Obstetrics & Gynecology:
Benefits of Term Delivery in Infants With Antenatally Diagnosed Gastroschisis
Huang, Jasmine MD; Kurkchubasche, Arlet G. MD; Carr, Stephen R. MD; Wesselhoeft, Conrad W. Jr MD; Tracy, Thomas F. Jr MD; Luks, Francois I. MD, PhD
Program in Fetal Medicine, and the Divisions of Pediatric Surgery and Maternal‐Fetal Medicine, Brown Medical School, Providence, Rhode Island.
Address reprint requests to: Francois I. Luks, MD, PhD, Brown Medical School, Division of Pediatric Surgery, 2 Dudley Street, Suite 180, Providence, RI 02905; E‐mail: firstname.lastname@example.org.
Received November 28, 2001. Received in revised form April 5, 2002. Accepted April 25, 2002.
OBJECTIVE: To test the hypothesis that term gestation offers the best outcome. The relationship between gestational age and the extent of bowel injury in fetuses with gastroschisis is a matter of debate. Early delivery and cesarean delivery have been recommended to limit intestinal damage, but their benefits are unclear.
METHODS: Data on all patients with gastroschisis seen at our institution from 1991 through 2001 were included. Patients were compared based on gestational age: less than 35 weeks, 35–37 weeks, and term (more than 37 weeks) with regard to age at definitive closure, age at first and full feedings, and hospital stay. Statistical significance (P < .05) was determined by analysis of variance and χ2 analysis.
RESULTS: Of the 57 patients, 19.3%, 43.8%, and 36.9% were born at less than 35 weeks, 35–37 weeks, and more than 37 weeks, respectively. Age at definitive closure was significantly higher at 35–37 weeks (5.9 ± 4.6 days) than at more than 37 weeks (1.5 ± 2.3 days) and less than 35 weeks (2.6 ± 2.5 days) (P < .05). A prosthetic pouch (silo) was used more often at 35–37 weeks than at more than 37 weeks or less than 35 weeks (P = .03, χ2). Age at first (P = .04) and full feedings (P < .01) and length of hospitalization (P < .01) were all significantly higher at 35–37 weeks than at more than 37 weeks.
CONCLUSION: Based on a homogeneous cohort of patients in whom gastroschisis was diagnosed antenatally, term delivery results in earlier closure of the defect and shorter time to full feedings. The benefit of early delivery postulated by others cannot be substantiated.
The relationship between gestational age and the extent of bowel injury in infants born with gastroschisis (and ruptured omphalocele) has long been a matter of debate. Timing of delivery continues to be controversial, as some believe that prolonged contact of the exposed bowel to caustic amniotic fluid increases the degree of bowel wall edema, inflammatory peel, and intestinal dysfunction. The edematous bowel is often difficult to reduce, necessitating the use of a silo and secondary closure to safely repair the defect. Preterm delivery of infants with antenatally diagnosed gastroschisis has been recommended to avoid staged procedures and complications.1 An alternative explanation for the intestinal findings implicates the size of the fascial defect. The constricting effect of a small defect on the protruding viscera might be more likely to cause venous congestion, bowel wall edema, and even intestinal necrosis. Despite extensive animal experimentation and multiple clinical studies,2–4 neither prolonged exposure, defect size, nor vaginal delivery as causes of bowel damage have been substantiated.5 Since 1991, we have not, therefore, recommended either early delivery or routine cesarean delivery. The present study, which represents a ten‐year experience with the management of infants born with gastroschisis, was undertaken to examine the relationship between gestational age and outcome.
MATERIALS AND METHODS
All patients with gastroschisis treated at our institution from 1991 through 2001 were studied. The study was approved by the institutional review board at Women & Infants' Hospital of Rhode Island, a tertiary care center with a level three neonatal intensive care unit and immediate availability of a pediatric surgical team. Collected data included gestational age at delivery, type of defect, antenatal diagnosis, birth weight, associated anomalies, hospital of delivery, mode and timing of definitive closure (primary versus delayed repair, preceded by coverage of the exposed viscera with a prosthetic pouch [silo]; if primarily closed, age at definitive closure was defined as 0), age at first oral feeding, and time to full feedings (off parenteral nutrition). Patients were compared based on their gestational age: less than 35 weeks, 35–37 weeks, and term (more than 37 weeks). This grouping was based on the previously proposed approach that infants with gastroschisis be delivered early, between 35 and 37 weeks; it is further substantiated by a scattergram of gestational ages plotted against the four main outcome variables in a pilot study. This preliminary report, on the first 25 patients, was presented as a poster at the Society of Perinatal Obstetricians in 1998.6 Statistical significance (P < .05) was determined utilizing single factor analysis of variance with Scheffé post hoc testing or χ2 analysis (3 × 2 tables), where appropriate. In case of zero numerators, the maximum risk, estimated at 3/n, was used.7 Use of analysis of variance was justified by the finding of similar variances between groups (see Results). Continuous variables are expressed as means ± standard deviations.
