Predictors of postnatal outcome in neonates with gastroschisis

Background/Purpose

The optimal management of neonates with gastroschisis is unclear, and there is a significant morbidity. We performed a review of neonates with gastroschisis treated at our center of pediatric surgery over the last 21 years to determine predictive factors of outcome.

Methods

Single-center retrospective analysis of 79 neonates with gastroschisis (1989-2009) was done. Length of hospital stay (LOS), days of parenteral nutrition (PN), and survival were outcome measures. Univariate and multiple regression analyses were used.

Results

Overall survival was 92%, and primary closure was achieved in 80%. Median LOS was 25 days, and median duration on PN, 17 days. Intestinal atresia, closed gastroschisis, secondary closure, and sepsis were the primary variables associated with poor outcome independent of other variables, but prematurity also affected outcome. Route of delivery and associated malformations were not related to poorer outcome. Necrotizing enterocolitis did not occur in any of our patients.

Conclusion

Outcome in our patients was favorable as measured by survival, LOS, and days on PN. Primary predictors of poor outcome were factors related to short bowel syndrome and secondary closure, indicating a need to further improve treatment of short bowel syndrome.     

Staged repair of giant omphalocele in the neonatal period

Background/Purpose

The aim of this study was to analyze the outcome of giant omphalocele repaired in the neonatal period.

Methods

Twelve consecutive (1997-2004) neonates with giant omphalocele (defect >6 cm with liver herniation) were reviewed. A silo of Prolene mesh (Ethicon) was attached to the fascia and the defect was closed without opening the amniotic sac after sequential reduction. In 2 neonates with ruptured omphalocele a plastic sheet was inserted below the mesh. Data are reported as median and range.

Results

Gestational age was 38 weeks (range, 32-40 weeks) and birth weight was 2.9 kg (range, 1.0-3.1 kg). The final closure was achieved at 26 days (range, 16-62 days). Three neonates (25%) died before final closure (causes: ruptured omphalocele, lung hypoplasia, cardiac anomalies, and intestinal failure). In the 9 surviving neonates, mechanical ventilation was required for 8 days (range, 2-20 days), hospital stay was 42 days (range, 23-73 days), and full enteral feeding was achieved on day 12 (range, 4-53 days). Complications included wound infection in 5 neonates and midgut volvulus in 1. Prophylactic Ladd's procedure was performed laparoscopically at a later stage in 4 children. At laparoscopy, intraperitoneal adhesions were minimal and the central liver did not preclude the operation. The 9 survivors are all well after 46 months (range, 12-67 months).

Conclusions

Giant omphalocele can be safely repaired in the neonatal period without opening the amniotic sac. Intestinal malrotation should be excluded and Ladd's procedure can be performed laparoscopically at a later stage.     

Delivery of gastroschisis patients before 37 weeks of gestation is associated with increased morbidities

Background

Despite advances in the care of neonates with gastroschisis, patients present with significant morbidities. Preterm delivery of neonates with gastroschisis is often advocated to avoid the intestinal damage that may be sustained with prolonged exposure to amniotic fluid. However, preterm delivery may impose additional morbidities to this disease process.

Methods

We conducted a retrospective review of patients with gastroschisis born from 1989 to 2007. Demographic and clinical data were collected. Preterm healthy neonates, with gestational age from 26 to 36 weeks, were used as controls.

Results

Preterm infants with gastroschisis had a 14 times higher risk for any of the recorded morbidities. As compared to term neonates with gastroschisis, preterm neonates with gastroschisis had a higher rate of sepsis, longer duration to reach full enteral feedings, and longer length of stay. Although the preterm infants with gastroschisis were less likely to be small for gestational age at birth, they were as likely as the term infants with gastroschisis to have failure to thrive at discharge and had a greater drop in weight percentile during hospitalization.

Conclusions

Preterm delivery should be avoided because there is no clear benefit to the gut in avoiding derivative injuries. Meticulous attention should be given to the nutritional needs of patients with gastroschisis.     

The incidence and morbidity of adhesions after treatment of neonates with gastroschisis and omphalocele: a 30-year review

Background/Purpose

Adhesive small bowel obstruction (SBO) is a feared complication after correction of abdominal wall defects in neonates. Knowledge of its incidence and potential risk factors in a well-documented group with strict follow-up is needed to guide preventive measures.

Methods

Records of 170 neonates with abdominal wall defects, 59 gastroschisis (GS) and 111 omphalocele (OC), were reviewed focusing on SBO. Risk of SBO was calculated, and potential risk factors were analyzed. Long-term complaints possibly associated with adhesions were assessed through questionnaire.

