Author Affiliations: Department of Pediatric General Surgery (Chen QJ, Gao ZG, Tou JF, Qian YZ, Li MJ, Xiong QX), and Department of Pediatric Thoracic Surgery, Children's Hospital, Zhejiang University Shool of Medicine, Hangzhou 310003, China (Shu Q)

 

Corresponding Author: Qiang Shu, Department of Pediatric Thoracic Surgery, Children's Hospital, Zhejiang University Shool of Medicine, 57 Zhugan Xiang, Hangzhou 310003, China (Tel: +86-571-87033296; Fax: +86-571-87033296; Email: shuqiang@zju.edu.cn)

 

doi: 10.1007/s12519-014-0499-4

 

 

 

 

 

 

 

 

Congenital duodenal obstruction (CDO) is one of the most common anomalies in newborns, affecting approximately one in 2500 to 10 000 live births, and accounting for nearly half of all cases of neonatal intestinal obstruction.^[1] Several intrinsic and extrinsic congenital lesions can lead to complete or partial obstruction of the duodenum, including embryologic defects in the development of the foregut, disturbances in recanalization or rotation, and abnormal embryologic relationships with neighboring anatomical structures.^[2,3] Intrinsic duodenal obstruction can be caused by duodenal atresia, stenosis, or diaphragm, whereas extrinsic duodenal obstruction can be caused by annular pancreas, malrotation, or preduodenal portal vein.^[4] Advances in surgical management, intensive care medicine, and postoperative nutritional support over the last decades have contributed greatly to improved survival of newborns with CDO. However, CDO still carries significant morbidity and mortality because of low birth weights, associated complex anomalies, and postoperative complications.^[2,5] It is imperative to make a timely diagnosis and adopt an aggressive therapeutic strategy to maximize the outcome.

We reviewed our experience in managing CDO in 287 neonates over 10 years, with a focus on characterizing the clinical, radiologic, and prognostic findings.

 

 

Methods

We conducted a comprehensive retrospective review of the clinical data of all infants with a diagnosis of CDO admitted to Children's Hospital of Zhejiang University Medical School between 2003 and 2012. Patients with abdominal wall defects (omphalocoele, gastroschisis) and diaphragmatic hernias accompanied by malrotation were excluded from the study. We recorded the history of the mother's pregnancy and the neonates' sex, birth weight, clinical manifestations, associated anomalies, details of preoperative examination, intraoperative findings, type of operative intervention, length of postoperative hospitalization, short- and midterm complications, and final outcomes. The follow-up period ranged from 6 months to 5 years. The project was approved by the Ethics Committee of the Hospital.

All data were collected, tabulated and compared, and numerical data were presented as mean±SD. Statistical analysis was completed using the Statistical Program for Social Sciences (SPSS, version 13.0, Chicago, IL) to perform analysis of variance (ANOVA) or the chi-square tests; P<0.05 was considered to be statistically significant.

 

 

Results

Between January 2003 and December 2012, a total of 287 neonates (193 boys and 94 girls) with CDO were treated at our hospital. Patients' age ranged from 1 to 28 days (mean, 8.42±6.54); mean gestational age was 37.84±2.62 weeks (range, 28-41), average birth weight 3048.33±625.40 g (range, 950-4850). Fifty-three patients were born prematurely between 28 and 36 weeks' gestation. Prenatal ultrasound scans revealed a fetal double-bubble signal in 67 patients (Fig. 1A). Neonates diagnosed prenatally were admitted to the hospital and treated much earlier than neonates who were diagnosed postnatally (mean, 3.15±2.11 days vs. 7.03±5.71 days, 6.08±4.49 days vs 11.36±6.58 days, respectively), and the differences were significant (F=18.304, P<0.001; F=33.6, P<0.001). Among the entire cohort, 174 (60.63%) patients were diagnosed with malrotation, 55 (19.16%) had duodenal web (type I duodenal atresia), 66 (23.00%) had annular pancreas, 7 (2.44%) had type III duodenal atresia, 2 (0.70%) were diagnosed with duodenal stenosis, 2 (0.70%) had preduodenal portal vein, and 1 (0.35%) had congenital band compression. In addition, 20 patients (6.97%) had various combinations of lesions, mainly malrotation in combination with duodenal web, annular pancreas, preduodenal portal vein, or duodenal atresia (Table 1).

