Full text
Clinical Profile and Outcomes of Neonates with Intestinal Obstruction: A Retrospective Study from Karami hospital Mohammed Jabbar Kadhim Assistant professor in pediatric surgery, Department of surgery, College of Medicine, Wasit University, Kut, Iraq Abstract Background: Neonatal intestinal obstruction (NIO) is a major percentage of the surgical emergencies that are experienced during the neonatal days. It is a broad range of both inherited and acquired defects of the gastrointestinal tract. The timely surgical intervention and prompt diagnosis is essential to avoid bowel ischemia, perforation, sepsis and short bowel syndrome. Even with the development of neonatal treatment and surgical methods, NIO still remains a challenge to diagnosis and treatment, especially in resourcerestricted cases. Methods: The proposed study is a retrospective observational study that will be performed at the Karami hospital where the medical records of all neonates diagnosed with intestinal obstruction between January 2019 and December 2023 will be reviewed. The information gathered was on demographic data, antecedent history, clinical picture, radiological results, operative data, histopathological diagnoses, postoperative results, and mortality. The statistical analysis was conducted in SPSS 26 version and the variables were categorical (demonstrated by frequencies and percentages), and continuous (mean and standard deviation). Resuls: The study involved 87 neonates; a majority of them were male (59.8). Etiological causes Jejunoileal atresia (34.5%), malrotation with volvulus (21.8%), and Hirschsprungs disease (16.1), were the most common etiological causes. Abnormalities that were identified by the use of antenatal ultrasounds were indicative of an intestinal obstruction in 32 percent of the cases. Main clinical manifestations consisted of bilious vomiting (87.4%), abdominal distension (70.1%), and inability to pass meconium (56.3). Initial evaluation was made universal with plain abdominal radiography with contrast enema and upper gastrointestinal series playing a crucial role in establishing the degree and cause of obstruction. Every patient was subjected to definitive surgical management, such as resection and anastomosis, Ladd procedure, duodenoduodenostomy, or pull-through surgeries. One quarter of the patients experienced postoperative complications (25.3) with wound infection (9.2), anastomotic leak (5.7), and sepsis (6.9). The total mortality rate stood at 6.9, and it mainly was related to late presentation and septic problems. The average stay was 12.4 days in months. Conclusion: Neonatal intestinal obstruction is a severe condition with a high index of suspicion and multidisciplinary approach that is necessary. Diagnostic intervention, resuscitation and surgery can and should be performed timely. Our statistics show acceptable morbidity and mortality rates relative to the published international standards. The further progress in the areas of enhancing prenatal diagnosis, the optimal perioperative treatment, and the prolongation of the follow-up work are necessary to increase survival rates and the quality of life of the affected infants. More Information How to cite this article: Kadhim MJ. Clinical Profile and Outcomes of Neonates with Intestinal Obstruction: A Retrospective Study from Karami hospital. Eur J Med Health Res, 2025;3(4):238-45. DOI: 10.59324/ejmhr.2025.3(4).35 Keywords: Neonate, intestinal obstruction, atresia, malrotation, Hirschsprung’s disease, outcome, surgery. This work is licensed under a Creative Commons Attribution 4.0 International License. The license permits unrestricted use, distribution, and reproduction in any medium, on the condition that users give exact credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if they made any changes.
