Auxiliary transplantation of the fetal liver II. Functional evaluation of an intraabdominal model

Auxiliary Transplantation of the Fetal Liver II. Functional Evaluation of an Intraabdorninal Model By Alan W. Flake, Michael R. Harrison, Loie Sauer, N. Scott Adzick, Jean-Martin LaBerge, Thomas M. Krummel, and M. Michael Tbaler San Francisco, California 9 To evaluate the use of the fetal liver as an auxiliary graft, w e have developed a model of intraabdominal heterotopic transplantation of late gestational fetal lamb livers into weanling lambs, Thirty-eight transplants have been performed of which 31 were technically successful, T w e n ty-three grafts functioned for intervals of 5 to 22 days after transplantation. Grafts w e r e functionally evaluated by analysis of total bile acid and bilirubin excretion. To determine whether host liver excretory function would influence function of the graft, common bile duct ligated recipients were compared with recipients with normal host liver function. W e found that (1) intraabdominal auxiliary transplantation of the fetal lamb liver is technically feasible; (2) the fetal liver graft is capable of rapid adaptation and can assume a significant portion of host excretory function; and (3) excretory function of the fetal liver is proportional to the functional demands of the host. Auxiliary transplantation of the fetal liver is a promising alternative to current methods of liver transplantation. 9 1987 by Grune & Stratton, Inc, INDEX WORDS: Fetal organ transplantation; auxiliary liver transplantation.

RTHOTOPIC LIVER transplantation currently offers the only hope for children dying of liver O disease. ~ Although results of orthotopic transplantation have steadily improved, serious problems remain. 2 The most serious limitation of liver transplantation in children is organ availability. As many as a third of children in need of a transplant die waiting for a suitable organ) Others are anatomically unsuitable for standard orthotopic transplants and die for lack of an alternative treatment. Transfusion requirements remain high in orthotopic transplantation4 increasing the threat of transfusion related complications. And finally, orthotopic liver transplantation is among the most expensive procedures in medicine and carries a high cost for the patient and society. 4 The recent expansion in liver transplantation centers will not solve these problems. They can only be solved by finding new sources of donor organs and developing strategies for their utilization. One potential source of donor organs that is currently completely wasted is the anencephalic fetus. The use of livers from anencephalic fetuses could potentially solve the current limitations of size discrepancy and organ availability for the pediatric recipient. To investigate the feasibility of heterotopic fetal liver transplantation, we recently developed a sheep model in which we transplanted fetal livers into the adult Journal of Pediatric Surgery, Vo122,No 6 (June), 1987:pp 559-565

neck. 5This model supported the technical feasibility of fetal liver transplantation and allowed insight into the anatomic and physiologic requirements for fetal liver survival in the postnatal environment. In this report, we describe (1) a successful lamb model of intraabdominal auxiliary fetal liver transplantation; and (2) an excretory functional evaluation of the intraabdomihal auxiliary fetal liver graft. MATERIALS AND METHODS Fetal lambs of 135 to 145 days gestation were used as liver donors in all experiments. Recipients were weanling lambs weighing from 10 to 20 kg (5 to 8 weeks old). Anesthetic and fetal surgical techniques have been described elsewhere. 6 All livers were harvested and transplanted immediately with cold ischemia times of <2 hours.

Fet a l Liver H a r v e s t and Preservation Our techniques for fetal liver harvest and preservation have been previously described) Briefly, the fetus is delivered by hysterotomy and one of its two umbilical veins is cannulated. Heparin is administered and the fetal chest is opened by median sternotomy. The suprahepatic inferior vena cava is severed at its junction with the right atrium resulting in rapid fetal exsanguination. Simultaneously, umbilical perfusion is initiated with cold Ringers lactate solution and the umbilical cord divided. Umbilical cold perfusion is maintained throughout liver harvest. The donor is opened longitudinally in the midline and the diaphragm incised with ligation of the phrenic veins. The ligamentous attachments of the liver are divided. The infrahepatic cava is exposed and divided above the renal veins. The portal vein and common bile duct are ligated and divided and the hepatic artery dissected free with ligation of its branches. A segment of donor aorta is harvested in continuity with the hepatic artery. The liver is placed in an ice bath until transplantation.

