RATIONALE FOR A NETWORK MODEL FOR MANAGEMENT OF INTESTINAL FAILURE
Intestinal failure (IF) is the result of long term or irreversible loss of intestinal function either because the intestine has been largely removed or its motility or digestive absorptive function is severely compromised (1,2). IF is a complex condition requiring a multidisciplinary approach. Management is made difficult by the low number of cases, heterogeneity of clinical presentation and need for advanced clinical, nutritional and surgical expertise (3,4). The major challenges in managing a child with permanent IF are the diagnostic approach, particularly in children with intractable diarrhea, and the nutritional management (5).
An increasingly broader spectrum of primary intestinal diseases can lead to permanent IF, and often advanced techniques are required to detect the primary etiology. Nonetheless, substantial progress has been made in parenteral nutrition (PN), and this has contributed to an increase in the number of surviving patients and to improved quality and quantity of life(1,6-9). Intestinal transplantation has also substantially improved from an experimental approach to a feasible option for IF (10-13).
The increasing number of children living with permanent IF is associated with enormous costs and the need to rethink health care models to optimize management, both in the short term and in the long term. The report of the working group of the Federation of International Societies of Pediatric Gastroenterology and Nutrition (FISPGHAN) indicated the major needs of medical intervention, research and education on the field of IF on a global scale (14). The report strongly calls for close international cooperation to reach the indicated goals (14).
In several countries, reference centers with advanced expertise in gastroenterology and PN evolved spontaneously, developing specific competencies in gastrointestinal diagnosis, clinical nutrition, gastrointestinal surgery and, finally, in intestinal transplantation. This happened without any specific planning in most western countries. In Italy, there is a relatively high number of centers with advanced competencies in pediatric gastroenterology and nutrition, and small numbers of IF patients are managed by individual centers (15). Specific knowledge and diagnostic techniques that are essential to manage a child with IF are scattered all over the country. To improve the management of children with IF and to build an integrated and collaborative approach to the problem, major centers merged in a national network (1). A consensus-based definition of IF was initially achieved, and a classification system based on etiology was agreed upon. Specifically, IF was defined as "a primary intestinal disease, leading to the failure of digestive-absorptive processes and/or impairment of water and electrolyte homeostasis, and requiring parenteral administration of at least 75% of total caloric intake for at least 4 weeks, or 50% of total caloric intake for at least 3 months." (1) Individual centers were identified to serve the network for specific diagnostic investigations (Fig. 1). This allowed the development and maintenance of high technical standards while helping to advance knowledge in specific fields in individual centers. A common diagnostic algorithm was defined for all cases of IF secondary to the so-called intractable diarrhea of infancy (IDI). In this article, we report the main results of the collaborative work of the Italian network: the diagnostic protocol for IDI, the data on the natural history of IF and the approach to the main complications of PN.
Diagnosis of Primary IF With Intractable Diarrhea
Intractable diarrhea of infancy was defined as severe and protracted diarrhea (stool output higher than 150 mL/kg per day, lasting more than 2 weeks), unresponsive to drug or dietary treatment, in a child needing PN for at least 50% of the daily energy intake (14). The diagnostic protocol for the child with IDI included a specific step by step evaluation (Table 1). To optimize this shared approach, a common operative protocol for biological samples was also defined (Table 2).
Because diarrhea may be the result of a secretory or osmotic process, separating into these 2 categories may help in the initial diagnostic approach (16).
Early-age Onset Diarrhea
Maternal hydramnios suggests congenital chloridorrhea or congenital sodium diarrhea, which are both transmitted with autosomal recessive pattern. The first is characterized by hypokalemia, hypochloremia, metabolic acidosis and high concentration of Cl- in the stool (exceeding Na+ and K+), the latter by hyponatremia, metabolic acidosis and large amount of Na+ in the stools. Intestinal biopsy is normal in both conditions.
In case of massive albumin intestinal losses, as shown by high levels of fecal α1antitrypsin, congenital enterocyte heparan sulphate deficiency should be considered. Intestinal biopsy shows normal histology and the etiology can be obtained only with specific staining for sulphated glycosaminoglycans.
In a male infant or young children with diabetes mellitus, thyroid disease, eczematous ichthyosis or hemolytic anemia, consider autoimmune enteropathy and specifically IPEX (Immune dysregulation, Polyendocrinopathy, Enteropathy, X-linked), a syndrome due to a mutation of FOXP3 gene. Intestinal histology shows severe villous atrophy with a marked infiltration into the lamina propria of activated T cells. Antienterocyte antibodies can be detected in serum samples.
