Short bowel syndrome (SBS) describes the status of malabsorption caused by resection of large parts of the small intestine resulting in a high mortality rate. During the first months of life, accurate management has been shown to be most important for the child's survival. After survival and growth are ensured through enteral and parenteral nutrition (PN), the next step for each patient is to achieve enteral autonomy (1). Of course, an adequate adaptation reaction of the gut is mandatory; this means growing in length in infants and increasing mucosal surface by dilatation of the bowel.
The aim of surgery in SBS is to delay transit time and to increase the contact time of enteral alimentation to the mucosa. In patients with dilated bowel, the aim is to increase propulsive activity and to avoid gastrointestinal stasis leading to vomiting, bacterial overgrowth, and recurrent sepsis.
Surgical attempts have been the reversed insertion of bowel segments or colon interpositions, constructions of intestinal valves, and tailoring against intestinal stasis (2–4). Other techniques are intended to increase bowel length. In 1980 the technique of longitudinal intestinal lengthening and tailoring (LILT) was described by Bianchi (5) and gained recognition as surgical therapy. By 2003, the serial transverse enteroplasty (STEP) had been established as a surgical option, and the first results have been published by the international STEP data registry (6–8).
In our institution we have performed LILT for more than 20 years. The aim of this report is to describe our series of 53 patients, the long-term follow-up, and factors for prediction of potential outcome.
PATIENTS AND METHODS
Between 1987 and 2006, 53 patients underwent intestinal LILT in our department. The indications for LILT were dependence on PN. A weaning period for at least 6 months with adequate conservative therapy was imposed after bowel continuity was reconstituted.
A gastrointestinal contrast study was performed in every case to evaluate the length and diameter of the remaining small bowel. Doppler imaging of the central vessels was performed to exclude potential thrombosis when indicated. Serial blood cultures and nasopharyngeal cultures were obtained to exclude highly resistant bacteria, especially because most patients were referred from other institutions and had spent most of their lives in a hospital.
Percutaneous liver biopsy and single coagulation factors have recently been included in the preoperative evaluation to exclude severe liver injury. Our recommendations to prevent severe liver injury before LILT is performed are summarized in the following list:
- Oral feeding is increased until stool frequency is as high as 10 bowel movements per day.
- Ursodeoxycholic acid is given orally in a concentration of 30 mg · kg−1 · day−1.
- Bacterial overgrowth is treated by a selective decontamination of the small bowel using paromomycin.
- A hepatosparing PN regimen is mandatory:
- Carbohydrate load is limited to a maximum of 12 g · kg−1 · day−1.
- Fat is restricted to 1 g · kg−1 · day−1.
- PN is cycled: in patients 6 months or older, an interruption of infusions 4 to 6 hours/day as a minimum is recommended.
The anatomic requirement for successful lengthening is dilatation of the small bowel to a diameter of at least double the normal size. The operative technique has been described elsewhere (9).
Fashioning of the new bowel loops was accomplished by inverted running sutures or stapler anastomoses. Anastomoses were performed in an isoperistaltic manner. Fibrin sealing covered long suture lines. Postoperative feeding with formula was started after return of bowel function. Solid food was introduced when formula was well tolerated by the patient.
The medical records of 53 patients were retrospectively reviewed and analyzed. Descriptive data including sex, weight and height, primary diagnosis leading to SBS, length of short bowel before and after the Bianchi procedure, existence of ileocecal valve, and age at time of Bianchi procedure were recorded.
Actual outcome data were evaluated, including survival, successful weaning from PN, weight, height, and parameters indicating quality of life. The data are presented as number of patients (%) or mean ± standard deviation (range).
Descriptive statistics (mean, median, standard deviation) were computed for all of the relevant parameters. Survival times were analyzed by use of the method of Kaplan and Meier, including the log rank test for group-specific differences with respect to ileocecal valve and PN. The relation between survival status and bowel length was investigated by the 2-sample t test and the U test; moreover, a simple logistic regression was performed to test the influence of bowel length on the probability of survival. The Fisher exact test was performed to evaluate the significance of the colon for survival of our patients. All tests were 2-sided. In general, a P value below 0.05 was considered to be significant. All of the calculations were done within the SAS system (version 8.2).
In all, 53 consecutive patients were included for evaluation in this study (27 boys, 26 girls). After intestinal lengthening, 41 patients are alive, and 29 are in our follow-up program and responded to our questionnaire. The follow-up time was 79.95 months (6–234 months). Survival of the remaining 12 patients was checked at the registration office. The diagnoses leading to SBS are given in Table 1.
The LILT was performed at an average age of 24 months (4−144 months). The mean small bowel length was 30.6 ± 15.1 cm before LILT and 54.71 ± 26.95 cm after LILT. The ileocecal valve was retained in 8 patients.