Preparation of the Recipient During fetal liver harvest, a second team prepares the recipient. The lamb is placed on its left side and a long right subcostal incision is performed. To create space and facilitate use of the right renal artery, a right nephrectomy is performed. The inferior vena cava is exposed by blunt dissection. Attention is then turned to the porta hepatis. An

From the Department of Surgery, Division of Pediatric Surgery, University of California, San Francisco. Presented at the 35th Annual Meeting of the Surgical Section of the American Academy of Pediatrics, Washington, DC, November 1-2, 1986. Address reprint requests to Michael R. Harrison, MD, 585 HSE, University of California, San Francisco, CA 94143. 9 1987 by Grune & Stratton, Inc. 0022-3468/87/2206-0023503.00/0

559

560

FLAKE ET AL

adequate length of portal vein is exposed and the recipient common duct is cannulated proximally and distally and divided. A recipient cholecystectomy is performed.

;NARE ON P IMON

Auxiliary Transplantation of the Fetal Liver Cold umbilical perfusion is maintained until the final anastomosis (umbilical vein to portal vein) is begun (Fig 1). All anastomoses are performed using x3.5 magnification and standard vascular techniques. The fetal liver is placed in the prepared space so that the donor and recipient vena cava are aligned, and the donor suprahepatic vena cava is directed caudad. A long side-to-side (5 to 7 cm) recipient infrahepatic to donor intrahepatic vena cava anastomosis is performed. The liver is then rotated approximately 120~ clockwise so that the umbilical vein is in alignment with the portal vein. The donor aorta on its hepatic artery pedicle is swung inferiorly around the caval anastomosis and an end-to-side donor aortic to recipient right renal artery anastomosis is performed. Finally, a partially occluding vascular "C" clamp is placed on the portal vein, the umbilical per fusion is stopped, and the umbilical vein is anastomosed end-to-side to the portal vein. With this technique, cold perfusion is maintained until the final 20 minutes of the procedure. After release of the clamps, a biliary drainage catheter is placed in the fetal gallbladder. If incremental host liver defunctionalization is planned, a snare is placed around the portal vein distal to the umbilical vein anastomosis and the hepatic artery is ligated (Fig 2). Catheters are tunneled externally and the incision closed.

I 31LE

DONOR BILE DRAINAGE

Fig 2. Modifications of the basic model. Proximal and distal host common bile duct catheters allowed analysis of host liver excretory function, obstruction of host excretory function, and return of daily bile output to the enteropehatic circulation. Incremental host liver defunctionalization was achieved by hepatic a r t e r y ligation and placement of a snare around the portal vein that could be tightened postoperatively.

way)7 Bilirubin determinations were by the method of Billing et al.8 All statistical comparisons were by the Student's independent ttest.

Monitoring the Transplant Bile flow was indicative of liver viability and a successful transplant. Catheterization of the host common duct allowed external collection of host liver bile excretion, or, obstruction of host liver bile flow. Daily bile volumes and, in selected animals, serum and bile concentrations of total bile acids and bilirubin were compared with and without host liver biliary obstruction. Bile was returned to the recipient once a day through a catheter in the distal common bile duct (CBD). Tissue for histology of the graft was obtained at sacrifice.

Bile Acid and Bilirubin Determinations Serum and bile total bile acid determinations were performed by Sterognost Kit Assay (Nyegaard Diagnostics Division, Oslo, Nor-

PO AL N ,

.~ LONG SIDE-TO-S)DE

CAVAL-CAVAL

ANASTOMOSIS-

ANASTOMOSIS

J

~

;!

, !

~;

~-~/~ ~,,L"

I ,

C~'-'-;'~'.,,;"-..;

HOST

V~NA~AVA~--

jUMBLCALVEN

.~~..~\

//~,~,,~-~-~

~ I ~ '

~q" -~ ~ " ~' ~ ' ~'J

~\\~ \ ~ \ ,,~

I ,i ~

I

\J~

" PANCREAS

~ ~"

,,,l/ FETAL VENAOAVA

\ \ DRAINAGE \\ CATHETER

I',

I[

ii

Fig 1. Schematic diagram of the intraabdominal fetal liver transplant model. Note the long vena caval anastomosis for graft fixation and the use of the umbilical vein to achieve portal inflow.