Congenital microvillus atrophy (CMA) is inherited as an autosomal recessive trait. Histology shows villous atrophy with little crypt hyperplasia, in absence of a marked inflammatory infiltrate and an abnormal brush border with PAS staining. On EM, grossly abnormal microvilli and numerous vesicular bodies (secretory granules) and the characteristic microvillus inclusions are found in surface epithelial cells. The typical histological pattern may be detected in the large bowel.
Cytomegalovirus infection may induce severe diarrhea even in an immunocompetent host. Duodenal and colonic histology reveals severe inflammatory changes and typical nuclear and cytoplasmatic inclusions can be present. Cytomegalovirus infection role is supported by early antigen immunostaining or by PCR in intestinal specimens.
The presence of sugars in the stools despite the general nutrient manipulation, suggest glucose-galactose malabsorption. The disease, characterized by a normal intestinal mucosa, is inherited as an autosomal recessive trait.
Osmotic diarrhea may be the result of tufting enteropathy or, in the presence of abnormal phenotypic aspect, syndromic diarrhea. Tufting enteropathy is inherited as an autosomal recessive trait and is also referred to as intestinal epithelial dysplasia. Total or partial villous atrophy may be evident with crypt hyperplasia and normal or slightly increased density of inflammatory cells in the lamina propria. The histological hallmark is focal crowding and disorganization of the surface enterocytes which results in a tear-drop configuration of the affected cells (tufts), pseudocystic formation of the glands and abnormal regenerative crypts; on immunohistochemistry, faint and irregular laminin deposition is observed at the epithelium-lamina propria interface, whereas deposits of heparan sulfate proteoglycan are large and lamellar. Syndromic diarrhea is associated with low birth weight, facial defects and woolly, easily removable hair with trichorrexis nodosa associated with an unusual immunodeficiency with predominantly humoral abnormalities and a cellular deficiency. Intestinal histology shows moderate-to-severe villous atrophy with slight chronic inflammation and normal or hypoplastic crypts without epithelial abnormalities.
Evidence of nonintestinal autoimmune disease may be associated with autoimmune enteropathy. This may be the presenting syndrome or with a more subtle condition with few intestinal symptoms. Severe infiltration of activated T lymphocytes is observed in the lamina propria. The histopathology is similar to celiac disease except for a relative paucity of intraepithelial lymphocytes. Duodenal biopsy shows total villous atrophy, crypt hyperplasia and a dense lymphoplasmacellular infiltrate into the lamina propria. Crypt abscesses are identified in severe cases. Numerous cells express CD25 whereas HLA-DR is increased on crypt epithelium. Lesions can be seen in the stomach and colon as well. Circulating antienterocyte autoantibodies are detected in the majority of patients. Associated autoantibodies are mainly directed against nucleus, DNA, smooth muscle or mitochondria.
Other possible diagnoses include late onset CMA and postenteritic syndrome. The latter may present as a protracted diarrhea after an acute infection and is always associated to a severe villous atrophy with lymphoplasmacytic inflammation of the lamina propria.
Onset after weaning or formula containing polysaccharides with presence of sugars in the stools suggests sucrase-isomaltase deficiency. Duodenal biopsy is normal, and diagnosis is based on enzyme activity determination in intestinal mucosa.
Late-onset tufting enteropathy should be considered when there is no evidence of reducing substances in the stools.
EPIDEMIOLOGY AND NATURAL HISTORY OF IF
The network allowed us to record long-term data providing insight on the history of IF. Based on etiology and pathophysiology of IF, a classification system was proposed, discussed and applied. The etiologic classification of IF included short bowel syndrome; disorders of intestinal motility, including chronic intestinal pseudo-obstruction and aganglionosis; and structural defects of the enterocyte, including CMA and tufting enteropathy, multiple food intolerance/intestinal infections and autoimmune enteropathy. The usefulness of such classification system was tested, based on the hypothesis that a close link exists between etiology and natural history of IF. Such hypothesis testing, implying the opportunity to predict the outcome of patients based on etiology, was investigated through a prospective study. The classification of IF was finally validated and natural history of IF analyzed in relation to etiology. More than 100 patients with IF were enrolled by the Network in 5 years. The number of cases enrolled each year was relatively constant at about 15 (Fig. 2). Short bowel syndrome was the most frequent cause of IF, accounting for approximately 50% of cases (Fig. 3). Other etiologies were less frequent and their incidence was similar. A relatively small number of cases could not be included in any of the major etiologic groups, and in few children no etiologic diagnosis was obtained.