In our series the overall survival was 77.36%. Thirty-six patients were weaned successfully from PN, and 23 of 29 in our follow-up program stayed free of PN. The median weaning time was 10.4 months (1.5–60 months). Successful weaning was defined as 4 weeks of PN without weight loss.
After LILT, 12 of 53 patients died. One patient died perioperatively because of acute liver failure with intractable intestinal bleeding, and 3 other patients died early after surgery because of liver failure. These patients experienced severe liver insufficiency, and as a result of our learning curve such patients would no longer be accepted for a lengthening procedure.
All patients who did not survive died within the first 22 months after LILT. The causes leading to death are given in Table 2. None of the surviving patients underwent small bowel transplantation. One patient was referred to a transplantation center 7 years after undergoing LILT. Weaning from PN failed, and venous access was limited; the patient has not been included in the transplantation program because of multiple intestinal thromboses. The patient is still receiving PN and is in poor general condition; jaundice has developed. Survival time in patients was analyzed by use of the Kaplan-Meier function. The calculated curve with a long-term survival of 77.36% is shown in Figure 1.
Weight and Height
The weights and heights of the patients before the Bianchi procedure were registered and compared with the percentiles at the time of last follow-up. The median height before LILT was at the 17th percentile (1−97 percentile). Median weight was at the 10th percentile (1−41 percentile). After LILT, the weights of the patients remained below the 3rd in 24% of patients; however, 58% of the patients gained weight to a higher percentile. The individual percentiles of the weight are shown in Figure 2. Of interest, all of the patients who are still receiving PN show a significant lack of weight gain.
Quality of Life
According to our follow-up, 22 of 29 (75.86%) patients are completely free of central venous access. Sixteen patients need additional enteral iron or vitamin supplementation. None of the patients have required water and electrolyte substitution, even during hot summers or while experiencing general pediatric infectious diseases. The need for a special diet or a tendency to diarrhea is described in almost all of the patients. Salty food is preferred by 29 of 29 patients, and all of the patients have 5 to 8 meals per day.
Of the school-age patients, 75% attend a regular school, and only 25% are in a school for handicapped children. Two patients are already employed without restrictions. Retardation and reduced performance in school are due to former concomitant diseases such as intracerebral bleeding or perinatal asphyxia.
In the self-evaluation, 41% of the patients or parents describe normal physical strength, 49% report a mildly reduced physical constitution, 10% report a severely reduced physical constitution, and 44% participate in for sports once or twice weekly.
Table 3 shows SBS-related complications after LILT. Disturbances are recurrent episodes of abdominal pain or cramps and recurrent vomiting caused by transport problems or bacterial overgrowth that can be managed by special diet and recurrent oral antibiotic therapy.
Acidosis (D-lactate acidosis) tends to be exacerbated as a result of errors in the diet, summer heat, or infectious diseases. Generally, this condition can be managed by special diet, sufficient oral fluid intake, electrolyte substitution by tea enriched with electrolytes, and the oral administration of bicarbonate.
Owing to high stool frequency, perianal eczema remains a problem over a long period. There are no general recommendations other than intensive care for skin, including zinc-containing ointment.
One patient receiving PN experienced jaundice as a result of reduced liver function. This patient has been referred to a transplantation center, but intestinal transplantation is considered to be impossible because of his intestinal vascular situation.
The presence of the ileocecal valve has been often discussed to be responsible for the good outcome in patients with SBS. In our series, no significance or evidence was found for that postulation; the remaining ileocecal valve had no impact on the chance to be weaned from PN. A log rank test for group-specific differences indicated that the ileocecal valve was not significant (P = 0.6113).
Inasmuch as the colon is necessary for reabsorption of water and electrolytes, the impact of colon length was investigated. The patients were divided into groups: 1 group with a complete colon or only a little resection and 1 group with half or less than half of the colon remaining. The data from 48 patients were evaluated and showed that 38 patients had a complete colon or only a little resection on the right hemicolon and10 patients had only parts of the left colon. The Fisher exact test showed a significant better survival rate in patients with a complete colon or only a small resection on the right-sided colon (P = 0.039).
It is obvious that in patients with SBS, survival and nutrition are dependent on bowel length. In our statistical analysis with logistic regression, we found a factor of 0.6582 + 0.0131 * cm bowel as a factor for surviving. In the group of patients who required LILT, a 2-sample t test and the U test were performed. There was no statistically significantly better outcome in surgical patients with longer bowel compared with patients with shorter bowel. Apart from extremely short segments, this fact can be interpreted as follows: a child does not have a statistically better chance whether it has a 30-cm or 40-cm bowel length, probably because other factors such as length of colon, liver status, and possibly underlying or concomitant diseases have a higher influence on the survival of these patients than length of residual gut alone.