EUR J MED HEALTH RES Volume 3 | Number 4 | 2025 239 Introduction Neonatal intestinal obstruction (NIO) is one of the most frequently occurring surgical crises in underage patients, comprising about 3500 surgical admissions out of the total 8000 to 15000 pediatric surgical admissions in the first month of life, respectively [13]. It is all mechanical obstruction of the gastrointestinal tract that fails to allow intestinal contents to pass normally and can occur at any point between the duodenum and the rectum [3]. A large variety of congenital defects can lead to the condition including atresia, stenosis, malrotation or such functional disorders like Hirschsprung disease [4]. NIO normally has non-specific clinical presentation especially during the initial stages, and therefore, it is difficult to be diagnosed, thus requiring clinicians to have high index of suspicion. The typical symptoms are bilious vomiting, abdominal distension, inability to pass on the meconium within the 24–48 hours of birth, and poor feeding. Early diagnosis or mismanagement can cause severe complications like bowel ischemia, perforation, peritonitis, sepsis and short bowel syndrome, particularly when the extent of resection is extensive [6]. Early detection and treatment play a vital role in intestinal obstruction in newborns as, after 48 hours, the death rate is high in case of a late diagnosis. NIO etiology is diverse with respect to the geographic location, genetics, and the availability of prenatal care. Antenatal ultrasound has enhanced the detection of diseases during the early stages of pregnancy in developed nations, enabling the preparation and management of the disease at birth in a better way than before [8]. Nevertheless, late presentation is still prevalent in most lowand middle-income environments, owing to a lack of prenatal imaging and awareness of the primary healthcare provider [8]. The modalities used in diagnosis are plain abdominal radiography, upper gastrointestinal contrast studies, contrast enemas, and abdominal ultrasonography. Every one has its advantages and disadvantages, and the decision made will be influenced by the level and nature of the obstruction suspected to exist [5]. Most complete mechanical obstruction cases are still treated by surgery, whose outcome is highly contingent upon the speed of surgical intervention, and the comorbidities responsible for the obstruction, including prematurity or congenital heart disease ¹ Although the surgical practice and neonatal intensive care have been improved, mortality in neonates with advanced illness or with the problems of the septic process remains high [12]. Follow-up of long term is necessary especially in patients who have been involved in long resections or patients with underlying motility disorders. Lack of nutritional support, failure to grow, frequent intestinal problems risk these infants to lifelong care treatment with pediatric surgeons, gastroenterologists, dietitians, and developmental specialists [6]. The given study is an attempt to provide our institutional experience of the neonatal intestinal obstruction within the five-year interval examining demographics, clinical presentation, diagnosis methods, surgery, and short-term outcome. Our research will add to the existing literature on this serious condition and will shed light on the present-day management practice in our location. Materials and Methods This was a retrospective observational study that was carried out at the Department of Pediatric Surgery, Alkarame Teachings hospital which is the tertiary referral center in case of neonatal surgical emergency. The research will take a period of five years, i.e., January 1, 2019, to December 31, 2023. and will comply with the ethical principles described in the Declaration of Helsinki. Study Design and Setting A chart review was conducted on all neonates less than 28 days of age and diagnosed with intestinal obstruction who received surgical intervention within the study time frame [1]. Clinical suspicion was used to diagnose the disease based on radiological imaging and confirmed during surgery or histopathologically [2]. They were identified using hospital electronic medical records and surgical logbooks [3]. Inclusion and Exclusion Criteria • Inclusion criteria: • Neonates aged ≤28 days • Mechanical intestinal obstruction diagnosed. • Experienced final surgical care in our institution. • Exclusion criteria: • Functional bowel disorders (e.g. meconium plug syndrome) • Unfinished or unavailable medical records. • Conservative surgical management of patients. Data Collection A structured data extraction form that was designed by the research team was used to collect the data. Variables included: Demographic Data: • Gender • Gestational age • Birth weight • Mode of delivery • Apgar score at 5 minutes Antenatal History: • Maternal age and parity • Presence of polyhydramnios