Results

One hundred forty-seven neonates were operated on, 12 were treated nonoperatively, and 11 patients died shortly after birth. Defects were primarily closed in 128, 7 neonates needed prosthetic mesh, and 12 had a silastic sac inserted. Twenty-six (18%) neonates had SBO, 14 (25%) of 55 with GS, and 12 (13%) of 92 with OC (P = .06). Of the 26 with SBO, 26 (88%) needed laparotomy. Four patients died because of SBO. Most episodes (85%) were in the first year. Sepsis and fascia dehiscence were predicting risk factors for SBO. Abdominal pain and constipation were frequent long-term complaints not significantly associated with SBO.

Conclusions

Adhesive SBO is a frequent and serious complication in the first year after treatment of congenital abdominal wall defects. Sepsis and fascial dehiscence are predictive factors.     

Does gastroschisis reduction require general anesthesia? A comparative analysis

Background

Outcome after gastroschisis repair without general anesthesia is controversial, and published conclusions are variable with no comparative studies.

Aim

The aim of this study was to present a comparative study evaluating outcome after gastroschisis repair with and without general anesthesia.

Methods

An ambispective nonrandomized study of a cohort of 51 neonates born with gastroschisis between July 1998 and December 2003 was performed. Twenty-four neonates (group 1) had conventional reduction under general anesthesia, and 27 (group 2) cotside minimal intervention reductions were without general anesthesia.

Results

Groups were comparable regarding gestational age, birth weight, and quality of eviscerated bowel. Statistical significance (P < .05) was seen between groups 1 and 2 with regard to age at reduction of gastroschisis (5.6 ± 2.5 vs 3 ± 1 hours) and time taken for completion of gastroschisis reduction (58.1 ± 15 vs 49 ± 14 minutes). No statistical significance (P > .05) was seen with respect to start of feeds (10.4 ± 3.6 vs 10.9 ± 4.1 days), duration of total parenteral nutrition (21.5 ± 7.3 vs 22.4 ± 6.8 days), and total hospital (stay 29 ± 10 vs 30 ± 13 days). Admission to the intensive care unit was required in 92% in group 1 for 1 to 6 days vs 7% in group 2 for 3 to 6 days. There was 1 death in group 1 (4%). Total hospital cost in group 1 was £12,283 ± £2438 vs £6208 ± £2120 in group 2 (P = .013).

Conclusions

Neonates with gastroschisis, whose bowel was reduced without general anesthesia, have similar outcomes to those whose bowel was reduced under general anesthesia. Both approaches appear to be safe and effective, but reduction without general anesthesia was cost-effective.     

Demographic and environmental risk factors for gastroschisis and omphalocele in the National Birth Defects Prevention Study

Background

Primary prevention efforts for both gastroschisis and omphalocele are limited by the lack of known risk factors. Our objective was to investigate associations between potential maternal risk factors and gastroschisis and omphalocele within a large population-based sample of participants enrolled in the National Birth Defects Prevention Study (NBDPS).

Methods

Demographic, health-related, and environmental exposure data from the NBDPS were collected from women with expected delivery dates between October 1997 and December 2003. Data were collected on 485 cases of gastroschisis, 168 cases of omphalocele, and 4967 controls.

Results

Women who had offspring with gastroschisis were younger (adjusted odds ratio [AOR], 0.84; 95% confidence interval [CI], 0.81-0.86) and less likely to be black (AOR, 0.54; 95% CI, 0.34-0.85) than controls. They also were more likely to have smoked (AOR, 1.51; 95% CI, 1.12-2.03), taken ibuprofen (AOR, 1.61; 95% CI, 1.23-2.10), and consumed alcohol (AOR, 1.38; 95% CI, 1.06-1.79) than controls. Women who had offspring with omphaloceles were more likely to have consumed alcohol (AOR, 1.53; 95% CI, 1.04-2.25) and be heavy smokers (AOR, 4.26; 95% CI, 1.58-11.52) than controls.

Conclusions

Our results suggest a moderately increased risk of gastroschisis among women who used tobacco, alcohol, and ibuprofen during early pregnancy. A modestly elevated risk was observed for omphaloceles among women who used alcohol during the first trimester and among women who were heavy smokers.     

