Short bowel syndrome (SBS) is defined as a spectrum of small intestinal dysfunction for a major portion of small intestines being resected and the patients need chronic total parenteral nutrition (TPN) in lifetime. However, a long-term use of TPN is associated with significant morbidity and mortality[ ^{1 - 3}], such as system infection and progressive cholestatic liver disease and many patients could not afford the high cost of the nutrition support. Several bowel elongation procedures were successful in increasing the absorptive surface area, but these methods were not widely applied for serious complications. Small intestinal transplantation may be a valid option, but the result is unsatisfactory for immunologic problems.

The animals survived after the operation with no evidence of peritonitis or sepsis. All the rabbits had gradual weight loss from median 2.1 to 1.8 kg and the survival after the operation was no longer than 8 wk, while the body weight of pigs increased from median 42 to 51 kg which were executed at 3 mo after the surgery according to the scheme.

In group A, severe adhesions were found between the graft and surrounding intestine. There was almost no blood supply in the grafts and the lumens were full of dry stool or pus at 2 and 4 wk after surgery. At the 8th postoperative week, the grafts totally disappeared, similar to group B and one pig in group C and there were only two blind-ended pouches. The other one in group C seemed to be the most successful one considering the small intestinal continuity was restorative. But the graft shrank from 3 to 0.2 cm and became a narrow-ring without flexibility (Figure 3).

Histological observation showed acute inflammation in the tissue during the first two weeks. In the first week, there were focal fibrinoid necroses and infiltration of a large amount of neutrophils and leukomonocytes (Figure 4A and B). In the second week, there was a further development of the inflammation. The fibrinoid necroses and cell infiltration became dispersed and inflammatory exudates of neutrophils and pus cells could be seen (Figure 4C and D). At the same time, a lot of micrangiums and fibroblasts appeared in the scaffold. By the 4th postoperative week, the infiltration of leukocytes and leukomonocytes mitigated. No well-formed structure could be seen and all the layers were contorted in the narrow-ring of group C (Figure 5). The thickening serosa and infiltration of a small amount of leukocytes and leukomonocytes could be discovered. Immunocytochemical stain with smooth muscle actin showed smooth muscle cells in the narrow-ring of ADM with disordered structure.

Several surgical procedures have been used to improve the SBS[ ^{5 , 6}]: (1) to decrease transit time of the bowel content by creation of intestinal valves or interposition of bowel segments; (2) to increase mucosal surface area; and (3) to increase the bowel length or to transplant the bowel. Objective assessment of these surgical techniques was deficient due to a relatively small number of patients. The small intestine transplantation is becoming a less morbid therapy in recent years, but it remains complicated and is limited due to the donor availability.

ADM is a dermal biomaterial in which all of the cellular elements have been removed. The biologic properties of ADM, including its ability to support tissue regeneration repopulation with fibroblasts, revascularization, new collagen deposition and eventual absorption and replacement with native tissue permit its use in tissue reconstruction. Now ADM has been used successfully in repairing the defect or damaged tissues and organs both in preclinical animal studies and in clinical applications[ ^{7 - 11}], such as burn surgery, herniorrhaphy, urethroplasty, pelvic floor reconstruction, contaminated abdominal wall defects and fistula.

We supposed that the ADM can be used as a scaffold for tissue regeneration with the normal morphology of the mucosa, submucosa, muscle and serosa layers. An in vivo study in dog models showed that an extracellular matrix (ECM) plus autologous muscle could facilitate the reconstruction of the esophagus with near normal biomechanical properties and structure[ ¹²]. Pahari et al[ ¹³] used the acellular human dermis in the form of tubular scaffolds for intestinal elongation. Intact mucosa with well-formed crypts was evident at 6 mo after operation[ ¹³].

But it did not develop like what we had designed in the study. The grafts were almost completely absorbed in two or three months. Pus could be seen in the lumens of ADM and histological examination showed the inflammatory exudates of neutrophils and pus cells. Thus, the ADM might be a source of infection in this study, though we have tested that the ADM can be used in contaminated wound for its resistance to infection[ ¹⁴]. Inadequate blood supply and the rapid absorption were the most likely reasons. All the ADM depends on the blood infiltrating from the two anastomotic ends and the supply was very limited comparing with the granulation tissue of the wound surface. Immunocytochemical stain with smooth muscle actin showed smooth muscle cells in the narrow-ring at 3 mo after surgery, which is an exciting event and consistent with our previous study [ ¹⁴].

The biggest challenge of bowel tissue engineering is the recovery of peristaltic activity of the regenerated bowel, which needs both smooth muscle and reinnervation of the regenerated bowel. It seemed that there was no evidence of innervations of the neointestine to date. Even the neointestine had well-formed structures, the dysmotile segment may become dilated, resulting in fecal stasis, bacterial overgrowth and malabsorption.