EUR J MED HEALTH RES Volume 3 | Number 4 | 2025 240 Categories of data: Fetal ultrasonography. • Mother comorbidity (e.g. diabetes) Clinical Presentation: • Age at presentation • Symptoms (bilious vomiting, abdominal distension, failure to pass meconium) • Findings of physical examination. Diagnostic Investigations: • Plain abdominal X-ray • Gastrointestinal series, upper. • Contrast enema • Abdominal ultrasound Surgical Data: • Type of procedure performed • Findings of operation (degree and type of obstruction). Intraoperative complications Intraoperative complications are associated with inadequate planning during the preoperative phase or during the execution of the surgical operation. Intraoperative complications Intraoperative complications Intraoperative complications refer to wrong planning in the preoperative phase or in the surgical operation itself. Histopathological Findings: • Diagnostic confirmation (e.g. Hirschsprungs disease, atresia etc.) Postoperative Outcomes: • Length of hospital stay • Complications (infection of the wound, anastomotic leakage, sepsis) • In-hospital death rates in 30 days of surgery. Two investigators extracted all data independently and cross-verified them to reduce the chances of biasing [5]. Diagnostic Evaluation The first examination was carried out with a history and physical examination. Clinical suspicion was followed by the orders of radiological investigations: Plain abdominal X-ray was carried out in all the patients as the initial-line investigation [7]. • Selective upper gastrointestinal series was administered when the diagnosis was duodenal atresia or malrotation. Contrast enema was used to mark the grade and character of distal obstruction, especially in cases of suspected Hirschsprung disease or colonic atresia [9]. • In the presence of suspicion of perforation or free fluid, abdominal ultrasound was used [10]. Surgical Management Every patient was subjected to conclusive surgery based on the underlying pathology. Procedures included: • End to end anastomosis and resection of jejunoileal atresia. Malrotation with volvulus: the procedure by Ladd. • Duodenal atresia treated with duodenoduodenostomy. • Transanal or Swenson pull-through of Hirschsprung disease. • Colostomy or ileostomy in the case of critically ill patients, or when the procedure has to be done in stages. Operative notes recorded intraoperative findings and were compared with histopathological reports where there are operative histopathological reports. Surgical procedures were done by senior pediatric surgeons or under their direct supervision [11]. Statistical Analysis The analysis of data was done using IBM SPSS statistics version 26.0 (IBM Corp., Armonk, NY, USA). Demographic and clinical characteristics were summarized by using descriptive statistics. Frequencies and per centages were used to represent categorical variables and mean ± standard deviation was used to represent continuous variables. The p-value was taken as statistically significant at a p-value of less than 0.05. Results Demographic Characteristics Table1 was Showered Demographic Characteristics of Patients, there were 52 cases of male patients and 35 cases of female patients (59.8% and 40.2% respectively). About 24.1% (n=21) of the patients were preterm (less than 37 gestation weeks), and the rest (75.9) (n=66) were full term. The average weight at birth was 2850/510 grams and mean age at operation was 4.2 days/2.1 and most of the operations were done during the first 72 hours of life. Table 1: Demographic Characteristics of Neonates with Intestinal Obstruction (n = 87) Variable n (%) or Mean ± SD Male 52 (59.8%) Female 35 (40.2%) Preterm (<37 weeks) 21 (24.1%) Term (>37 weeks) 66 (75.9%) Mean Birth Weight (grams) 2850 ± 510 Mean Age at Surgery (days) 4.2 ± 2.1 Etiology of Intestinal Obstruction Jejunoileal atresia was the most prevalent cause of any intestinal obstruction, 34.5% (n=30) of all. The subsequent ones were malrotation with volvulus (21.8%, n=19) and Hirschsprung disease (16.1., n=14) . In 12.6% (n=11) of the patients, duodenal atresia was the cause, whereas meconium ileus (6.9, n=6), imperforate anus (4.6, n=4), and NEC-related obstruction (3.4, n=3) were less common causes Table 2 and Figure 1.