Canadian Pediatric Surgical Network: a population-based pediatric surgery network and database for analyzing surgical birth defects. The first 100 cases of gastroschisis

Purpose

Outcomes studies for gastroschisis are constrained by small numbers, prolonged accrual, and nonstandardized data collection. The aim of this study is to create a national pediatric surgical network and database for gastroschisis (GS) that tracks cases from diagnosis to hospital discharge.

Methods

The 16-center network serves a population of 32 million. Gastroschisis cases are ascertained at prenatal diagnosis. Perinatal data include maternal risk and fetal ultrasound variables, delivery plan and outcome, a postnatal bowel injury score, intended and actual surgical treatment, and neonatal outcomes. Institutional review board-approved data collection conforms to regional privacy legislation. Deidentified data are centralized and accessible for research through the network steering committee.

Results

To date, 114 cases of pre- and/or postnatal gastroschisis have been uploaded. Of 106 live-born infants (40 [38%] by cesarean delivery), 100 had complete records, and overall survival to discharge was 96%, with a mean survivor length of stay (LOS) of 46 days. Infants treated with attempted urgent closure (61%) had significantly shorter LOS (42 vs 57days; P = .048) but comparable LOS compared with those treated with silos and delayed closure. Fetal bowel dilation 18 mm or greater did not predict a difference in outcome.

Conclusion

Population-based databases allow rapid case accrual and enable studies that should aid in the identification of optimal perinatal treatment.     

Outcome of peripherally inserted central venous catheters in surgical and medical neonates

Background/Purpose

Peripherally inserted central venous catheters (PICCs) are commonly used for neonatal vascular access. The aim of this study was to look at PICC line complication rates and possible predictors of PICC infection in a neonatal intensive care unit.

Method

This was a prospective study of 226 neonates who had PICCs on our neonatal intensive care unit between January 2006 and June 2009. Complete data was available on 218 neonates who had 294 PICC lines. Criteria for catheter-related sepsis was positive blood cultures (peripheral/central) and/or a positive catheter tip culture after removal in the presence of a clinical suspicion of line sepsis.

Results

Of 218 neonates, 132 (169 lines) were medical, and 86 (125 lines) were surgical. Our PICC line infection rate was 17 infections per 1000 catheter-days. Surgical neonates had infection rates of 24.8% compared with 18.3% of medical neonates (P < .18). The odds ratio for a PICC infection was 3.1 (95% confidence interval, 1.64-5.87) if the catheter was in situ for 9 days or more, P < .01. Coagulase-negative staphylococcus was isolated from 55 (89%) of 62 blood cultures.

Conclusions

Our PICC infection rate was 17 per 1000 catheter-days. The length of catheter stay was the only predictor of PICC infection.     

Effects of multidisciplinary prenatal care and delivery mode on gastroschisis outcomes

Background/Purpose

This study examined the effects of multidisciplinary prenatal care and delivery mode on gastroschisis outcomes, with adjustment for key confounding variables.

Methods

This retrospective cohort study included all gastroschisis patients treated at a single tertiary children's hospital between 1999 and 2009. Prenatal care, delivery mode (vaginal vs cesarean section before labor vs after labor), patient characteristics, and clinical outcomes were determined by chart review. Time to discontinuation of parenteral nutrition (PN) was the primary outcome of interest. Effects of multidisciplinary prenatal care and delivery mode were evaluated using Cox proportional hazards regression models that included gestational age, birth weight, sex, concomitant intestinal complications, and year of admission.

Results

Of 167 patients included, 46% were delivered vaginally, 69% received multidisciplinary prenatal care, and median time to PN discontinuation was 38 days. On multivariable modeling, gestational age, uncomplicated gastroschisis, and year of admission were significant predictors of early PN independence. Delivery mode and prenatal care had no independent effect on outcomes, although patients receiving multidisciplinary prenatal care were more likely to be born at term (49% vs 27%, P = .01).

Conclusions

Gestational age and intestinal complications are the major determinants of outcome in gastroschisis. Multidisciplinary prenatal care may facilitate term delivery.     

Determination of risk factors for deep venous thrombosis in hospitalized children

Purpose

Our objective was to determine the time trend and risk factors for deep venous thrombosis (DVT) of the lower extremities among pediatric inpatients.

Methods

This cross-sectional study used the data from the Health Care Cost and Utilization Project Kids' Inpatient Database for the years of 1997, 2000, and 2003 to estimate the DVT prevalence and crude and adjusted prevalence ratios. Patients between the ages of 1 and 17 years and who were hospitalized for at least 4 days were included.

