Materials and Methods

This was a retrospective cohort study of patients with fetal omphalocele referred to the Center for Prenatal Pediatrics at Columbia University Medical Center from May 2003 to July 2010. Clinical investigation was approved by the Columbia University Institutional Review Board (protocol AAAE0499).

All cases of prenatally diagnosed fetal omphalocele were identified using our computerized obstetric ultrasound and perinatal databases. Subjects were included in the cohort if they underwent at least 1 antenatal sonographic evaluation at the referral center that yielded an antenatal diagnosis of omphalocele. Inclusion also required that the pregnancy result in a live birth with subsequent neonatal care at Columbia. Exclusion criteria included elective termination of pregnancy, subjects with insufficient postnatal data, a documented plan for postnatal comfort care, postnatal change in diagnosis, and intrauterine demise or spontaneous pregnancy loss prior to 24 weeks’ gestational age.

Sonographic examinations were performed by staff perinatal sonographers using an Acuson 128XP (Siemens Medical Solutions USA, Inc, Malvern, PA), an HDI 3000 and HDI 5000 (Philips Medical Systems, Bothell, WA), an HD11 (Philips Medical Systems), and a Voluson 730 Expert, Voluson E6, Voluson E8, and Voluson I (GE Healthcare, Wauwatosa, WI) ultrasound units. All sonographers were registered by the American Registry for Diagnostic Medical Sonography.

For each subject, omphalocele diameter, AC, FL, and HC were recorded for all antenatal ultrasound studies possessing satisfactory archived measurements. The fetal omphalocele diameter was defined as the maximal omphalocele diameter in an axial plane at the level of omphalocele insertion into the fetal abdominal wall.

The AC, FL, and HC were measured according to the definitions established by the American Institute of Ultrasound in Medicine (AIUM) practice guideline for performing obstetric ultrasound. According to the AIUM, the AC “should be determined at the skin line on a … transverse view at the level of the junction of the umbilical vein, portal sinus, and fetal stomach when visible.” Therefore, when fetal omphalocele distorted the expected fetal intraabdominal anatomy, the AC was captured at an axial level that best included these usual landmark structures. Figure 1 illustrates how the omphalocele diameter and AC measurement were obtained in a sample patient. Ultrasound studies were also reviewed for the presence of extracorporeal liver contained within the omphalocele as well as associated fetal structural malformations.

Fetal omphalocele diameter and abdominal circumference

A , Fetal omphalocele diameter. Fetal omphalocele diameter was defined as the maximal omphalocele diameter in an axial plane at the level of omphalocele insertion into the fetal abdominal wall. In this image of a 30 week fetus with omphalocele, the dotted line indicates an appropriate omphalocele diameter measurement (5 cm). B , Abdominal circumference. The abdominal circumference measurement is indicated by the dotted yellow circle in the same fetus. Because of the distortion resulting from the ventral wall defect, the stomach and vascular structures are anteriorly located within the fetal abdomen. Red arrows indicate venous return vessels.

Images are courtesy of Chia-Ling Nhan-Chang, MD.

Montero. Fetal omphalocele ratios. Am J Obstet Gynecol 2011.

Maternal and neonatal electronic medical records (Eclipsys Xa; Eclipsys Corp, Atlanta, GA) were reviewed for pregnancy and postnatal outcomes data. Maternal electronic medical records were reviewed to establish estimated delivery dates as well as to record information regarding the gestational age at time of each sonographic evaluation, fetal karyotype, and obstetrical outcomes. Delivery outcomes were determined, including termination of pregnancy, intrauterine death, and deliveries resulting in live births. Maternal records were screened to determine whether there was a documented decision prior to delivery to pursue postnatal comfort care. Pregnancy dating was established using a standardized institutional ultrasound dating policy.

Neonatal electronic medical records were reviewed for postnatal outcomes including type of surgical repair and selected neonatal complications. The primary outcome was defined as inability to perform primary operative repair of the omphalocele within the neonatal period. Staged surgical repair or delayed closure with silo procedures therefore both satisfied the primary outcome.

Secondary outcomes that were collected included length of neonatal intensive care unit (NICU) stay, time to first oral feed, time requiring mechanical ventilation, number of surgical procedures prior to age 1 year, and neonatal death. Neonatal records were similarly screened to determine whether there was a plan in place for postnatal comfort care.

Characteristics of subjects who underwent primary operative closure were compared with those who required more than one operation to achieve abdominal closure. Continuous variables were compared using the student t test where appropriate. The Fisher’s exact test was used for comparing categorical data between groups. All tests were 2 tailed, with P < .05 considered to be statistically significant. Statistical analyses were performed using STATA 10 (StataCorp LP, College Station, TX).

Because of varying published definitions of giant omphalocele in the literature, several established definitions were tested as predictors of inability to achieve primary operative omphalocele closure. Omphalocele of 4 cm or greater diagnosed prior to 24 weeks’ gestational age, omphalocele of 4 cm or greater diagnosed at any gestational age, omphalocele of 5 cm or greater diagnosed prior to 24 weeks’ gestational age, omphalocele of 5 cm or greater diagnosed at any gestational age, and omphalocele containing extracorporeal liver were all evaluated. For each candidate giant omphalocele, test performance characteristics were calculated.

Fetal omphalocele ratios were calculated using available data from each ultrasound study by dividing omphalocele diameter by AC, FL, or HC measurements, resulting in omphalocele (O)/AC, O/FL, and O/HC ratios, respectively. For each omphalocele ratio test (O/AC, O/FL, and O/HC), a receiver operating characteristic curve (ROC) was generated to identify the optimal test-specific cutoff in the prediction of the primary outcome. In cases involving multiple omphalocele ratios determinations throughout a given pregnancy, the first available measurement was considered for ROC generation.

Using the respective optimal test cutoffs to determine pregnancies that screened positive for the inability to achieve primary operative omphalocele repair, test performance characteristics were analyzed for each omphalocele diameter ratio, as described for the giant omphalocele candidates above. Ratios were also tested for their ability to predict secondary neonatal morbidities.