Immunosuppression Cyclosporin (10 mg/kg/d IV) and solumedrol (20 mg/kg/d IV) were administered to all recipients beginning at the time of transplantation. Drugs were continued until animal sacrifice. RESULTS A f t e r n u m e r o u s u n s u c c e s s f u l a t t e m p t s w i t h a variety o f o t h e r a n a t o m i c a r r a n g e m e n t s , 38 a u x i l i a r y i n t r a a b d o m i n a l f e t a l liver t r a n s p l a n t s w e r e p e r f o r m e d as d e s c r i b e d above. T h i r t y - o n e o f t h e s e w e r e t e c h n i c a l l y successful w i t h t h e g r a f t r e m a i n i n g v i a b l e until t h e a n i m a l s d e a t h . O f these, 23 g r a f t s s e c r e t e d bile for i n t e r v a l s o f 5 to 22 d a y s a n d c o n t r i b u t e d to o u r f u n c t i o n a l analysis. O f t h e seven t e c h n i c a l failures, five w e r e d u e to a n a s t o m o t i c p r o b l e m s r e s u l t i n g in t h r o m b o s i s o f t h e d o n o r a o r t i c s e g m e n t ( N = 22), u m b i l i c a l v e i n ( N = 2), or b o t h ( N = 1) w i t h f a i l u r e o f t h e g r a f t to f u n c t i o n . T w o w e r e a n e s t h e t i c d e a t h s t h a t in r e t r o s p e c t were avoidable. F i v e o t h e r e a r l y f a i l u r e s w e r e d u e to o v e r v i g o r o u s efforts at host liver d e f u n c t i o n a l i z a t i o n b y h e p a t i c a r t e r y l i g a t i o n a n d p o r t a l v e i n o c c l u s i o n at t h e t i m e of transplantation. Although these animals survived the procedure, they appeared encephalopathic postoperatively a n d on d e a t h h a d n e c r o t i c h o s t livers a n d f u n c t i o n i n g grafts. T h e final t h r e e e a r l y f a i l u r e s w e r e d u e to g r a f t d r a i n a g e c a t h e t e r o c c l u s i o n by c l o t in t w o cases, a n d sepsis s e c o n d a r y to i n a d v e r t a n t e n t e r o t o m y in one case.

AUXILIARY TRANSPLANTATION OF THE FETAL LIVER

The majority of recipients with successful transplants were killed when evidence of sepsis or purulent bile drainage was noted. Other late causes of death were small bowel obstruction in three cases, arterial line bleeding in two cases, and an intraabdominal abscess in one animal. In each case the fetal liver was viable and secreting bile at the time of death.

561

Bile v o l u m e ( t a l / 2 4 h r s ) ZOO

180I / ~ : ) ~ 140 120 IQO 80 80 40 ~0 1 2 3

Rejection and Immunosuppression Our first two recipients were not immunosuppressed and both grafts were vigorously rejected with cessation of bile flow after four days and graft necrosis in 5 to 7 days. This was similar to our experience with auxiliary fetal liver transplants in the neck. In the next ten transplants we developed an immunosuppression regimen. We found the sheep to be very difficult to immunosuppress. The combination of azothioprine and steroids was completely ineffective with no prolongation of graft survival. In fact, doses of solumedrol in excess of 20 mg/kg/d were required to produce lymphopenia suggesting that the sheep is steroid resistant. We next tried the combination of cyclosporin A and solumedrol and found that cyclosporin (10 mg/kg/d IV) and solumedrol (20 mg/kg/d IV) would prevent acute rejection and allow prolonged graft function although septic complications, ie, cholangitis proved limiting. In addition, periportal infiltration by lymphocytes was evident by seven days after transplantation and all grafts ultimately had evidence of chronic rejection. Higher doses of cyclosporin resulted in renal toxicity. Using this regimen the grafts remained relatively normal histologically for the first week after transplantation and, in spite of chronic rejection and cholangitis, continued to excrete bile for up to 22 days after transplantation. This allowed us to meaningfully study graft excretory function during the first week and make some interesting observations thereafter. Bile Volume Studies Bile volume excretion by the fetal graft was compared in recipients with host liver excretory function obstructed by CBD ligation (N = 7) and those with host liver excretory function intact (N = 3; Fig 3).