Results

The weighted prevalence of DVT was 4.2 per 1000 hospital discharges (95% confidence interval [CI], 3.4-3.7). Independent of age, the prevalence of DVT was significantly greater in 2000 and 2003 compared to 1997, prevalence ratio (PR) of 1.2 and 1.4 (95% CI, 1.1-1.3 and 1.2-1.4). Using only the 2003 database, adjusted analysis revealed that patients at highest risk were those in the age range of 15 to 17 years (PR, 2.0; 95% CI, 1.6-2.4) and with the following comorbid conditions: obesity (PR, 2.1; 95% CI, 1.5-2.8), inflammatory bowel disease (PR, 1.8; 95% CI, 1.2-2.7), hematologic malignancy (PR, 2.5; 95% CI, 2.0-3.1), and thoracoabdominal (PR, 1.8; 95% CI, 1.6-2.2) or orthopedic (PR, 2.2; 95% CI, 1.7-2.8) operations. Predictors not associated with DVT included sex and diagnosis of trauma.

Conclusions

The discharge diagnosis of DVT of the lower extremities has significantly increased since 1997. In addition, teenagers with underlying disorders are at highest risk for DVT.     

Peroperative Gastrograffin bowel lavage in gastroschisis

Aims

The aim of the study was to evaluate potential benefits in the use of peroperative bowel lavage with Gastrograffin in neonates with gastroschisis.

Methods

A retrospective analysis of newborns with gastroschisis was performed over a 10-year period in 2 centers in the United Kingdom. Two groups were studied wherein one had peroperative bowel lavage with Gastrograffin and the other did not.

Results

Data were collected on 116 patients of whom 93 were suitable for analysis. There were no statistically significant differences in primary closure rate, duration of ventilation, parenteral nutrition, or hospital stay. Intestinal obstruction occurred more frequently in the nonlavage group.

Conclusion

Gastrograffin lavage peroperatively in gastroschisis offers no potential advantage in reducing ventilatory requirements, parenteral nutrition, and hospital stay. It also does not achieve greater primary closure rates, but may reduce the incidence of intestinal obstruction.     

Effects of intrauterine treatment on interstitial cells of Cajal in gastroschisis

Aim

An experimental study was performed to investigate the effects of amnio-allantoic fluid exchange and intrauterine bicarbonate treatment on intestinal damage and interstitial cells of Cajal (ICC) in gastroschisis.

Materials and Methods

Thirteen-day-old fertilized chick eggs were randomly allocated into 4 groups as control, gastroschisis, gastroschisis + amnio-allantoic fluid exchange, and gastroschisis + bicarbonate treatment groups. In the treatment groups, amnio-allantoic exchange and bicarbonate treatments were performed for 3 days, after creating gastroschisis. Specimens were processed for hematoxylin-eosin and c-kit immunohistochemistry on the 18th day of incubation, after macroscopic examination. The intestines were evaluated with light microscopy for the presence of mucosal congestion and muscular and serosal edema. Mean muscular thickness and density of ICC were measured.

Results

Mean muscular thickness significantly increased in the gastroschisis group when compared with control and treatment groups. Labeling intensity, morphology, and localization of the ICC were similar in all groups. Mean ICC density significantly decreased in the gastroschisis group when compared with the control group (P < .01), and it significantly increased after amnio-allantoic fluid exchange treatment (P < .01).

Conclusions

The decrease in ICC density encountered in damaged intestinal loops in gastroschisis was prevented with intrauterine treatment. The beneficial effects of amniotic exchange on intestinal motility may depend on both prevention of intestinal damage and preservation of ICC density and function. The density of ICC might be a reliable numeric parameter both to predict intestinal motility disorders in gastroschisis and to compare the effectiveness of intrauterine treatment methods.     

Familial recurrence of gastroschisis: Literature review and data from the population-based birth registry “Mainz Model”

Background

Familial forms of gastroschisis are considered rare. A search for these forms in a population-based birth registry in 1993 found a recurrence risk of 3.5% among first-degree relatives. Since then, similar investigations in population-based registries have led to contradictory results.

Methods

A search of the population-based birth registry “Mainz Model” for familial cases of gastroschisis and a systematic review of the literature were performed.

Results

The Mainz Model database yielded 1 familial recurrence out of 27 gastroschisis cases. From the literature, 37 affected families could be retrieved. Among 412 gastroschises from population-based registries, 10 familial recurrences have been found. These translate into a recurrence risk of 2.4%, with a strong tendency toward underestimation.