EUR J MED HEALTH RES Volume 3 | Number 4 | 2025 241 Table 2: Distribution of Etiological Causes of Intestinal Obstruction Among Studied Neonates (n = 87) Diagnosis n (%) Jejunoileal Atresia 30 (34.5%) Malrotation with Volvulus 19 (21.8%) Hirschsprung’s Disease 14 (16.1%) Duodenal Atresia 11 (12.6%) Meconium Ileus 6 (6.9%) Imperforate Anus 4 (4.6%) Necrotizing Enterocolitis (NEC) 3 (3.4%) Figure 1: Distribution of Causes of Neonatal Intestinal Obstruction Antenatal and Clinical Presentation Intestinal obstruction was identified in Tab,e 3 consisting of 32 cases (n=28) of abnormalities that were identified using antenatal ultrasound. These were polyhydramnios among 28 patients and bowel distension of the fetus in 19 cases. The most common presenting symptom was bilious vomiting with 87.4 (n=76) appearing in all of the patients. Also, 56.3% (n=49) and 70.1% (n=61) of neonates were observed to fail meconium within the first 48 hours and abdominal distension respectively (See Figure 2). Table 3: Antenatal Findings and Clinical Presentation Features in Neonates with Intestinal Obstruction (n = 87) Feature n (%) Polyhydramnios 28 (32.2%) Fetal Bowel Dilatation 19 (21.8%) Bilious Vomiting 76 (87.4%) Failure to Pass Meconium 49 (56.3%) Abdominal Distension 61 (70.1%) Diagnostic Modalities Used All patients underwent plain abdominal radiography as the initial diagnostic tool. Additional investigations included (Table 4 and Figure 3) • Contrast enema in 56.3% (n=49) of cases • Upper gastrointestinal series in 16.1% (n=14) • Abdominal ultrasound in 25.3% (n=22) Plain X-ray typically showed multiple air-fluid levels or a double-bubble sign in duodenal cases, while contrast studies helped delineate the level and nature of obstruction. Figure 2: Radiographic Findings in Common Diagnoses (Double-bubble sign for duodenal atresia, Corkscrew appearance for malrotation with volvulus,Transition zone for Hirschsprung’s disease,Multiple air-fluid levels for mid or distal intestinal obstruction) Table 4: Diagnostic Imaging Modalities Utilized in the Evaluation of Neonatal Intestinal Obstruction (n = 87) Investigation n (%) Plain X-ray 87 (100%) Contrast Enema 49 (56.3%) Upper GI Series 14 (16.1%) Abdominal Ultrasound 22 (25.3%) Figure 3: Comparison of Mortality Rates Across Different Types Surgical Procedures Performed Definitive surgical correction was done on all patients. Resection and end-to-end anastomosis (39.1% n=34) was the most commonly used procedure, and the next processes were Ladd procedure on malrotation (21.8%, n=19) and duodeno-duodenostomy (12.6%, n=11). In Hirschsprung disease, Swenson or transanal pullthrough had been done in 16.1% (n=14) of cases. In 6.9 percent (n=6) of the patients, temporary stomas were
EUR J MED HEALTH RES Volume 3 | Number 4 | 2025 242 designed because of poor general condition or intraoperative findings( Table 5 and Figure 4). Table 5: Types of Surgical Procedures Performed According to Underlying Pathology (n = 87) Procedure n (%) Resection and Anastomosis 34 (39.1%) Ladd’s Procedure 19 (21.8%) Duodeno-duodenostomy 11 (12.6%) Swenson / Transanal Pull-through 14 (16.1%) Colostomy / Ileostomy 6 (6.9%) Other 3 (3.4%) Figure 4: Length of Hospital Stay by Diagnosis Postoperative Complications The rate of postoperative complications was 25.3 (n=22) (Table 6). The most frequent complications were: • Wound infection in 9.2% (n=8) • Anastomotic leak in 5.7% (n=5) • Sepsis in 6.9% (n=6) • Postoperative ileus in 3.4% (n=3) The complications were more commonly noted in preterm infants and in those who came late or had the signs of sepsis. (See Figure 5). Table 6: Postoperative Complications Following Surgical Correction of Intestinal Obstruction (n = 22, 25.3%) Complication n (%) Wound Infection 8 (9.2%) Anastomotic Leak 5 (5.7%) Sepsis 6 (6.9%) Postoperative Ileus 3 (3.4%) Outcome Summary The overall mortality rate was 6.9% (n=6), primarily among patients who presented late or developed septic complications. Most deaths occurred in cases of malrotation with volvulus and NEC, likely due to associated bowel ischemia or perforation. The mean hospital stay was 12.4 days ± 6.2, with longer stays observed in patients who developed postoperative complications or required intensive care support (See Table 7, Figure 6 and Figure 7). Figure 5: Timeline of Patient Management from Admission to Discharge Table 7: Short-Term Outcomes Following Surgical Management of Neonatal Intestinal Obstruction (n = 87) Outcome n (%) or Mean ± SD Discharged Well 81 (93.1%) Died 6 (6.9%) Mean Length of Hospital Stay 12.4 ± 6.2 days Figure 6: Trends in Annual Incident of Neonatal Intestinal Obstruction (2019-2023) Figure 7: Age at Surgery vs. Gestational Age
EUR J MED HEALTH RES Volume 3 | Number 4 | 2025 243 Discussion Neonatal intestinal obstruction (NIO) is among the most severe cases of surgical emergencies in infants since its timely detection and intervention are required to avoid serious complications like bowel ischemia, perforation, sepsis, and short bowel syndrome, which may be life-threatening in infants [1]. The paper is a retrospective study that presents a summary of the clinical presentation, diagnostic results, operative procedures, and prognosis of 87 cases of NIO while admitted to Al-karame Teachings hospital within a fiveyear interval. Demographic Characteristics and Etiology Our cohort (59.8) is dominated by males which is consistent with some studies carried out