Vomiting was the first presenting symptom in 86.06% of the patients (247/287), including bilious vomiting in 207 (83.81%). Other symptoms and signs included epigastric distention in 47 (16.38%), hematochezia in 25 (8.71%), dehydration and acid-base disorder in 85 (29.62%), azotemia in 17 (5.57%), weight loss (>20% birth weight) in 50 (17.42%), hyperbililubinemia (total bilirubin >255 μmol/L) in 88 (30.66%), and aspiration pneumonia in 9 (3.14%). Symptoms appeared within the first 24 hours of birth in 42.51% of the patients and by the third day in 64.46%. Patients with annular pancreas or duodenal stenosis or atresia (including duodenal web) exhibited symptoms earlier than patients with malrotation (mean 1.64±0.21 days vs. 3.98±0.54 days and 1.31±0.17 days vs. 3.98±0.54 days, respectively).

X-ray examination and/or ultrasonic evaluation were obtained to confirm the diagnosis. Of the 209 plain film abdominal radiographs, 143 (68.42%) demonstrated a typical double-bubble sign, single-bubble sign, or air-fluid levels in the upper abdomen (Fig. 2A, B). Other nonspecific findings included signs of a dilated stomach, less intestinal gas, and uneven intestinal gas distribution (Fig. 2C, D). Compared with malrotation, there were more positive findings on upright abdominal radiographs for annular pancreas and duodenal atresia/stenosis (87.80 % vs. 48.15%, 89.80% vs. 48.15%, respectively), and significant differences were observed (X²=19.33, 24.61, respectively, P<0.001). Upper and/or lower gastrointestinal contrast studies were conducted in 64.11% patients to confirm the diagnosis. Abnormal findings included positioning of the duodenojejunal flexure to the right of the spine, a "coil spring" appearance of the proximal jejunum, and abnormal positioning of the ileocecal junction in patients with malrotation (Fig. 3A-C). Obstruction of the second or third part of the duodenum was present in patients with annular pancreas or duodenal atresia/stenosis (Fig. 3D-F). Among the 174 cases of malrotation, 113 had color Doppler ultrasound imaging, which revealed 81 cases of dilated duodenum with inversion of the superior mesenteric artery and vein (the whirlpool sign) (Fig. 1B, C).

Half of the patients (145, 50.52%) had one or more associated anomalies. The majority of these were cardiac defects (46, 31.72%), of which atrial septal defect was the most common. Forty-three neonates had associated gastrointestinal malformation, and 22 patients had urogenital system abnormalities. Defects of the neural, orthopedic, vertebral, and endocrine systems were also detected. Nine patients had Down's syndrome.

Surgical repair depended on the type of anomaly. Ladd's procedure with and without intestinal resection was performed in cases of malrotation; duodeno-duodenoduodenostomy or duodenoplasty was performed for annular pancreas (diamond-shaped duodenoplasty) and duodenal atresia; and web resection and duodenoplasty were performed for duodenal web. Laparoscopic surgery was performed in 21 cases (including 10 cases of malrotation, 7 cases of annular pancreas, and 4 cases of duodenal web). Seven patients with megaduodenum (≥5cm in diameter) underwent triangle-tapered duodenoplasty. Combination procedures were performed in patients with multiple gastrointestinal anomalies.

Postoperative parenteral nutritional support was achieved over a period of 9.32±3.45 days. Patients started enteral feeding after an average of 8.51±3.17 days, and mean postoperative hospital stay was 14.87±5.60 days. Postopative hospital stay was significantly longer for annular pancreas compared with malrotation (mean, 15.02±4.67 days vs. 12.76±4.07 days; F=34.4, P<0.001), and for duodenal atresia compared with malrotation (17.38±6.22 days vs. 12.76±4.07 days, F=44.4, P<0.001).

Seventeen patients died postoperatively. The cause of death included prematurity, sepsis, pneumonia, brain hemorrhage, extensive bowel gangrene, short bowel syndrome, and other serious anomalies. Eight patients died of serious life-threatening abnormalities or of premature and its sequelae, such as intracranial hemorrhage and respiratory distress syndrome; 8 patients died mainly as a result of volvulus and bowel necrosis, peritonitis and sepsis, aspiration pneumonia, disturbances of electrolyte and acid-base balance, and the complications resulting from delayed diagnosis. One patient died of anastomotic leak and peritonitis (Table 2). Surviving patients were followed up from 6 months to 5 years. Near- and mid-term complications were identified in 23 patients, and 14 patients required re-operation for wound dehiscence, adhesive intestinal obstruction, anastomotic leak, cholestasis, missed diagnosis of duodenal web, or recurrence of malrotation (Table 3).

