Small bowel transplantation : Current clinical status

D SJGALET. Small bowel transplantation: Current clinical status. Can J Gastroenterol 1991;5(4):154-160. W ith recent refinements in immunosuppression techniques, the first successful reports of small bowel transplantation in humans have now heen ma<le, increasing interest in bowel t ransplantation among clinicians and patient:i alike. This article reviews recent J evelopments in umlerstan<ling of the functional capabilities an<l requirements for effective immune suppression in bowel transplantation. Both experimental an<l cl inical experience with transplantation are <liscusse<l, as are the areas which appear to offer the most promise for future developments. Finally gui<lelines for consi<leration of patient selection for this procedure are reviewed.

T l IE SI !ORT BOWEL SYNDRl)MI that re~ulcs from massive losses of smal l howel continues to he a difficult problem despite the c,)mnHm use of long term total parenteral nutrition (TPN).TPN is expensive (1 ), limit, the lifestyle of the patient anJ the pacienc's fami ly (2), and requires contin ued long term venous access.Lung term TPN in children 1s even mmc prohlcmatic, with increased nutrient req uirements, difficulty with patient compliance, and the risk of as.,oc..iatcd li ve r damage, espec ially in the very young infant (3 ).Because of these factors, the lifelong mortali ty from direct complicatiuns nf long term TPN in the pediatric age group exceeds 15% (4).ln rhe neonate, mass ive resection of the smnl I howcl most commonly results from such condirnms as nccrmizing entcrocolitis, malmtauon with volvulus, anJ strangulated abdominal wall defects (4).ln the ad ul t population, mas~ive howcl resection most commonly resuh s fro m mesenteric vasc ular accidents, inflammatory bowel dise,N: and trauma (5,6).T hese condit i,1n, result in a stable incidence of new patients who require long term nutritional support.
Small bowel Lransplantation as a treatment for shorr bowel syndrome has been considered for many years, but only in the past year have successful atte mpts at bowel transplantation been reported (7,8).Inevitably, such resu lts provoke inquiry from patients and fam ilies about the applicability of ~uch new therapies to their own situation.Since the last major reviews of small bowel transplantation were pub I isheJ, much new information has been presented (9,10).This article will summarize the research which has led to the success of the cases reported, the current Status of bowel transplantation for treatment of short bowel syndrome, and possible d irections that future developments may take.

EXPERIMENTAL TECHNIQUES AND MODELS
Transplantation of vascu larized organs, including the bowel, was fi rst artempted by the French surgeon Alexis Carrel ( 11).The problem of rejection was soon recognized, anJ interest languished until the late 1950s, when Lillehei's grnup ( 12) in M innesota in-vestigateJ the effects of ischemi a on gut organs.They found that cooling and perfusion with heparinized saline would all ow reliable preservation of the sma ll bowel for 4 h; this preserved bowel coulJ be re-impla nted and would function indefinitely as an autografL (12).Their model consisted of a one stage operation; the superior mesenteric vesse ls were isolated, clamped and divided, and the bowel was flushed and then revascularizeJ using the mesenteric vessels of a similarly prepared recipient, re-establishing bowel continuity using end-co-end anastomosis of the native duodenum and ileum Lo the graft.They a lso used isolated loops of bowel placed in the neck, permitting the study of immunosuppressive agents and graft function in a con trolled fashion such that rejection of the graft would nor lead to the death of the animal (13 ).They had no success with a llografts, and with others found minimal survival aJ-  CsA Cyclosporine, GVHD Groff-versus~host disease vantage using the immunosuppressants available at chat time -azathinprine and steroids (14-16).
The dog model was used to evaluate th e rejection process in detail, and after the introduction of cyclosporine in rhe 1970s, the orthotopic model of bowel transplantation in the Jog was the first used to assess the effects of cydosporine on small bowel transplantation ( 17,18).The significant prolongation of graft survival demonstrated by Reznick er al (17,18) frnm ToronLo in a landmark study wa~ encouraging, hut the overnll succe~s raLe wa~ low (further details are discussed in the sec Lion on immunosuppression).
The description of hetermopic bowel tramplantaLion in rhe rat by Monchik and Russell in 1971 ( 19) greatly facilitated study in this field.The rat model has since served as the standarJ for initial inveHigation~ of immunosuppression, function and techniques.T echnically, the procedure is similar to that descr ibed for larger animals, the main problem being the small size of the vessels.The aorta is used as the conduit for the superior mesenteric artery, and the grafr is reva~cularized using rhe recipient's inferior vena cava and aorta.The bowel is left as a Thiry-Vi lla fistula with proximal and distal stomas (hereroropic graft).Initial attempts to re-establish gastrointestinal continuity immetliately using the graft were plagued with a high failure rate (20); however, with experience th is improved (21,22).This mndel then became the standard for (indefinite survival In seven of 16) (39) investigation of small bowel transplantation in rats.Within the rat model the availahility of genetically defined strains of animals has greatly faci li tated investigaLion of the immunological consequences of small bowel transplantation ( 18,23 ).The current availahil ity of monoclonal antibodies to various cell populations in the raL should provide further valuable information.
The pig is an excellent model of h uman bowel physiology, with a more defined genetic lineage than the dog (24).Although earlier attempts had hecn made ( 2 5), Ricour and colleagues (26) were the first tn perform successful small howel transplantation in the pig and ach ieve :illogrnft ~urv ival.The 1cchniques used paralleled those used in Lhe dog.
In reviewing reports of experimental models of small bowel transplantation, tine must remember that rejection responses and function va ry considerably depending on the model.Inbred strains of rats have varying rejection responses, while outhrcd larger animals generally undergo a more vigorous reaction ( 13,19,25).Simi larly, different models within the same species may exhibit different rejection responses: grafts drained via the portal circulation of the recipient may have a survival advantage, while caval drainage permits more vigorous rejec-Lion (27,28).This may be due to an effect of the liver on soluble antigen processing: liver transplantation has long heen known to permit the survival of subsequent grafts from the same donor, with normal reJection of 'thi rd parry' grafts (29,30).Finally, graft survival in heteroLOpic models does nor imply that the graft is necessarily capable t)f supporting Lhe animal nutritionally.

