Almond, Sarah L.*; Haveliwala, Zeni†; Khalil, Basem*; Morabito, Antonino*
When the functional capacity of the intestine is inadequate for required nutrient absorption, the outcome is intestinal failure (IF) (1,2). The most common cause in the paediatric population is short bowel syndrome (SBS), following extensive intestinal resection. More rarely, IF occurs secondary to congenital enteropathies and motility disorders (3). The mainstay of therapy for IF is parenteral nutrition (PN), in combination with enteral nutrition to stimulate gut adaptation (3). Other strategies, used primarily in the SBS group, include surgery to lengthen the residual bowel, thus increasing transit time and maximising the absorptive capacity of the gut (4–6); however, despite these therapies, many children remain dependent on PN (7,8). The consequences of long-term PN are well described, and include failure to thrive, central venous catheter–related sepsis, loss of venous access for PN- and IF-associated liver disease (9,10). Ultimately, a proportion of these children require small bowel and/or liver transplantation (2).
The cutoff point for severe SBS is generally defined as 40 cm of small intestinal length. Children with >40 cm of bowel are expected to undergo bowel adaptation and eventually achieve nutritional autonomy (11,12); however, a subset of patients has IF despite “adequate” bowel length. The most common indications for paediatric bowel resection are neonatal pathologies, particularly gastroschisis, intestinal atresias, and necrotising enterocolitis (2,13). These conditions are associated with dysmotility of the residual gut (14,15). Dysmotility can lead to progressive bowel dilatation, with associated intraluminal stasis and bacterial overgrowth (3,5,14,16). Thus, enteral nutrition is poorly tolerated and adaptive processes fail.
This group of patients with IF and bowel dilatation despite adequate intestinal length for adaptation can be defined as having short gut–associated bowel dilatation (SGABD). For such patients, resection of the dilated bowel is often precluded by the length of the affected segment (15); however, SGABD may be amenable to intestinal reconstructive surgery, to reduce bowel diameter while preserving or enhancing intestinal length. We aimed to identify the proportion of patients referred to our unit with SGABD and investigate whether surgery to reduce dilatation could improve outcomes for these children.
Data were collected prospectively on all patients with IF referred to our intestinal rehabilitation unit for a 7-year period (2004–2011). Preliminary assessment of bowel length using previous operative notes enabled identification of patients with >40 cm small intestine. These patients were investigated by upper gastrointestinal contrast study, to further define the residual gut length and to establish the presence and degree of gut dilatation. Children with a maximal small intestinal diameter >2 vertebral bodies were considered to have dilated bowel. Patients with IF and bowel dilatation despite adequate bowel length for adaptation (SGABD) were thus identified.
Initial management of these patients followed our standard protocols, including prompt restoration of bowel continuity (6,17). Our standard low-fat “hepatosparing” PN regimen providing 1.5 to 2.5 g · kg−1 · day−1 of fat, 3.75 g · kg−1 · day−1 of protein, and 14 g · kg−1 · day−1 of glucose was used (6), in conjunction with scrupulous central line care, including daily alcohol line locks (18). Liver function was monitored biweekly and serum bilirubin levels >20 μmol/L considered abnormal. Strategies for managing deteriorating liver function included reducing the lipid content of PN, cycling PN (commenced routinely from 3 months of age), administration of ursodeoxycholic acid, and aggressive treatment of suspected line sepsis with intravenous antibiotics (9). Unresolving or recurrent (same causative organism) episodes of line sepsis were indications for line change.
All SGABD patients were subsequently considered for reconstructive surgery, aiming to restore normal gut motility and physiology. The principles of such surgery were 3-fold:
1. Confirm adequate bowel length for nutritional autonomy
2. Assess patency of anastomoses and refashion where necessary
3. Reduction of bowel diameter with preservation of intestinal length
All of the procedures were performed by a single surgeon. Surgical procedure was determined intraoperatively, taking into consideration the length and condition of the bowel and the underlying pathology.
Postoperatively, ongoing administration of PN was accompanied by attempts to establish feeds, with increasing enteral nutrition/gradual reduction of PN, and judicious use of prokinetic agents (domperidone and/or erythromycin, dependent on individual patient response). Our preferred option for infant feeding is breast milk and/or lactose-free hydrolysed formula (Pepti-Junior, Cow & Gate, Wiltshire, UK). Solid diet is introduced gradually after the age of 6 months. Whenever possible, feeds are given orally, to minimise the development of food aversion; however, feeding gastrostomies are placed at the time of surgery for those patients with severe food aversion, to facilitate weaning from PN. Notwithstanding these general guidelines, enteral feeding is tailored to the individual child with extensive input from specialist dieticians. All feeding interventions are assessed with respect to the child's growth, weight gain, and developmental milestones.
Outcome measures assessed both pre- and postoperatively included episodes of blood culture–confirmed sepsis, incidence of IF-associated liver disease, bowel length, time to achieve nutritional autonomy, and growth centile. Data were analysed using GraphPad Prism version 5 (GraphPad Software Inc, San Diego, CA) software and are expressed as mean (range), unless otherwise specified, using 2-tailed paired t test. P < 0.05 was significant.