Discussion

CDO is a relatively common abnormality in newborns admitted to the neonatal surgical unit, affecting boys more frequently than girls. Failure to identify duodenal obstruction in the immediate newborn period may result in clinically significant metabolic disturbances, aspiration pneumonia, growth retardation, and even short bowel syndrome in cases of midgut volvulus. Delayed diagnosis is a significant cause of increased morbidity and postoperative mortality.^[6,7] Although recent advances in surgical technology, neonatal intensive care unit medicine, respiratory support, and nutritional therapy have dramatically improved CDO outcomes, the relatively high incidence of postoperative re-operation and death still remains a major challenge for pediatric surgeons.^[8]

CDO can be detected as early as twenty weeks of gestation, and the obvious benefits of prenatal diagnosis have been wildly recognized.^[9] Antenatal diagnosis allows for parent counseling and pre-delivery planning, and enables additional to determine whether associated pathology is present. Adequate preoperative preparation and earlier surgery may help decrease overall morbidity, hospital length of stay, and costs.^[10-12] In our group, 67 cases were diagnosed prenatally. The majority of these had serious obstructions, i.e., duodenal web/atresia, annular pancreas, or malrotation associated with intestinal atresia. As compared with those patients diagnosed postnatally, patients diagnosed prenatally tended to to have a decreased gestational age and lower birth weight at delivery, and a higher rate of associated serious congenital anomalies. Despite earlier diagnosis confirmed and definite operation performed, there were no significant differences in total outcomes, and the post-operative hospital stay seems longer. These results were similar to those reported by Cohen-Overbeek's.^[13] We reasoned that a higher morbidity and mortality would be observed if the prenatal diagnosis was not achieved.

The postoperative re-operation rate for CDO is reported to be between 4% and 14%.^[2,8] The main reasons for re-operation are overlooked combination lesions, anastomotic leak or dysfunction, adhesive obstruction, wound dehiscence, and gastroesophageal reflux disease.

Congenital duodenal obstruction is the result of several embryologic defects in foregut development, canalization, or rotation. Multiple factors likely contribute to this pathogenesis, and it is widely recognized that congenital duodenal obstruction has a high incidence of concomitant anatomic abnormalities. Despite this knowledge, missed lesions during the initial surgery are still the major reason for re-operation, especially in cases of malrotation and duodenal web or multiple weds.^[2,8,14] Both duodenal web and malrotation cause incomplete duodenal obstruction and cannot usually be identified using preoperative gastrointestinal contrast. This kind of combination lesion is liable to be missed when performing Ladd's procedure without discovering the web or only removed proximal web but failed to find other webs. Obviously, the number of re-operations could be decreased with more thorough preoperative and intraoperative evaluation of the gastrointestinal tract and more careful attention to operative details. In our experience, we routinely irrigated the distal bowel with warm saline and pulled the nasogastric tube deeply into proximal jejunum during surgery to rule out distal obstruction, and we dissected suspicious sites when necessary. Our reoperative rate decreased dramatically in recent years after we fully realized this phenomenon and strengthened intraoperative exploration.

Anastomotic leak or dysfunction is another major cause of re-operation. In many CDO cases, the proximal duodenum is massively dilated and hypertrophied and demonstrates leather-like changes, and is always associated with secondary degeneration of ganglion cells and smooth muscle, features similar to those seen in cases of megaduodenum.^[15] The dilated proximal duodenum not only affects anastomotic healing, but it also results in ineffective coaptation of the dilated duodenum with a consequent inadequate aboral pressure gradient and functional obstruction.^[16] Delayed transit is one of the most problematic issues following duodenal obstruction repair, and is usually associated with a persistently dilated and dyskinetic proximal duodenum. The chronically obstructed duodenum cannot produce effective peristaltic waves, which may require prolonged total parenteral nutrition or surgical revision for persistent abdominal pain and failure to thrive. Previous studies^[15,16] have demonstrated that tapering or contouring the duodenum may ameliorate these problems. In recent years, we have routinely performed a tapering duodenoplasty when megaduodenum (≥5 cm diameter) is present. After trimming the length of the duodenum laterally to avoid damage to the main blood supply of the duodenum and the major duodenal papilla, we used full-thickness, interrupted sutures to repair the duodenal wall and create an anastomosis. There were no duodenal leaks or other complications, and postoperative feedings were initiated within 7 days for all patients who had tapering duodenoplasty. Pathological analysis of the dilated proximal duodenal tissue identified vacuolar degeneration of the enteric neuron.