REJECTION AND GRAFT-VERSUS-HOST DISEASE
The histology of small bowel allograft rejection has been well described, and can be divided into three phases ( 19,3 1 ).In the rat at day 3, there is no change detectable with routine light microsct)py, but at days 6 and 7, lymphocytes and plasma cells begin to infil trate the lamina propria.In phast• 11, over days 8 and 9, the mfiltrare intensifies and extends to the muscularis propria.There is associated shortening and blunting of the villi and scattered epithelial sloughing.In phase Ill, which occurs after the 10th Jay, there is complete mucosa!destruction and transmural infiltration with lym phocytes and polymorphonuclear leukocytes.Rejection is similar in J ogs and pigs, but has an accelerated time course; rejection is complete by days 7 to9(l 2-H,26).
The small bowel is unique among transplanted vasculnrizeJ organs because of the large population of immunocomperent cells which can become activated followi ng transplantation into a non-major hisrocompatibility complex (MHC ) identical recipient.This leads to a phenomenon known as 'graft-versus-host disease', initially described following bone marrow transplantation.This problem was recognized by the early investigators of small bowel transpl:rnrntitm ( 12), and examined in derail using inbred strains of rats (1 9).When both rejection and graft-versus-host disease me possible, rejection is the dominant resptinse; however, when animals are treated with potent immunosuppressants such as cyclospor'lne, graft-versus-host disease becomes important.This results in an activation of graft-derived T cells directed against cell surface antigens of host epidermal cells (possibly primitive seem cell markers (32)).The disease complex seen after small bowel transplantation is a phase of poor appetite, 156 red ears and snout, diarrhea, weight loss and hunched poswre, from which most animals recover spontaneously (31 ).There may be differences in the intestinal manifestations of the graft-versushost disease seen after small bowel versus bone marrow transplantation (34 ); this possibility requires further investig:ition.

REJECTION MONITORING
In the clinical sening it is important to monitor for rejection.The nonspecific nature of the early stages of acute rejection noted above were found to limit the usefulness of suct ion biopsy in monitoring for rejection in the one clinical case that has been well documented (35,36).This problem, coupled with the difficulty and possible hazards of obtaining biopsy material, have prompted invest igarors to search for alternate methods of monitt)ring for rejection.The functional capacity of the bowel has been ex ploited; the best characterized of these markers is the maltose absorption rest (37).This test measures the ability of the bowel mucosa co split maltose into glucose and then transport it across the enterocyte into the circulation, where it is detected hy monitoring of serum glucose levels.The resulting rise is diminished when rejection is in the initial stages.
A simpler methtx.lmonitors changes in the permeability of the howcl wall by measuring the urinary excretion nf chromium-5 l EDT A instilled in the lumen of the graft.The bowel is normally impervious to this compound; an increase in absorption occurs with early graft rejection (38).This test has been used clinically and has been found to correlate well with biopsy evidence tif rejection (7).

