), prematurity (P<.01), the need for a Gore-Tex patch at r
epair (P<.05), prenatal diagnosis at less than 25 weeks' gestation (P<
.01), and the occurrence of an intracranial hemorrhage (P<.01), decrea
ses the chances of survival.