The natural history of IF was next investigated with reference to following outcome parameters: achieving intestinal sufficiency with permanent weaning from PN, permanent need for total or supplemental PN, intestinal transplantation, and death. The results showed a close association between etiology of IF and its outcome (Fig. 4). Children with structural enterocyte defects had the worst outcome with no chance of weaning from PN. Specifically, children with CMA had an increased risk of an early poor outcome, which was worse than those with epithelial dysplasia. On the other hand, all children with food intolerance were able to shift back to oral nutrition. Between these 2 extremes, a broad pattern of increasing severity was observed (Fig. 4). Thus our data suggest that children with structural defects of the enterocyte should be considered for early intestinal transplantation. This is now included in the indications for intestinal transplantation (2,17).
The knowledge of the natural history of IF provides clues for planning specific therapeutic approaches. However, a well-conducted and well-tolerated PN regimen probably remains the best lifelong option in children with permanent IF. The major limitations of PN are related to complications. Understanding these complications was another target of the collaborative Network.
COMPLICATIONS OF PN
The main complications associated with PN, thereby leading to death, or to the need for intestinal transplantation, include PN-associated liver disease, vascular thrombosis and recurrent central venous catheter (CVC) sepsis (Table 3). Hepatobiliary complications encompass a wide spectrum of hepatobiliary abnormalities, from asymptomatic biochemical changes, to steatosis, or cholestasis, with possible progression to end stage liver disease (2,18). Biliary sludge and cholelithiasis are frequent features in patients on long-term PN, and approximately 25% of short bowel patients will develop gallstones eventually requiring cholecystectomy (19). The pathogenesis of liver disease in patients on long-term PN is multifactorial, and it involves patient-dependent factors and PN-dependent factors (2,18). Potentially effective measures to prevent liver disease include: (9) preventing overfeeding; using 20% lipid emulsion containing medium- and long-chain triglycerides; providing pediatric amino acid solutions with age-appropriate amino acid concentrations including taurine; PN cycling; decreasing aluminum content in PN solutions; providing some (even minimal) oral/enteral intake to stimulate the enterohepatic axis; cycles of antibiotics or probiotics to suppress/prevent bacterial overgrowth which is a specific complication of children with short gut or with impaired intestinal motility; and performing prophylactic cholecystectomy in patients with short bowel syndrome. The only pharmacologic option proposed for PN-associated liver disease is ursodeoxycholic acid (UDCA). We first demonstrated that administration of UDCA to children on PN is effective in improving cholestasis, possibly reducing the occurrence of end-stage liver failure and, thus, modifying the natural history of IF (20). Further data supported this finding although it is not clear yet whether early administration of UDCA at the appropriate doses would be consistently effective to prevent PN-related liver complications in the long term (21) Central line-related sepsis is the other main severe complication of long-term PN, whose incidence is relatively related to the duration of PN itself (2,22). Infections can originate from skin or surrounding tissue of the CVC exit site, in the pocket (in cases of implanted CVCs) and in the tunnel or alternatively from bacteria translocating from the intestinal lumen (19,23). Systemic infection is defined as a CVC-related bloodstream infection and can be diagnosed by semiquantitative blood cultures from peripheral vein and from the catheter (19,23). Prevention of CVC-related sepsis requires strict asepsis during both CVC insertion and filter and infusion set replacements. Within the Network, the application of specific standardized protocols for the prevention of CVC-related infections was implemented and effective in reducing by 50% the incidence of severe infectious complications and resulted in a better nutritional management of children with IF (24).
The approach to the child with IF requires advanced professional expertise and close collaboration among specialists. In parallel, advanced technology for diagnosis, PN and surgery is also required. The care of patients with IF is a rapidly evolving field in which cooperation is essential for continuous technologic upgrading and improving the outcome for these children. The Network model was proposed to address the many aspects of IF based on the strategy that was adopted with for children with AIDS at the start of epidemics in the 1980s. Similar to AIDS in the 1980s, IF is a chronic severe condition, not curable, severely affecting the whole family life, requiring an enormous amount of resources and for which a broad spectrum of medical and technical competence is needed. The network approach to IF permitted sharing knowledge and resources of centers that are seeing a limited number of children, upgrading the complex diagnostic approach to the child with intractable diarrhea, development of effective strategies to improve PN and prevent its complications and limited the need for referring children to centers located far from home. The network approach provided scientific results as well, including new data on the epidemiology and etiology of IF and its natural history. Although challenges of capacity and cooperation exist, the transfer of this national (Italian) model to the international (European) level might be an effective option to maximize the advantages of network-based care.