Conclusion

The existing data support the hypothesis that familial recurrence of gastroschisis is much more likely than previously thought.     

Method to our madness: an 18-year retrospective analysis on gastroschisis closure

Background

The advent of preformed silos has facilitated routine bedside placement often without any attempt of intestinal reduction. It is unclear whether a strategy of routine silo (RS) placement with delayed fascial repair is beneficial over attempted primary repair (aPR) and silo placement only for those patients who cannot be reduced. We retrospectively compared clinical outcomes of neonates having aPR to those having RS placement to determine the impact of routine silo use and silo duration on gastroschisis care.

Methods

Neonatal records from patients with gastroschisis at a single children's hospital between 1990 and 2008 were reviewed. Demographic and outcome data were recorded and subjected to statistical analyses. Documentation of attempted intestinal reduction was used as a surrogate marker for aPR. The remaining patients were placed in the RS group.

Results

Two hundred forty-eight neonates with gastroschisis were identified. Thirteen were excluded for congenital or clinical issues which precluded aPR. Of the remaining 235 patients, neonates with RS had significantly more ventilator days (6.2 vs 4.4; P = .0011), more time of total parenteral nutrition (36.5 vs. 28.5; P = .0018), longer length of stay (LOS, 46.5 vs. 40.5; P = .0011), and greater hospital charges ($216,000 vs $172,000; P < .0001) than patients who had aPR. There was no significant difference observed in complications or survival. Linear regression modeling demonstrated that time to closure was significantly related to LOS as an independent variable. Each day to closure was associated with 2.2 extra days of hospitalization and approximately $9557 in hospital charges.

Conclusion

Although limited by retrospective biases, this study demonstrates that time to closure is the most significant variable related to LOS in gastroschisis. This relationship is intuitive since longer time to closure is probably determined by the severity of gastroschisis. The method of closure, by primary repair or silo, is of secondary importance. Conversely, unnecessarily increasing the time to closure may increase the LOS. The speed of reduction, whether through primary repair or by silo, should be guided by physiologic principles.     

Amniotic fluid ferritin as a marker of intestinal damage in gastroschisis: a time course experimental study

Background/Purpose

Intestinal damage (ID) is closely related to morbidity and mortality in gastroschisis. This study was performed to determine the intraamniotic substances that may correlate ID and also to verify their time course levels that would be useful for determining when ID starts in gastroschisis.

Methods

In this study, 13-day-old fertilized chick eggs were used. The amnioallantoic membrane was perforated to create amnioallantoic cavity in all embryos. Gastroschisis was created in gastroschisis group to simulate human gastroschisis. Amnioallantoic fluid samples were collected from the embryos on the 13th to 19th gestational days, and the intestines of each group were harvested for evaluation. Amnioallantoic levels of interleukin-8, ferritin, alkaline phosphatase, and amylase were measured. Serosal thickness of the intestines in each group was evaluated.

Results

Increasing amnioallantoic fluid levels of interleukin-8, alkaline phosphatase, and amylase were found in both groups. In contrast to control group, ferritin levels, as a sign of inflammation, were found increased only in gastroschisis group. Histopathologic examination of intestines in the gastroschisis group showed a significant increase in the serosal thickness especially after the 16th day.

Conclusion

Increases in amnioallantoic fluid levels of ferritin show promise as a marker for determining ID encountered in gastroschisis but warrant further investigation.     

Development and validation of a risk stratification index to predict death in gastroschisis

Background

Gastroschisis is a rare congenital anomaly, the improved surgical management of which has contributed to a survival rate greater than 90%. Development of an accurate risk stratification system to help identify the subset of patients at greatest risk for death may lead to further improvements in outcome.

Methods

Infants with gastroschisis were identified from 16 years of the National Inpatient Sample database and the Kids' Inpatient Database using the International Classification of Diseases, Ninth Revision, Clinical Modification procedure code 54.71 (repair of gastroschisis) and an age of less than 8 days. Logistic regression analysis determined which coexisting diagnoses were significantly associated with death. Odds ratios from the logistic regression model were simplified and used as weighting factors to create an additive index. The index was validated using the 2003 Kids' Inpatient Database data set.

Results

Intestinal atresia, necrotizing enterocolitis, rare cardiac anomalies, and lung hypoplasia were strongly associated with death and used to create a scoring system with a potential range of 0 to 10. Every point increase on the scale of gastroschisis risk stratification index is associated with a 95% relative increase in the likelihood of death.