During a ten‐year study period, 60 cases of isolated gastroschisis were identified. All but three of these cases (95%) were diagnosed antenatally, and all but one mother delivered at our institution. Gestational age at diagnosis ranged from 15 to 26 weeks; there was no correlation between age at diagnosis and outcome (data not shown). Three fetuses died in utero of unknown causes. The cesarean delivery rate was 24.5% (indications included placenta previa, fetal distress, breech, and postdate induction). This rate was evenly distributed among the three gestational age groups. Twenty‐four patients underwent primary closure of the defect and 33 patients underwent silo placement.
The postnatal survival rate was 96.5%: Two infants (one born at less than 35 weeks and one at 35–37 weeks) died from sequelae of antenatal midgut volvulus and short bowel syndrome. Intestinal atresia was seen in two infants born at less than 35 weeks and in one born at 35–37 weeks. Of the 57 patients evaluated, 19.3% were born at less than 35 weeks, 43.8% between 35 and 37 weeks, and 36.9% at term (more than 37 weeks). Variances were 1.23 weeks in the less than 35 weeks group (33.4 ± 1.1 weeks' gestation), 0.7 weeks in the 35–37 weeks group (36.3 ± 0.8 weeks), and 1.07 weeks in the term group (38.3 ± 1.0 weeks). The differences in variance were not significant (F test). Although the decision to stage the procedure was based on surgical judgement and was therefore not without bias, the use of a silo was more common at 35–37 weeks (19 of 25) than in either less than 35 weeks (six of 11) or more than 37 weeks (eight of 21) (P = .03, χ2 analysis) (Table 1). Age at definitive closure, a more objective measure, was significantly higher at 35–37 weeks (5.9 ± 4.6 days) than at less than 35 weeks (2.6 ± 2.5 days) or more than 37 weeks (1.5 ± 2.3 days; P < .05, analysis of variance) (Figure 1 and Table 1). Age at initiation of feeding (Figure 2 and Table 1) was significantly less in term infants (13.0 ± 5.0 days) than at 35–37 weeks (23.4 ± 19.5 days, P = .04); the difference between less than 35 weeks and 35–37 weeks (15.1 ± 6.0 days) was not significant. Time to full feedings (Figure 3 and Table 1) and length of hospitalization (Figure 4 and Table 1) were both significantly greater at 35–37 weeks than at more than 37 weeks (P < .01).
In gastroschisis and ruptured omphalocele, the main concern is to avoid undue trauma to the exposed intestines during pregnancy and at the time of delivery. Possible therapeutic interventions include the alteration of timing, mode, and place of delivery.5 The debate over place of delivery has been for the most part resolved. The consensus favors delivery of infants with known abdominal defects at or close to a regional perinatal center with immediate access to neonatal intensive care and pediatric surgery capabilities.8 The potential benefits of cesarean delivery have been suggested by some9,10 but disputed by others.11 To date, no conclusive randomized study has established the superiority of one mode of delivery over the other, and most centers consider cesarean delivery in gastroschisis for obstetric reasons only.12
The cause of intestinal damage to some but not all infants with gastroschisis continues to be debated. Some suggest that damage to the intestine is due to a tight fascial ring leading to mesenteric venous congestion or ischemia.12 Although this has been studied experimentally in fetal lambs,3,13 rabbits,4,14 and chicks,15 conclusive evidence has yet to be offered that size of the defect affects outcome of gastroschisis in humans. This may be hampered, in part, by the difficulty in assessing and reliably quantifying the defect on serial ultrasound examinations, as was our experience. Others believe that the damage to the intestinal wall in many of these patients is due to prolonged exposure to the amniotic fluid, and have recommended early delivery1 or even, in selected cases, serial amnioinfusion to dilute the caustic effect of amniotic fluid.16,17 The findings of recent studies on the caustic effect of amniotic fluid2,18,19 have revived the previously held notion that premature delivery of infants with gastroschisis (as soon as pulmonary maturity can be established, between 35 and 37 weeks20) helps minimize intestinal damage.