1. Guarino A, De Marco G, For the Italian National Network for Pediatric Intestinal Failure
. Natural history of intestinal failure
, investigated through a network-based approach. J Pediatr Gastroenterol Nutr
2. Goulet O, Ruemmele F, Lacaille F, et al. Irreversible intestinal failure
. J Pediatr Gastroenterol Nutr
3. Sudan D, DiBiaise J, Torres C, et al. A multidisciplinary approach to the treatment of intestinal failure
. J Gastrointest Surg
4. Koehler AN, Yaworski JA, Gardner M, et al. Coordinated interdisciplinary management of pediatric intestinal failure
: a 2-year review. J Pediatr Surg
5. Walker-Smith JA. Intractable diarrhoea of infancy: a continuing challenge for the paediatric gastroenterologist. Acta Paediatr
6. Dorney SF, Ament ME, Berquist WE, et al. Improved survival in very short small bowel of infancy with the use of long-term parenteral nutrition
. J Pediatr
7. Zlotkin SH, Stallings VA, Pencharz PB. Total parenteral nutrition
. Pediatr Clin North Am
8. Burnes JU, O'Keefe SJ, Fleming CR, et al. Home parenteral nutrition
a 3-year analysis of clinical and laboratory monitoring. J Parenter Enteral Nutr
9. Koletzko B, Goulet O, Hunt J, et al, For the Parenteral Nutrition
Guidelines Working Group. Guidelines on Paediatric Parenteral Nutrition
of the European Society of Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) and the European Society for Clinical Nutrition and Metabolism (ESPEN), supported by the European Society of Paediatric Research (ESPR). J Pediatr Gastroenterol Nutr
10. Kaufman SS, Atkinson JB, Bianchi A, et al. Indications for pediatric intestinal transplantation
: a position paper of the American Society of Transplantation. Pediatr Transplant
11. Goulet O, Baglin-Gobet S, Talbotec C, et al. Outcome and long-term growth after estensive small bowel resection in the neonatal period: a survey of 87 children
. Eur J Pediatr Surg
12. Bond GJ, Mazariegos GV, Sindhi R, et al. Evolutionary experience with immunosuppression in pediatric intestinal transplantation
. J Pediatr Surg
13. Grant D, Abu-Elmagd K, Reyes J, et al. On behalf of the Intestine Transplant Registry. 2003 report of the intestine transplant registry: a new era has dawned. Ann Surg
14. Kocoshis SA, Beath SV, Booth IW, et al. Intestinal failure
and small bowel transplantation, including clinical nutrition: Working Group report of the second World Congress of Pediatric Gastroenterology, Hepatology and Nutrition. J Pediatr Gastroenterol Nutr
15. Catassi C, Fabiani E, Spagnuolo MI, et al, and the Working Group of the Italian Society of Pediatric Gastroenterology and Hepatology (SIGEP). Severe and protracted diarrhea: results of the 3-year SIGEP multicenter survey. J Pediatr Gastroenterol Nutr
16. Guarino A, De Marco G. Persistent diarrhea. In: Allan Walker W, Goulet O-J, Kleinman RE, et al eds. Pediatric Gastrointestinal Disease: Pathophysiology, Diagnosis, Management
, 4th ed. BC Decker Inc, 2004:180-93.
17. Ruemmele FM, Jan D, Lacaille F, et al. New perspecives for children
with microvillous inclusion disease: early small bowel transplantation. Transplantation
18. Cavicchi M, Beau P, Crenn P, et al. Prevalence of liver disease and contributing factors in patients on home parenteral nutrition
for permanent intestinal failure
. Ann Intern Med
19. Howard L, Ashley C. Management of complications in patients receiving home parenteral nutrition
20. Spagnuolo MI, Iorio R, Vegnente A, et al. Ursodeoxycolic acid for treatment of cholestasis in children
on long-term total parenteral nutrition
: a pilot study. Gastroenterology
21. Sordino D, De Marco G, Bruzzese E, et al. Efficacy of Ursodeoxycholic acid for parenteral nutrition
-associated cholestasis in children
with intestinal failure
. J Pediatr Gastroenterol Nutr
22. Colomb V, Ricour C. Home parenteral nutrition
. Clin Nutr
23. Marolis J, Kalfarentzos F. Complications of parenteral nutrition
at the end of the century. Clin Nutr
24. Mambretti D, Aceto B, Bruzzese E, et al. Impact of parenteral nutrition
and its septic complications on the long term nutritional outcome of children
with intestinal failure
Annual Meeting of ESPGHAN. Prague June 4-7 2003. Abstract Book p.251.