Dependence on Parenteral Supplementation
Of our patients, 75.86% are without permanent central venous access and have been weaned from PN. We compared the survival of patients who still require PN with that of patients who have been totally weaned from PN 18 months after LILT. A log rank test showed that patients without central venous access have a statistically highly significantly better survival rate than do patients still receiving PN after 18 months (P = 0.0032).
In population-based studies, the incidence of SBS has been 24.5 (12.1–36.9) per 100,000 live births, with a much higher incidence in babies born before 37 weeks of gestation compared with term newborns (10). Actual cohort studies show a high mortality of SBS (6–45%) (11). Supported by modern PN at home, surgical techniques for autologous gastrointestinal reconstruction such as LILT and STEP or small bowel transplantation were established.
The techniques of autologous gastrointestinal reconstruction have been shown to augment intestinal adaptation, to improve intestinal motility, and to reduce bacterial overgrowth. (12,13).
Early reports of the LILT showed a high complication rate with anastomotic leakage, abscess formation, and fistulization, but with more experience the complication rate has been reduced in other series (1,14,15). The perioperative mortality is low, and only 1 patient in our series died because of intractable bleeding in a later stage of liver insufficiency. We also had 3 more patients who died because of liver insufficiency within a short period after gastrointestinal reconstruction. Like other authors, we therefore postulate the exclusion of patients with irreversible liver damage before LILT (1,16). This can be underlined by the fact that patients with liver cirrhosis have a high mortality and morbidity in elective nonhepatic surgery. Therefore, in adult patients different scores have been established to evaluate the perioperative risk for these patients (17,18). Unfortunately, that has not yet been proven or modified for pediatric patients.
We intended to show whether there are prognostic factors for successful LILT, especially because intestinal transplantation has been postulated as the therapy of first choice in patients with intestinal length <50 cm (16).
We present the largest series of patients undergoing LILT so far, to our knowledge. The median follow-up was 79.9 months, and the bowel length was 30.6 cm before LILT. According to the Kaplan-Meier function, the survival in our series is 77.36%. This is higher than has heretofore been reported in infants and children with SBS and PN and in patients with SBS after intestinal transplantation, respectively (19,20). Besides the survival, we have shown that after LILT, 75% of patients can participate in normal life, with a regular education in school and a good physical constitution.
Weight and height increased in 58% of patients, and only one fourth remained under the 3% percentile. Of interest, the patients who could not be weaned from PN within 18 months after LILT are the patients who do not gain sufficient weight, as is shown in Figure 2.
Identifying those patients who will profit from LILT before the operation is performed still remains a problem. In our statistical analysis, the existence of the ileocecal valve did not influence the prognosis. By contrast, the length of the colon, especially the remaining right hemicolon, had a statistically significant impact on the survival of our patients. Absolute bowel length is important, of course, but within the group of patients who required LILT, the preoperative bowel length had no statistically significant influence. That is in strong contrast to other workers, who postulate intestinal transplantation in patients with bowel length <50 cm (16).
A further preoperative prognostic parameter remains the severity of liver damage. Inasmuch as we observed a high postoperative mortality resulting from liver insufficiency and intractable diffuse bleeding during our learning curve, severe liver injury became an exclusion criterion for LILT. Regarding liver function, we had the impression that minor liver damage improves after LILT. In our follow-up time, liver enzymes and bilirubin concentration improved after successful weaning from PN, but statistical analysis has not yet been performed.
The only significant prognostic parameter after LILT was persistent dependence on PN. Inasmuch as mean weaning time from PN was 10.4 months and only some outliers needed more than 18 months, we defined 18 months as the upper limit for successful weaning. All of the patients who exceeded 18 months of PN showed a significantly worse prognosis for survival similar to that in patients who have never had LILT (10).
With this favorable outcome in survival and quality of life, LILT remains the surgical treatment of first choice in patients with SBS who meet the necessary criteria for autologous intestinal reconstruction. It will be interesting to compare LILT with the newer STEP technique showing promising results in the early postoperative period (7,21,22). In our institution we prefer LILT as the technique of first choice, and we see the major indication for the STEP procedure as a second step (23) in patients who again show dilatation of the small intestine after LILT.
Beside autologous intestinal reconstruction, transplantation of the small bowel with or without liver has been shown to be clinically successful with increasing time of patient survival. Still, the 5-year graft survival is <50%, and there is a general lack of donor organs (19). Rejection remains a persistent threat for the recipient, with the need for lifelong immunosuppressive therapy. Furthermore, there is the permanent risk for infectious diseases and secondary malignancies after intestinal transplantation, such as myeloproliferative syndrome or de novo malignancies (24).