METHODS OF IMMUNOSUPPRESSION
At present, the factor which I imits the use of small bowel transplantation clinically is the lack of a reliable protocol for immunosuppression.After the description of the surgical techniques required for successful small bowel transplantation in dogs, a number of different immunosuppressants were used, with limited success.These included steroids, azathinprme, anltlymphocyte globulin and graft irradiation (14-16,39).It was not until the intro• duct ion of cyclosporine in the late 1970s that a significant prolongation ti small bowel allograft survival was achieved ( l 7).The average survival in I J dogs treated with cyclosporine (25 mg/kg/day, tntramuscul ar injection for the first 28 days, then oral administration) was 91 d,1ys, while umreated controls survived an average of 12. 5 Jays.
However, it is important to note that only three dogs survived more than 60 days, and two of these succumbed to rej ection at 210 days.The importance of using parenteral cyc kisponne wa~ demonstrated in a fol ltlw-up stud y from the same group, in which a third set of animals was given oral cyclosporinc: seven of JO di ed of acute rejecllon an average of 30 days post transplant (18).
Once it had been shown that cyclospnrine prolonged the su rvi val of intestinal allogrnfb in dogs, a serie~ of studies in the more controlled rat model appeared.A dosage nf cyclo-~porinc of 15 mg/kg/day allowed indefinite survival of grafts in unidirectional and two-way rejection and grafL-versus• host dbca:,e models ( 22 ).A similar dosage of cyclosporine was ab() shown lO control grnfr,versus,host disease (40).
The model of bowel tramplantation available whic h most closely resembll's Lhc situmion in humans 1s the pig.Mnnotherary with high dosage intrn• vern)uscyclosporine ( 15 mg/kg/day) for LO da ys post transplant, followed by continued high dosage oral therapy (30 mg/kg/day) , has allowed successful trnnsplanrntion in chis model ( 41 ).These high dosages see m necessary: protocob ustng lower doses have an increased rmc nf ,mimal death, ;i i though this hns nnt been shown to he due to rejection (42).In the pig, combi ned stemid and cyclosporine use increased the rnte of infectious complications anJ did nor reduce the rat e of rejectton (4l) , while graft irradiation was nnc useful ( 43 ).It is interesting to note that stopp111g cyclosponne after two to three months of cominuous therapy did not lead to a ll ngra fr reject itln ( 41 ) .The recipient may develop tolerance to the bowel a llografr in a fashion similar Ln that described for liver transplams (29).
A tremendous effort ts underway to improve methuJs of imm unosuppre~sion (44).These methods have in large part focused on pharmacological control of rejection post transplantation.The new drugs FK506, rapamycin and 15-deoxyspergualin have already been shown to allow successful small hnwcl transp lantation in rat models, and arc currently being reviewed in large an imal and human studies (45)(46)(47).Alternative strategics, such as pretreatmenl of grafr with monoclonal antibod ies ( 48) or o( host with donorspecific transfusion, arc also usefu l experimentally ( 49).The challenge m present will be tn develop effective combined strategics which optimize both pretreatment of donor and recipient, and post transplamation immunosuppression.