across the globe2 3. Although it is not yet clear what causes this gender gap, it might be connected to the natural biological distinctions or the trends in references, which give preference to men infants in some cultures4. The proportion of preterm infants was 24.1% and that is a little bit lower than some low-resource settings with a limited antenatal care and more frequent late presentation have been reported9. The most frequent cause of NIO in our series (34.5 per cent) was jejeunoileal atresia, then malrotation with volvulus (21.8 per cent) and Hirschsprung disease (16.1 per cent). These results are relatively congruent with the published literature in both highand low-income countries 12, but deviations are possible because of differences in prenatal detection of the condition and referral. Duodenal atresia was less common whereas NEC-associated obstruction was uncommon and only seen in preterm infants, presumably because of the related immaturity of the intestine and other risk factors including hypoxia or infection. The geometry and socioeconomic results are very relevant in the distribution and presentation of neonatal intestinal obstruction [4]. Clinical Presentation and Diagnosis The most frequent symptom of presentation (87.4%), which is a criterion of red-flagging symptoms of intestinal obstruction, was bilious vomiting. Inability to pass meconium in the first 48 hours was observed in 56.3 percent of patients especially those with Hirschsprung disease or distal atresia. There was 70.1 percent abdominal distension that is normally more intense in distal obstruction. Abnormalities identified using pregnancy ultrasound to suggest intestinal obstruction were identified in 32% of cases, mostly polyhydramnios and fetal bowel dilation. This rate is similar to the one recorded in developed nations but lower than what can be attained when third-trimester scanning is done on a regular basis [8]. Prenatal diagnoses can be performed early and enhance postnatal outcomes as well as perinatal planning [9]. Nevertheless, ultrasound access and well trained ultrasonographers are scarce in most resource starved environments. Initial evaluation was done by plain abdominal radiography which was universally used, and in selected cases, contrast enema and upper gastrointestinal series were used. Contrast enema were specifically effective in the diagnosis of distal obstruction, particularly in the diagnosis of Hirschsprung disease or colonic atresia. Ultrasound was also used in situations of possible perforation or of free fluid and was used as an appendix in making management decisions [5]. Contrast enema is the gold standard in the diagnosis of distal obstruction particularly in suspected Hirschsprung's disease [12]. Surgery and Surgery Results Every patient was exposed to definitive surgery correction based on the underlying pathology. Resection and end-to-end anastomosis were the most common ones, and the next most common procedure was the Ladd procedure in cases of malrotation and the duodenal atresia duodenal duodenostomy. The Hirschsprung disease was treated with pull-through operations, and temporary stomas were undertaken in the very ill patients or those who needed time to undergo management. The decision to follow this or that procedure was made depending on the extent of obstruction and the overall condition of the patient during surgery. Complications were 25.3 percentage of the post operation patients, wound infection (9.2), anastomotic leak (5.7), and sepsis (6.9). Such complication rates can be compared to the recent reports7. The complications were more common in preterm babies and those born late, or with the signs of septicemia. There were 6.9 recorded cases of mortality with most of them being patients who reported advanced disease or with septic complications. The majority of deaths were associated with malrotation with volvulus and NEC, probably because of the bowel ischemia or perforation. These results have revealed the significance of timely diagnosis and timely surgical intervention of survival. The average length of stay was 12.4 days, with the extended stay noted among those patients who either got postoperative complications or needed the intensive care facility. These findings are in line with the published statistics in both resource-limited and highincome environments. Late surgery after 48 hours is a major cause of death in neonatal obstruction of the intestine7. Restrictions and Future Research There are a number of limitations associated with this study. First, it is a retrospective chart review, which is