The rapid expansion of laparoscopic surgery in the pediatric population has led to decreased morbidity and a shorter hospital stay for a number of procedures. Most duodenal obstruction operations that were classically performed via open laparotomy can be done endoscopically.^[17,18] There have been several reports indicating that laparoscopic Ladd's procedure and duodenoduodenostomy can be performed with success rates equal to those achieved with laparotomy, without increasing operative time.^[19-22] In the current series, we used laparoscopy in 21 patients in the last 3 years. One case required conversion to open laparotomy and 2 cases required re-operation due to postoperative volvulus and missed duodenal web. Although laparoscopic management of duodenal obstruction has proven to be safe and effective, it does have some evident disadvantages. The small size of the neonatal abdominal cavity affects the manipulation of laparoscopic instruments and results in prolonged operation time for beginners. It is difficult to widen the mesenteric base adequately and thoroughly assess the distal bowel for other atretic segments, resulting in a relatively high rate of conversion and re-operation.^[23-25] We propose that laparoscopy is preferable for treating duodenal obstruction in select patients, providing that a low threshold for conversion to open surgery is maintained.

The mortality rate of duodenal obstruction has gradually decreased since Ladd published his classic article more than 70 years ago, and the survival rate now exceeds 90%.^[26,27] The risk factors for death are complex cardiac anomalies, prematurity, sepsis, pneumonia, and surgical complications such as short bowel syndrome, anastomotic leak, and gastro-duodenal dysfunction.^[2,27,28] Multiple congenital anomalies and sepsis are still the main risk factors for mortality, accounting for nearly 80% of all postoperative deaths. In the present study, nearly half of the 17 patients who died postoperatively suffered from intestinal necrosis, aspiration pneumonia, or severe electrolyte and acid-base imbalance when they were admitted to our hospital. These situations could be avoided or minimized if the parents paid more attention to neonatal vomiting and if the primary care physician fully recognized this disease. In our opinion, although multiple congenital anomalies and prematurity are the major factors influencing the prognosis for duodenal obstruction, which is an established fact that can not be altered artificially. Beyond this prerequisite, we should put more emphasis on the importance of early diagnosis and urgent surgical treatment. It is key to improving prognosis and reducing postoperative complications of CDO.

 

 

Funding: This study was supported by grants from Zhejiang Province Science and Technology Programs (no. 2012C13G2010119, 2012C13024-1 and 2012C23096) and Zhejiang Provincial Program for the Cultivation of High-level Innovative Health talents (Qiang Shu).

Ethical approval: This study was approved by Ethical Committee of Chilren's Hospital, Zhejiang University School of Medicine.

Competing interest: None declared.

Contributors: Chen QJ wrote the first draft of this paper. All authors contributed to the intellectual content and approved the final version. Shu Q is the guarantor.

 

 

References

1   Lawrence MJ, Ford WD, Furness ME, Hayward T, Wilson T. Congenital duodenal obstruction: early antenatal ultrasound diagnosis. Pediatr Surg Int 2000;16:342-345.

2   Bailey PV, Tracy TF Jr, Connors RH, Mooney DP, Lewis JE, Weber TR. Congenital duodenal obstruction: a 32-year review. J Pediatr Surg 1993;28:92-95.

3   Mustafawi AR, Hassan ME. Congenital duodenal obstruction in children: a decade's experience. Eur J Pediatr Surg 2008;18:93-97.

4   Ladd WE. Congenital duodenal obstruction. Surgery 1937;1:878-885.

5   Kaddah SN, Bahaa-Aldin KHK, Aly HF, Hassan HS. Congenital duodenal obstruction. Ann Pediatr Surg 2006;2:130-135.

6   Bittencourt DG, Barini R, Marba S, Sbragia L. Congenital duodenal obstruction: does prenatal diagnosis improve the outcome? Pediatr Surg Int 2004;20:582-585.

7   Kilbride H, Castor C, Andrews W. Congenital duodenal obstruction: timing of diagnosis during the newborn period. J Perinatol 2010;30:197-200.