The difficulty in interpreting many reports on outcome of infants with gastroschisis relates to the heterogeneity of the study populations. Most large cohorts mix inborn patients (in whom the diagnosis is often known antenatally) with patients born in peripheral institutions, and in whom the diagnosis was not known.20 The present cohort is unique in that all but three patients had been diagnosed antenatally and all but one delivered at a tertiary care institution. All but two of these mothers had been counseled by a pediatric surgeon and a neonatologist (two infants were born before the parents had a chance to meet the pediatric specialists). Prenatal diagnosis allowed us to closely monitor the pregnancy (ultrasonography every 2 weeks until the third trimester, and weekly thereafter). Although we acknowledge the use of certain ultrasonographic criteria, such as bowel diameter and bowel wall thickness,3 we have not found these to be useful. In our cohort there were no instances of early delivery dictated by worsening ultrasonographic signs. The fact that, since 1991, we have stopped recommending routine cesarean and/or early delivery for gastroschisis has allowed us to critically evaluate the alleged benefits of preterm delivery. A cohort of 57 patients was divided into three groups, based on gestational age at delivery. (Whereas the numbers of infants born at 35–37 weeks and at more than 37 weeks are similar, there were only 11 patients in the less than 35 weeks group, possibly reducing the power of the analysis for that group.)
In our study, infants born at 35–37 weeks were less likely to undergo primary closure of the abdominal wall defect than infants born at term. If a silo was required, definitive closure was likely to occur later at 35–37 weeks than at more than 37 weeks or less than 35 weeks. Whereas in the latter two groups definitive closure could be achieved by 1.5–2.5 days, infants born between 35 and 37 weeks had closure, on average, at the end of the 1st week of life.
Age at first feedings and age at which a full enteral diet was tolerated were also greater at 35–37 weeks than at more than 37 weeks or less than 35 weeks. Very premature babies (less than 35 weeks) required more time to tolerate full feedings (42 days, versus 21 days in term babies), but even they tended to advance to full feedings more rapidly than infants born between 35 and 37 weeks (50 days). Length of stay was longer for premature infants than for those born at term, but delivery between 35 and 37 weeks did not result in earlier discharge than even the more preterm infants. In a recent study,21 length of hospitalization for infants with gastroschisis was correlated with gestational age. Although inverse linear correlation existed, a scattergram of that patient population suggested an increased length of stay between 35 and 37 weeks' gestation, much like the present study. Others have also found near‐term delivery to be detrimental to feeding and length of hospitalization, although their cohorts were not homogeneous and invited a possible bias due to postnatal transport in some patients.20
Although the presence of an inflammatory peel was not quantified, owing to the somewhat subjective nature of this finding, actual bowel damage was not seen with greater frequency in more mature infants. Intestinal atresia, which is associated with gastroschisis in 10–15%, was seen in three infants, none full term. Fetal midgut volvulus, resulting in severe short bowel syndrome at birth (and leading to death in infancy) was seen twice, in premature infants. None of the live‐born infants presented with major extraintestinal anomalies. In summary, prolonging gestation to term in these infants was not associated with a higher incidence of complications or associated defects.
We conclude that early delivery (before 37 weeks) offers no advantage in terms of primary closure of the abdominal wall defect, time to full bowel function, or length of hospitalization, and may in fact delay these milestones. Because duration of exposure of the exteriorized viscera to amniotic fluid does not correlate with bowel damage, we find it difficult to justify antenatal interventions such as early induction of labor or serial amnioinfusion to dilute amniotic fluid.
1. White JJ. Preterm C-section delivery for fetal gastroschisis. J Pediatr Surg 1998;33:1849.
2. Luton D, de Lagausie P, Guibourdenche J, Peuchmaur M, Sibony O, Aigrain Y, et al. Influence of amnioinfusion in a model of in utero created gastroschisis in the pregnant ewe. Fetal Diagn Ther 2000;15:224–8.
3. Langer JC, Bell JG, Castillo RO, Crombleholme TM, Longaker MT, Duncan BW, et al. Etiology of intestinal damage in gastroschisis, II. Timing and reversibility of histological changes, mucosal function, and contractility. J Pediatr Surg 1990;25:1122–6.