CED LIGA'~ON

~

(N=7)

N0 CBD L[GA'rI01d(N=3)

M~A.N:~iSgM -

5

6

Post operative day

Fig 3. Comparison of the volume of bile excretion from the fetal graft in recipients with host liver biliary obstruction (CBD ligation) v recipients with host liver biliary drainage intact * P < .05.

recipients with host CBD ligation rapidly increased, and were maintained at significantly higher levels than those seen in recipients with intact excretory function (Fig 5). In contrast, bilirubin concentrations in bile from fetal grafts in animals with host liver excretory function intact remained very low for the first week and only began to increase after 7 to 10 days. Finally, maximal fetal graft bile bilirubin concentrations were considerably higher than concentrations from nonobstructed recipient livers or common duct cannulated controls (12 to 25 mg/dL). Bile Acid Excretion Total bile acid concentrations in serum and bile were measured in two recipients with host CBD ligation and compared with concentrations in two recipients with intact excretory function. Bile acid concentrations in fetal graft bile were significantly higher throughout the study period of CBD ligated recipients than in recipients with host excretory function intact (Fig 6). These concentrations were comparable to concentrations of bile acids in bile from CBD cannulated cholecystectomized adults (18 to 26 mmol/dL). In spite of these high rates of fetal graft bile acid Serum

totalbiJirubin {~/dl) 10 Y~J~N +_ I SE~

--~=y-

CONTROLS (lq=3)

• 1SI~M

Bilirubin Excretion Measurements of serum bilirubin were performed in six recipients with CBD ligation and compared with three control animals with CBD ligation and no auxiliary graft. The fetal graft maintained near normal serum bilirubin levels in the CBD ligated recipients, which were significantly lower than the control levels (Fig 4). Bilirubin concentrations in bile from fetal grafts in

U Z

3

5

7

Post operative dey

Fig 4. Serum total bilirubin levels after transplantation in recipients with host CBD ligation v CBD ligated controls (no auxi ary g r a f t } * P < .01.

562

FLAKE ET AL

eUe [B~h-ubi~] ( ~ z g / d l )

[Bile A c i d s ] (mMol)

100

90 '

80 I

CBD M G A ~ O N (N=3) M~AN• 1SEM

CBD LIGATION (N=2) MEAN + 1 SEM 9

NO CBD UGATION(N=~)

NO CBD LIGATION(N=2) ~EAN + 1 SEM

MEKN•

70

1SEM

ZO 5O

to ~ 2O lO

o

~3

5

7

Post operative day

Fig 5. Comparison of the bilirubin concentrations in bile from the fetal graft of recipients with host liver biliary obstruction (CBD ligation} v recipients with host liver bi|iary drainage intact. * P < .05.

secretion, however, serum bile acid concentrations were abnormally elevated in the CBD ligated group.

Incremental Defunctionalization Two recipients survived for more than five days after ligation of the hepatic artery and placement of a snare around the portal vein allowing incremental defunctionalization. One of the two underwent host CBD ligation at the time of transplantation. The CBD ligated recipient survived for nine days after surgery with completion of snare tightening five days postoperatively. At autopsy the host liver was completely necrotic. The other animal was our longest survivor (22 days). Its course was particularly instructive and is detailed in Fig 7. Death was due to host liver cholangitis. At autopsy on the 22nd day, the right lobe of the host liver was necrotic, the left lobe of the host liver was viable due to an accessory hepatic artery but was compromised by severe cholangitis. The weight of the graft had increased from 125 g at transplantation to 318 g (host liver 508 g). DISCUSSION

The need for donor organs has rapidly increased with the increasing application of liver transplantation. Recent developments in prenatal diagnosis have made consideration of the anencephalic fetus as an organ donor a realistic possibility. The anencephalic is the ideal donor because it has no potential for ultimate survival and has essentially normal organ developmerit. 9 Obstetric sonography and maternal alphafetoprotein screening can detect 90% of anencephalic fetuses prior to delivery. ~~ The incidence of anencephaly is approximately 1 in 1,200 live births, or about 2,000 anencephalics born per year in the United States alone. H The use of even a fraction of these organs for liver transplantation would meet the current demand for pediatric liver transplantation (approximately 600/yr). Currently all of these organs are wasted.

0

2

4

6

8

10

12

Postoperative day

Fig 6. Comparison of total bile acid concentration in bile from the fetal graft of recipients with host liver biliary obstruction {CBD ligation) v recipients with host liver biliary drainage intact (all differences significant, P < .01 ).