A total of 26 patients with IF were referred to our intestinal rehabilitation unit between 2004 and 2011. All of the patients had undergone bowel resection (including resection of the ileocaecal valve and partial colectomy, Table 1) in the neonatal period and were referred with a diagnosis of IF presumed secondary to SBS. Of these, 8 (5 girls, 3 boys) fulfilled the criteria for SGABD (Tables 1 and 2). The most common underlying pathology in this cohort was necrotising enterocolitis (NEC, n = 3), followed by gastroschisis with atresia (n = 2). A single patient with simple (nonatretic) gastroschisis had IF following midgut volvulus within a silo. Three of the 8 patients had been assessed for transplant. One patient required initial surgery to anastomose the bowel; intestinal continuity had been restored before referral in all other cases. Previous attempts to establish enteral feeding over a minimum of 3 months had been unsuccessful, and all of the patients remained totally PN dependent.
All 8 patients with SGABD subsequently underwent intestinal reconstructive surgery to reduce the diameter of the bowel while preserving or enhancing gut length, with the aim of achieving nutritional autonomy. One patient documented as having 40 cm of small intestine was found to have only 35 cm when measured at the time of reconstructive surgery (Table 1). Median age at time of surgery was 273 days (103–1059). Three patients were managed by longitudinal intestinal lengthening and tailoring (LILT), 2 by serial transverse enteroplasty (STEP), and 3 underwent tapering enteroplasty (TE). Patients were followed up for a median of 51 months (18–84).
Postoperative outcomes are summarised in Table 3. No patient experienced a reduction in bowel length as a consequence of surgery. Overall, intestinal length increased significantly from 51 cm (35–75) to 73 cm (45–120); P = 0.02. Surgery led to a decline in the incidence of blood culture–positive sepsis from 2.8 (0–5) to 0.8 (0–2) episodes per patient; P = 0.01. This reduction in infection rate was also apparent when expressed relative to duration of PN (1.01 infections/100 preoperative PN days [0–2.15] vs 0.32 infections/100 postoperative PN days [0–1.23]; P = 0.02). In addition, peak serum bilirubin levels were significantly reduced postoperatively (peak preoperative 251 μmol/L [90–566] vs peak postoperative 58.5 μmol/L [10–204]; P = 0.005) with liver function ultimately normalised in all patients. From 100% PN requirements preoperatively, 7 patients have been completely weaned from PN, at a median of 110 days (35–537). Although 1 patient remains on PN 497 days after surgery, this is at a reduced requirement of 2 nights per week. All of the children are eating an unrestricted normal diet and are gaining weight appropriately. At the time of writing, there have been no significant surgical complications.
IF is a complex condition with a wide range of underlying causes. Present management focuses on nutritional support while allowing time for bowel adaptation to occur (2,3); however, this process can take many months and is fraught with complications (7,10,13). As a result, a significant number of patients eventually require transplantation (2). For patients with severe SBS (<40 cm small intestine), autologous intestinal reconstructive surgery to lengthen the remaining bowel can be a valuable adjunct to management (5,6); however, the role of surgery for patients with SBS with >40 cm small bowel is less well defined. Delineation of prognostic indicators may enable early identification of patients for whom surgical intervention may be beneficial, improving long-term outcomes.
We report a group of patients with IF associated with bowel dilatation, managed successfully using a variety of surgical techniques to reduce bowel diameter. All of the patients in our study had failed to establish enteral nutrition during a period of at least 3 months, because of clinical symptoms of dysmotility including recurrent vomiting, abdominal distension, and constipation. In addition, most had experienced episodes of sepsis and all had documented episodes of hyperbilirubinaemia. We hypothesised that reduction of bowel diameter would improve intestinal motility, reducing stasis and bacterial overgrowth and accelerating intestinal adaptation. Following surgery, episodes of sepsis and peak serum bilirubin measurements decreased. All of the patients presently have liver function within the normal range. In addition, 7 patients weaned from PN to a normal diet.
The type of reconstructive surgery used was at the discretion of the operating surgeon, dependent upon the following factors: initial pathology, total small intestinal length, length of the dilated segment, degree of dilatation, general condition of the bowel (adhesions, hypertrophy), and condition of the mesentery. Although all of the patients were considered to have sufficient bowel length to permit enteral autonomy, all had undergone previous resection of significant lengths of bowel including the ileocaecal valve. It was therefore imperative to avoid further intestinal resection. In all cases, the patency of previous anastomoses was established to exclude functional obstruction as a cause of dilatation. The reconstructive techniques used were LILT, TE, and STEP.
LILT was used in the management of 3 children with SGABD following management of antenatally diagnosed gastroschisis. The advantage of LILT in these cases, beyond reducing dilatation, was enhancement of intestinal length. TE is a well-described technique initially used in the management of high intestinal atresias (19). In our group, it was used for 3 patients with an initial diagnosis of necrotising enterocolitis. These patients had extensive adhesions and mesenteric thickening, unsuitable for LILT. Although TE preserves intestinal length, mucosal surface area is lost. In the context of patients with IF, this approach could be detrimental to future functional gut capacity. An alternative approach would have been STEP, but at the time, our experience of this procedure was limited. Subsequently, STEP was undertaken in 2 patients with previous intestinal atresias, enabling reduction of intestinal diameter with preservation of mucosal surface area. Although the increase in bowel length gained by STEP is reduced in comparison to LILT (4,20), it should be noted that this was not the primary aim of surgery.