EUR J MED HEALTH RES Volume 3 | Number 4 | 2025 244 prone to missing or incomplete data, particularly in the long-term follow-up. Second, the sample size is comparatively so small and represents one institution which reduces the generalizability of the results. Third, the lack of a control group does not allow comparing it directly with other centers or populations. In spite of these shortcomings, this research work adds to the body of knowledge on NIO, offering local epidemiological information, outlining the existing practices in management and finding ways of improvement. Further attempts at better prenatal diagnosis, better perioperative care, and better longterm follow-ups are necessary to improve the survival and quality of life of affected infants. Next generation prospective studies that are done in bigger sample size and are conducted in more than one center are necessary so that they can be used to prove these findings and also to investigate long-term functional outcomes, nutritional status, and developmental milestones among survivors. The creation of a national registry of neonatal surgical conditions would also be useful in enhancing the early referral systems and uniformity of the management procedures among institutions. The importance of multi-centered cooperation is in regional differences and better results in neonatal surgical illnesses [12]. Conclusion Neonatal intestinal obstruction is an acute and urgent surgical condition, which involves high index of suspicion, early diagnosis, and early intervention to avoid severe complications that may occur due to bowel ischemia, sepsis, and short bowel syndrome. NIO has diverse etiology and in our study the most prevalent etiology is jejunoileal atresia, and then malrotation with volvulus and Hirschsprung disease. Most of the patients were presented during the initial days of their lives with typical symptoms of bilious vomiting and abdominal distension. Initial assessment involved plain abdominal X-ray which was universally applicable, whereas contrast studies were important in establishing the degree and nature of obstruction. All patients were subjected to hard surgical treatment based on the pathology. Regardless of the limitations that are linked to neonatal surgery especially when dealing with preterm or very sick infants, our outcomes indicate that the morbidity and mortality rates are tolerable in relation to global standards. Preterm babies and infants who came late with or with the manifestations of sepsis were found to more often have complications. The significance of this study is that it highlights the need to: • Enhancing prenatal diagnosis by regular ultrasound screening. • Early postnatal diagnosis by clinical suspicion and relevant imaging. • Surgical intervention to curb the development of bowel necrosis. • Postoperative care which is multidisciplinary in order to minimize complications and enhance the outcome. Prospective research is necessary in future to compare the long-term functional outcomes, nutritional status and developmental milestones in survivors. The creation of a national registry of neonatal surgical cases can also be used to enhance the early referral systems and standard management procedures across hospitals. References [1] Moore SW, Pierro A. Neonatal intestinal obstruction. Surg Clin North Am. 2007;87(1):157-73. doi:10.1016/j.suc.2007.01.005 [2] Stringer MD. Neonatal intestinal obstruction: a surgical perspective. Arch Dis Child Fetal Neonatal Ed. 2016;101(3):F263-9. doi:10.1136/archdischild-2015308861 [3] Elhalaby EA, Coran AG, Blane CE, et al. Intestinal atresia: changing trends over three decades. J Pediatr Surg. 1995;30(3):410-4. doi:10.1016/00223468(95)90596-3 [4] Coran AG, Teitelbaum DH, London WT, eds. Pediatric surgery. 8th ed. Philadelphia: Elsevier; 2020. [5] Ziegler MM, Azizkhan RG, van Allmen D, eds. Pediatric surgery. Cham: Springer; 2017. [6] Bax NM, Coran AG. Pediatric minimally invasive surgery. Cham: Springer; 2019. [7] Levitt MA, Peña A. Management of disorders of the anorectum and pelvic floor in children. Cham: Springer; 2019. [8] Gupta DK, Biswas B, Saha AK, et al. Neonatal intestinal obstruction in India: a decade's experience. J Indian Assoc Pediatr Surg. 2014;19(2):72-7. doi:10.4103/0971-9261.129608 [9] Danzer E, Johnson MP, Bebbington M, et al. Prenatal diagnosis of intestinal atresia improves postnatal outcomes. J Pediatr Surg. 2010;45(7):1439-43. doi:10.1016/j.jpedsurg.2010.03.018 [10] Al-Salem AH. Neonatal intestinal obstruction in a developing country: a 10-year review. Saudi J Gastroenterol. 2011;17(2):117-21. doi:10.4103/13193767.77239 [11] Wester T, Bjørnland K, Emblem R, et al. Long-term outcome after neonatal intestinal obstruction. Eur J
EUR J MED HEALTH RES Volume 3 | Number 4 | 2025 245 Pediatr Surg. 2008;18(5):289-95. doi:10.1055/s-20081038490 [12] Snyder CL, Langer JC. Neonatal intestinal obstruction. Pediatr Clin North Am. 2019;66(2):291-308. doi:10.1016/j.pcl.2018.12.004 [13] Cole GJ, et al. Neonatal intestinal obstruction: clinical profile and outcomes. J Pediatr Surg. 2006;41(12):3294-9. doi:10.1016/j.jpedsurg.2006.09.020