FUNCTION OF TRANSPLAN TED BOWEL
In the nrst studies of small howel tran plantation Li ll ehci and coworkers ( 12) showed t hat animals could survive indefinitely fo llowing autot ransplantation.No specific study of nutr ient absorption was performed, but gross malahsorption of fat wa~ evident for two tu three weeks, and then suhsided.They were also able to demonstrate regeneration of severed lymphatics after three weeb (50).Ml>re detai led studies demonstrated that, fo llowing transplamation of the smal I bowel, dogs had a pe riod of two to three weeks of d iarrhea, we ight loss and abnormal motility with steatorrhea.These changes reversed over the ensuing months; hy six momhs fu nction had norma lized (51 ).An identical pmrern of changes cou ld he produced hy denervat ing the bowel and dividing the lymphatics; it was concl uded that the funct ion al alterariom of t rnnsplanintion observed were due to denervation aml lymphatic disruption, and were reversihlc.
Several invest igawrs have ,hmvn that th e t ransplanted bowel hccllmes electricall y a utonomous, and that thb may result in a reduction in glucose, glycine, water and sodium absorption (52,53).Chloride b most significmtly affected, with a net secretion in some instances, possibly due t1i a lo:;s of the nurmal inhibition of crypt chloride secretion by rhe autonllm1c nervous system following small howel transplantation.Limited studies ol bowel function al the cnterocyte level hnve been puhl ishcd: the dcctrophysiological parameters of the bowel seem norm.ii at nine days post transplam hut deteriorate rapidly if reJecnon 1lccurs (21).
In animab that arc nutritiunally dcpcndem on a small howel transphmt, long term reductions in fat, protein and carbohydrate absorption have hcen demonstrated ( 54-56).Cyclosporine itself may also affect nutrient absorption.Preliminary evidence wggcsts that cyclosporinc reduces glucose, alanine and fatty acid uptake in normal rnb, and in autogrnfteJ bowel in dogs (56,57).Overall, these findings suggest th at both the transplantation proce,, itself and rejection (if it occurs) significantly afft..•ct the motility, neural and transport functions of small howel.When the major port iun of the snrn II hnwel is trnn,plantcd, recipiem grnwth b near normal in most models (22,41 ); however, the length required to sustain the recipienr post transplant has n<ll been determined.Civen the present interest in triab of hl)Wcl tmnsplanuuion in humans, th is area requires further invest igat i1m.
As noted earlier, in considernuon nl 1hc possible advantages of porrn l drn111agc, l1ver trnnsph111tation reduce~ rCJl'Ction llf co-transplanted grafts.This may he by nomclf antigen proccs~ing hy t he liver or release of soluble MHC nntigens wh ich then hlnck cytotoxic antihlld ies 00).In cl inical t ransplantallon, improved survival of renal grafts following I iver transplanwt1on has bec:n dearly documented, even in the presence uf preformed antibodies to the graft (67) .In addition to this, anarnmicf!I considerations haJ defined the concept of a multivisceral transplant or 'cluster operation' for combined li ver and pancreaticoduodena I pathology (65,66).A complete visceral transplant -including stom ach , I iver, panc reas and various lengths of bowel -can be performed for extensive celiac pathology.In this situation , standard immunosuppressive therapy with cycl<)sporinc, prednisone and antilymphocyte globulin have allowed prolo nged surviva l of the associated howel (65,66) .It seemed logical co e xtend this to include a combined small bowel and I iver graft for short bowel syndrome with assoc iated primary liver disease, as reported by Grant (7).Grant described a typi cal immunosuppress ive protocol for isohned li ver transplantation which has prevented rejection in the as-soc iated bowel in three patients to da te (7, personal communication).
Altho ugh.initia l experience with the c luste r opera ti on was entirely clinical and no maj or problem with rej ection of the associated bowel was reported, an experimental study in rats has shown that the cluster operation does not ' protec t' the associated bowel (68 ).Howeve r, there are no experimental c.la ta dealing specifically with the issue of combined small bowel and li ver transplantation.C linical success a lready ac hieved certainly justifies continued interest in this area.Whethe r or not a person with short bowel syndrome and no associated liver disease shoulc.1 receive a combined liver-sma ll bowel graft, in hopes that the li ver will reduce the immunoge nicity of the bowe l graft, rema ins to be seen .The survival of patients receiving an iso-1990;3"~5 : 18 1-4.8. Sc hroeder P, Goulet 0, Lear PA.
Small bowel transplantation: Europe,111 experience.Lancet 1990;3 lated liver graft for primary liver disease (over 70% fi ve year survival 169]) demonstrates that the li ver tra nsplant portion of a combinec.lliveM mall bowel transplantation would be unlikely to harm the recipient.Selected patients with life-threate ning complications of short bowe l syndrome and primary liver disease can ethically he considered for suc h therapy (70).With short howel syndrome alone (and life.threatening complications which prevent combined support with TPN), isolated small bowel transplantation could be considered (8).Gi ven the complexities of the tec hnical and immuno logical aspects of the procedure, the preliminary work should he concentrated in a few experienced centres.O ngo ing e valua tion of the results will permit an accurate assessment of the risk/bene fit ratio of this procedure (70).