It could be argued that given more time, similar outcomes would have been achieved without surgical intervention; however, avoidance of bacterial overgrowth and stimulation by luminal nutrients are prerequisites for gut adaptation (3,14,16). None of our patients was able to tolerate significant volumes of enteral nutrition before surgery. Furthermore, all of the patients had significant bowel dilatation, indicating dysmotility and associated stasis. We therefore suggest that reconstructive surgery to reduce the bowel diameter facilitates the intestinal adaptive process, thus reducing PN requirements and associated morbidity. Nonetheless, trial data would be required to fully elucidate these issues.
We recommend that all patients with IF presumed secondary to SBS should undergo careful evaluation by a multidisciplinary team (10). A contrast study to assess for bowel dilatation is an essential part of this process. In our series, patients with SGABD comprise a significant proportion (8/26, 30%) of referrals. Such patients are amenable to intestinal reconstructive surgery to reduce bowel diameter, enabling earlier enteral autonomy with minimal complications.
1. O’Keefe SJ, Buchman,F AL, Fishbein TM, et al. Short bowel syndrome and intestinal failure: consensus definitions and overview. Clin Gastroenterol Hepatol 2006; 4:6–10.
2. Bines JE. Intestinal failure: a new era in clinical management. J Gastroenterol Hepatol 2009; 24 (suppl 3):S86–S92.
3. Kocoshis SA. Medical management of pediatric intestinal failure. Semin Pediatr Surg 2010; 19:20–26.
4. Sudan D, Thompson J, Botha J, et al. Comparison of intestinal lengthening procedures for patients with short bowel syndrome. Ann Surg 2007; 246:593–601.
5. Jones BA, Hull MA, McGuire MM, et al. Autologous intestinal reconstruction surgery. Semin Pediatr Surg 2010; 19:59–67.
6. Khalil BA, Ba’ath ME, Aziz A, et al. Intestinal rehabilitation and bowel reconstructive surgery: improved outcomes in children with short bowel syndrome. J Pediatr Gastroenterol Nutr 2012; 54:505–509.
7. Sala D, Chomto S, Hill S. Long-term outcomes of short bowel syndrome requiring long-term/home intravenous nutrition compared in children with gastroschisis and those with volvulus. Transplant Proc 2010; 42:5–8.
8. Wiskin AE, Cole C, Owens DR, et al. Ten-year experience of home parenteral nutrition in a single centre. Acta Paediatr 2012; 101:1651–2227.
9. Cowles RA, Ventura KA, Martinez M, et al. Reversal of intestinal failure-associated liver disease in infants and children on parenteral nutrition: experience with 93 patients at a referral center for intestinal rehabilitation. J Pediatr Surg 2010; 45:84–87.
10. Hess RA, Welch KB, Brown PI, et al. Survival outcomes of pediatric intestinal failure patients: analysis of factors contributing to improved survival over the past two decades. J Surg Res 2011; 170:27–31.
11. Goulet O, Baglin-Gobet S, Talbotec C, et al. Outcome and long-term growth after extensive small bowel resection in the neonatal period: a survey of 87 children. Eur J Pediatr Surg 2005; 15:95–101.
12. Sigalet D, Boctor D, Brindle M, et al. Elements of successful intestinal rehabilitation. J Pediatr Surg 2011; 46:150–156.
13. Casey L, Lee KH, Rosychuk R, et al. 10-year review of pediatric intestinal failure: clinical factors associated with outcome. Nutr Clin Pract 2008; 23:436–442.
14. Gariepy CE, Mousa H. Clinical management of motility disorders in children. Semin Pediatr Surg 2009; 18:224–238.
15. Phillips JD, Raval MV, Redden C, et al. Gastroschisis, atresia, dysmotility: surgical treatment strategies for a distinct clinical entity. J Pediatr Surg 2008; 43:2208–2212.
16. Bianchi A, Morabito A. The dilated bowel: a liability and an asset. Semin Pediatr Surg 2009; 18:249–257.
17. Ba’ath ME, Almond S, King B, et al. Short bowel syndrome: a practical pathway leading to successful enteral autonomy. World J Surg 2012; 36:1044–1048.
18. Cober MP, Kovacevich DS, Teitelbaum DH. Ethanol-lock therapy for the prevention of central venous access device infections in pediatric patients with intestinal failure. JPEN J Parenter Enteral Nutr 2011; 35:67–73.
19. Adzick NS, Harrison MR, deLorimier AA. Tapering duodenoplasty for megaduodenum associated with duodenal atresia. J Pediatr Surg 1986; 21:311–312.
20. Kim HB, Fauza D, Garza J, et al. Serial transverse enteroplasty (STEP): a novel bowel lengthening procedure. J Pediatr Surg 2003; 38:425–429.