Conclusion

We have developed a novel index, which is superior to previous classification systems in identifying patients with gastroschisis who are at highest risk for death.     

Infectious complications in infants with gastroschisis: an 11-year review from a referral hospital in southern Thailand

Main Purposes

The study aimed to (1) examine the incidence of infectious complications (ICs) in our referral hospital in southern Thailand in infants with gastroschisis, with analysis of the impact of these complications on outcomes, and (2) identify associated factors to improve the practice at our institution for dealing with this condition.

Methods

A retrospective review of consecutive gastroschisis cases at the major teaching and referral hospital in southern Thailand was conducted for an 11-year period (1996-2006). Cases referred after a primary operation at other hospitals were excluded. The study focused on postoperative nosocomial infections as identified by Centers for Disease Control and Prevention criteria.

Results

Sixty-eight patients with gastroschisis were operated on. Twenty-seven patients (39.71%) underwent primary closure. Mortality was 4 of 68 patients (5.9%). Infectious complication occurred in 43 patients (63.2%). The complications significantly increased mechanical ventilation days (10.8 vs 3.8 days in noncomplicated cases), need for parenteral nutrition (25.3 vs 14.5 days), and postoperative stay (33.7 vs 21.1 days). Common ICs were wound infection (32.35%), isolated septicemia (19.1%), and pneumonia (13.24%). Univariate analysis identified an association between the occurrence of IC and birth order (multigravida), time from birth until arrival at our center (5 hours or more), hypoalbuminemia, hypoglycemia, type of operation (staged closure), use of central venous line, and prolonged use of ventilator. On multiple logistic regression, prolonged referral time, use of a central venous line, multigravida, and staged closure independently predicted the risk of IC.

Conclusion

Infectious complication was significantly related to outcome in gastroschisis cases and should not be overlooked. Our data suggest that prompt referral, limiting central line practice on a selective basis, and an attempt to reduce wound infection in cases that require a temporary silo may improve the overall outcomes.     

Management of a giant omphalocele with an external skin closure system

Background/Purpose

The management of neonates with giant omphalocele remains challenging and multiple strategies have been described. We present the case of a 34-week-old neonate with isolated giant omphalocele managed with an external surgical skin closure system as a component of a staged closure strategy.

Case Presentation

An Inuit boy of 34 weeks gestation was born by urgent Caesarean delivery at an affiliated obstetrical hospital with a giant ruptured omphalocele and loss of abdominal domain. He was transferred to our institution and a silastic silo was fashioned and placed in the operating room. He returned to the operating room several times and was treated by placement of a combined Gore-Tex (WL Gore and Associates, Flagstaff, Ariz)/silastic inlay mesh. An eschar formed over this temporary closure, and we elected to place a dynamic skin closure device to continue gradual bedside reduction. The initial abdominal wall defect was 8.5 cm in transverse diameter and was reduced to 4.5 cm over 3 weeks. Complete closure was subsequently achieved without the need for skin grafting.

Discussion

The use of a dynamic reduction skin closure device has not been documented previously in the pediatric population or in the context of a congenital defect. We describe the use of an external surgical skin closure device in the context of the staged closure of a giant neonatal omphalocele and postulate that such a device may prove useful in the treatment of other congenital tissue defects.     

Predictors of gastroesophageal reflux in neonates with congenital diaphragmatic hernia

Purpose

Gastroesophageal reflux (GER) is observed in 22% to 81% of neonates with congenital diaphragmatic hernia (CDH). The purpose of this study was to identify factors that may predict GER requiring fundoplication in neonates with CDH.

Methods

A retrospective chart review was performed on all neonates with CDH treated at our hospital from June 1997 to June 2005. Preoperative respiratory status, side of the CDH, and method of repair were assessed as predictors of GER and the need for fundoplication.

Results

Of the 42 patients with CDH, 3 died before intervention, leaving 39 patients eligible for study. All but 1 patient survived until discharge. Twenty-one (54%) developed GER of whom 9 (23%) required fundoplication. Although the side of the CDH was not a determinant of GER or the need for fundoplication, patch repair and the need for extracorporeal life support were determinants of both.

Conclusions

Gastroesophageal reflux is common among babies with CDH, although symptoms often resolve without surgical intervention. Infants with CDH defects requiring a patch repair and those requiring advanced physiologic support, especially extracorporeal life support, are likely to develop severe GER necessitating fundoplication. Early recognition and treatment of GER among high-risk patients may shorten hospital stay and minimize patient morbidity. Early fundoplication should be considered for those patients at the highest risk.