8   Zhang Q, Chen Y, Hou D, Guo W. Analysis of postoperative reoperation for congenital duodenal obstruction. Asian J Surg 2005;28:38-40.

9   Grosfeld JL, Rescorla FJ. Duodenal atresia and stenosis: reassessment of treatment and outcome based on antenatal diagnosis, pathologic variance, and long term follow-up. World J Surg 1993;17:301-309.

10 Romero R, Ghidini A, Costigan K, Touloukian R, Hobbins JC. Perinatal diagnosis of duodenal atresia: does it make any difference? Obstct Gynecol 1988;71:739-741.

11 Phelps S, Fisher R, Partington A, Dykes E. Prenatal ultrasound diagnosis of gastrointestinal malformations. J Pediatr Surg 1997;32:438-440.

12 Lawrence MJ, Ford WD, Furness ME, Hayward T, Wilson T. Congenital duodenal obstruction: early antenatal ultrasound diagnosis. Pediatr Surg Int 2000;16:342-345.

13 Cohen-Overbeek TE, Grijseels EW, Niemeijer ND, Hop WC, Wladimiroff JW, Tibboel D. Isolated or non-isolated duodenal obstruction: perinatal outcome following prenatal or postnatal diagnosis. Ultrasound Obstet Gynecol 2008;32:784-792.

14 Spigland N, Yazbeck S. Complications associated with surgical treatment of congenital intrinsic duodenal obstruction. J Pediatr Surg 1990;25:1127-1130.

15 Zhang XW, Abudoureyimu A, Zhang TC, Zhao JR, Fu LB, Lin F, et al. Tapering duodenoplasty and gastrojejunostomy in the management of idiopathic megaduodenum in children. J Pediatr Surg 2012;47:1038-1042.

16 Alexander F, Difiore J, Stallion A. Triangular tapered duodenoplasty for the treatment of congenital duodenal obstruction. J Pediatr Surg 2002;37:862-864.

17 Frantzides CT, Madan AK, Gupta PK, Smith C, Keshavarzian A. Laparoscopic repair of congenital duodenal obstruction. J Laparoendosc Adv Surg Tech A 2006;16:48-50.

18 Hill S, Koontz CS, Langness SM, Wulkan ML. Laparoscopic versus open repair of congenital duodenal obstruction in infants. J Laparoendosc Adv Surg Tech A 2011;21:961-963.

19 Rothenberg SS. Laparoscopic duodenoduodenostomy for duodenal obstruction in infants and children. J Pediatr Surg 2002;3:1088-1089.

20 Spilde TL, St Peter SD, Keckler SJ, Holcomb GW 3rd, Snyder CL, Ostlie DJ. Open vs laparoscopic repair of congenital duodenal obstructions: a concurrent series. J Pediatr Surg 2008;43:1002-1005.

21 Kay S, Yoder S, Rothenberg S. Laparoscopic duodenoduod-enostomy in the neonate. J Pediatr Surg 2009;44:906-908.

22 Li B, Chen WB, Zhou WY. Laparoscopic methods in the treatment of congenital duodenal obstruction for neonates. J Laparoendosc Adv Surg Tech A 2013;23:881-884.

23 Jensen AR, Short SS, Anselmo DM, Torres MB, Frykman PK, Shin CE, et al. Laparoscopic versus open treatment of congenital duodenal obstruction: multicenter short-term outcomes analysis. J Laparoendosc Adv Surg Tech A 2013;23:876-880.

24 Burgmeier C, Schier F. The role of laparoscopy in the treatment of duodenal obstruction in term and preterm infants. Pediatr Surg Int 2012;28:997-1000.

25 Hagendoorn J, Vieira-Travassos D, van der Zee D. Laparoscopic treatment of intestinal malrotation in neonates and infants: retrospective study. Surg Endosc 2011;25:217-220.

26 Mooney D, Lewis JE, Connors RH, Weber TR. Newborn duodenal atresia: an improving outlook. Am J Surg 1987;153:347-349.

27 Escobar MA, Ladd AP, Grosfeld JL, West KW, Rescorla FJ, Scherer LR 3rd, et al. Duodenal atresia and stenosis: long-term follow-up over 30 years. J Pediatr Surg 2004;39:867-871.

28 Niramis R, Anuntkosol M, Tongsin A, Mahatharadol V. Influence of Down's syndrome on management and outcome of patients with congenital intrinsic duodenal obstruction. J Pediatr Surg 2010;45:1467-1472.