4. Albert A, Julia MV, Morales L, Parri FJ. Gastroschisis in the partially extraamniotic fetus: Experimental study. J Pediatr Surg 1993;28:656–9.
5. Kurkchubasche AG. The fetus with an abdominal wall defect. Med Health R I 2001;84:159–61.
6. Papadakis K, Luks FI, Carr SR, Lessin MS, Wesselhoeft CW, DeLuca FG. Effect of gestational age in gastroschisis: Benefit of term delivery. Am J Obstet Gynecol 1998;178:157S.
7. Hanley JA, Lippman-Hand A. If nothing goes wrong, is everything all right? Interpreting zero numerators. JAMA 1983;249:1743–5.
8. Quirk JG Jr, Fortney J, Collins HB 2nd, West J, Hassad SJ, Wagner C. Outcomes of newborns with gastroschisis: The effects of mode of delivery, site of delivery, and interval from birth to surgery. Am J Obstet Gynecol 1996;174:1134–8.
9. Sakala EP, Erhard LN, White JJ. Elective cesarean section improves outcomes of neonates with gastroschisis. Am J Obstet Gynecol 1993;169:1050–3.
10. Dunn JC, Fonkalsrud EW, Atkinson JB. The influence of gestational age and mode of delivery on infants with gastroschisis. J Pediatr Surg 1999;34:1393–5.
11. Bethel CA, Seashore JH, Touloukian RJ. Cesarean section does not improve outcome in gastroschisis. J Pediatr Surg 1989;24:1–3.
12. Snyder CL. Outcome analysis for gastroschisis. J Pediatr Surg 1999;34:1253–6.
13. Langer JC, Longaker MT, Crombleholme TM, Bond SJ, Finkbeiner WE, Rudolph CA, et al. Etiology of intestinal damage in gastroschisis. I: Effects of amniotic fluid exposure and bowel constriction in a fetal lamb model. J Pediatr Surg 1989;24:992–7.
14. Albert A, Sancho MA, Julia V, Diaz F, Bombi JA, Morales L. Intestinal damage in gastroschisis is independent of the size of the abdominal defect. Pediatr Surg Int 2001;17:116–9.
15. Aktug T, Ucan B, Olguner M, Akgur FM, Ozer E. Amnioallantoic fluid exchange for prevention of intestinal damage in gastroschisis II: Effects of exchange performed by using two different solutions. Eur J Pediatr Surg 1998;8:308–11.
16. Sapin E, Mahieu D, Borgnon J, Douvier S, Carricaburu E, Sagot P. Transabdominal amnioinfusion to avoid fetal demise and intestinal damage in fetuses with gastroschisis and severe oligohydramnios. J Pediatr Surg 2000;35:598–600.
17. Luton D, de Lagausie P, Guibourdenche J, Oury J, Sibony O, Vuillard E, et al. Effect of amnioinfusion on the outcome of prenatally diagnosed gastroschisis. Fetal Diagn Ther 1999;14:152–5.
18. Akgur FM, Ozdemir T, Olguner M, Aktug T, Ozer E. An experimental study investigating the effects of intraperitoneal human neonatal urine and meconium on rat intestines. Res Exp Med (Berl) 1998;198:207–13.
19. Morrison JJ, Klein N, Chitty LS, Kocjan G, Walshe D, Goulding M, et al. Intra-amniotic inflammation in human gastroschisis: Possible aetiology of postnatal bowel dysfunction. Br J Obstet Gynaecol 1998;105:1200–4.
20. Simmons M, Georgeson KE. The effect of gestational age at birth on morbidity in patients with gastroschisis. J Pediatr Surg 1996;31:1060–1.
21. Driver CP, Bruce J, Bianchi A, Doig CM, Dickson AP, Bowen J. The contemporary outcome of gastroschisis. J Pediatr Surg 2000;35:1719–23.
This article has been cited 3 time(s).
Cochrane Database of Systematic ReviewsElective preterm birth for fetal gastroschisisCochrane Database of Systematic Reviews
American Journal of Obstetrics and GynecologyMetaanalysis of the prevalence of intrauterine fetal death in gastroschisisAmerican Journal of Obstetrics and Gynecology
European Journal of Pediatric SurgeryIs there Unity in Europe? First Survey of EUPSA Delegates on the Management of GastroschisisEuropean Journal of Pediatric Surgery
© 2002 The American College of Obstetricians and Gynecologists