According to our results, transplantation should be considered in patients with severe liver damage and in patients in whom central venous access and PN cannot be discontinued after autologous gastrointestinal reconstruction. This recommendation is based on the strong correlation between persistent dependence on PN and reduced survival of these patients in our series.
1. Bianchi A. From the cradle to enteral autonomy: the role of autologous gastrointestinal reconstruction. Gastroenterology 2006; 130:138–146.
2. Mackby MJ, Richards V, Gilfillan RS, et al
. Methods of increasing the efficiency of residual small bowel segments: a preliminary study. Am J Surg 1965; 109:32–38.
3. Hammer JM, Seay PH, Johnston RL, et al
. The effect of antiperistaltic bowel segments on intestinal emptying time. Arch Surg 1959; 79:537–541.
4. Budding J, Smith CC. Role of recirculating loops in the management of massive resection of the small intestine. Surg Gyncol Obstet 1967; 125:243.
5. Bianchi A. Intestinal loop lengthening: a technique for increasing small intestinal length. J Pediatr Surg 1980; 15:145–151.
6. Kim H, Fauza D, Garza J, et al
. Serial transverse enteroplasty (STEP): a novel bowel lengthening procedure. J Pediatr Surg 2003; 38:425–429.
7. Modi BP, Javid PJ, Jaksic T, et al
. International STEP Data Registry. First report of the international serial transverse enteroplasty data registry: indications, efficacy, and complications. J Am Coll Surg 2007; 204:365–371.
8. Dutta S. The STEP procedure: defining its role in the management of pediatric short bowel syndrome
. J Pediatr Gastroenterol Nutr 2007; 45:174–175.
9. Hosie S, Loff S, Wirth H, et al
. Experience of 49 longitudinal intestinal lengthening
procedures for short bowel syndrome
. Eur J Pediatr Surg 2006; 16:171–175.
10. Wales PW, de Silva N, Kim J, et al
. Neonatal short bowel syndrome
: population-based estimates of incidence and mortality rates. J Pediatr Surg 2004; 39:690–695.
11. Wales PW, de Silva N, Kim JH, et al
. Neonatal short bowel syndrome
: a cohort study. J Pediatr Surg 2005; 40:755–762.
12. Weber TR, Powell MA. Early improvement in intestinal function after isoperistaltic bowel lengthening. J Pediatr Surg 1996; 31:61–64.
13. Chang RW, Javid PJ, Oh JT, et al
. Serial transverse enteroplasty enhances intestinal function in a model of short bowel syndrome
. Ann Surg 2006; 243:223–228.
14. Weber TR. Isoperistaltic bowel lengthening for short bowel syndrome
in children. Am J Surg 1999; 178:600–604.
15. Thompson JS, Pinch LW, Young R, et al
. Long-term outcome of intestinal lengthening
. Transplant Proc 2000; 32:1242–1243.
16. Bueno J, Guiterrez J, Mazariegos GV, et al
. Analysis of patients with longitudinal intestinal lengthening
procedure referred for intestinal transplantation. J Pediatr Surg 2001; 36:178–183.
17. Farnsworth N, Fagan SP, Berger DH, et al
. Child-Turcotte-Pugh versus MELD score as a predictor of outcome after elective and emergent surgery in cirrhotic patients. Am J Surg 2004; 188:580–583.
18. Keegan MT, Plevak DJ. Preoperative assessment of the patient with liver disease. Am J Gastroenterol 2005; 100:2116–2127.
19. Ruiz P, Kato T, Tzakis A. Current status of transplantation of the small intestine. Transplantation 2007; 83:1–6.
20. VanGossum A, Bakker H, Bozzetti M, et al
. Home parenteral nutrition in adults: a European multicentre survey in 1997. Clin Nutr 1999; 18:135.
21. Cowles RA, Lobritto SJ, Stylanios S, et al
. Serial transverse enteroplasty in a newborn patient. J Pediatr Gastroenterol Nutr 2007; 45:257–260.
22. Sudan D, Thompson J, Botha J, et al
. Comparison of intestinal lengthening
procedures for patients with short bowel syndrome
. Ann Surg 2007; 246:593–604.
23. Ehrlich PF, Mychaliska GB, Teitelbaum DH. The 2 STEP: an approach to repeating a serial transverse enteroplasty. J Pediatr Surg 2007; 42:819–822.
24. Abu-Elmagd KM, Zak M, Stamos JM, et al
. De novo malignancies after intestinal and multivisceral transplantation. Transplantation 2004; 77:1719–1725.