The use of fetal livers for transplantation raises several practical, physiologic, and ethical questions. The ethical questions have been addressed elsewhere by ourselves and others, n'13 The purpose of this work was to investigate the technical and physiologic feasibility of intraabdominal heterotopic fetal liver transplantation. Heterotopic transplantation has several theoretical advantages over orthotopic transplantation that are particularly relevant to fetal liver transplantation. First, it obviates the need for recipient hepatectomy, with its associated morbidity and mortality, technical difficulty, and blood loss. Second, preoperative host liver function would be preserved allowing time for maturation and growth of the immature fetal heterotopic graft. Finally, if the graft rejected or failed for technical reasons it could be removed without further compromise of the recipient. In spite of these advantages, clinical efforts at auxiliary liver transplantation using postnatal donor organs have been disappointingJ 4'~sThe primary problems have been technical, relating to size discrepancy as well as kinking of the vascular anastomoses in the absence of normal liver fixation. Our technical success in 31 of 38 transplants supports the technical feasibility of auxiliary fetal liver transplantation and suggests that use of the fetal liver may solve these problems. The small size of the fetal liver was easily accommodated by the abdominal space of the weanling lamb recipients, and fixation was achieved by the long caval to caval anastomosis. In addition, the use of the large thick-walled umbilical vein, rather than the small thin-walled portal vein, for portal inflow greatly simplified the procedure, and was less prone to kinking or compression. Although technically demanding, these procedures required 2 to 3 hours to perform and involved minimal blood loss for the recipient. Can the fetal liver completely support the functional requirements of the recipient? Under the conditions of our model, the answer is no. When the host liver was

AUXILIARY TRANSPLANTATION OF THE FETAL LIVER ~O z

POD 2 6 10 ~ HOST 15.3 20,3 18.5 < p_~Lu~ FETAL .8 3.4 1.4 Z ~ GRAFT LU ~ HOST 12.9 11.6 l 5.4 WWj ~

~ F~rAL GRAFT

~

13

3.3

563

15

16 12.5

1.4

17

19

21

21.0 17.5 19.1

18.2

5.9

18.3

19.1 ---]Total Bile Volume Fetal Graft Bile Volume

3O

15 -J

10 5 2 4 6 8101214

15

16 f Host Obstructed

17

1 Obstruction Relieved

19

i,_ i,_ ~ 20 l Host Obstructed

21

J 22 T Sacrifice

POST OPERATIVEDAY (POD) Fig 7. Course of 22-day survivor. This animal had hepatic a r t e r y ligation at the time of surgery and incremental tightening of the portal snare from postoperative d a y 2 to 10. The host common duct was drained for the first 15 days after transplantation. During this period the volume of graft excretion slowly increased to equal host liver volume excretion. However. bilirubin and bile acid excretion remained low. The host liver drainage catheter was obstructed between days 15 and 17, opened on day 18, and reobstructed on day 19. There w a s a dramatic increase in bilirubin and bile acid excretion after host fiver obstruction with an increase in both volume and concentration of bile excretion from the fetal g r a f t . With relief of obstruction, the volume decreased, however, the bile bilirubin concentration remained e l e v a t e d (bile acid data not available for day 18). On the final t w o days of life, fetal graft bile volume nearly equaled maximal combined previous bile output.

completely defunctionalized by portal vein, hepatic artery, and common duct ligation at the time of surgery, the recipients became encephalopathic and ultimately died, in spite of having a viable auxiliary graft. In addition, in other experiments in which we performed orthotopic fetal liver transplants, recipients died within 24 to 48 hours of transplantation, again, with apparent encephalopathy and viable grafts. 16 There are a number of possible explanations for the early functional inadequacy of the fetal liver grafts under these conditions: (1) the fetal graft was functionally too immature to support the host; (2) the graft may require time to functionally recover from cold ischemia; (3) in the case of heterotopic transplantation with host liver defunctionalization, toxic products from necrosis of the host liver may overwhelm the clearance capacity of the auxiliary graft; and (4) the recipient was approximately four times the size of the donor, thus the functional demand in our model would be far in excess of that normally encountered at birth. Whatever the reason, we feel that heterotopic transplantation of the fetal liver is not only theoretically advantageous, but probably necessary for success. In contrast to the fetal graft's functional inadequacy after complete host liver defunctionalization, partial host liver defunctionalization stimulated rapid increases in fetal graft excretory volume, bilirubin and bile acid excretion, and parenchymal growth. The fetal

liver graft was able to maintain near normal serum bilirubin levels in CBD ligated recipients and produced bile with higher bilirubin and bile acid concentrations than those of normal postnatal livers within days of transplantation. Even in the presence of a functional host liver, the fetal graft slowly increased its contribution to total excretory function. These results are encouraging and suggest that regardless of initial functional immaturity, if there is adequate host liver function present to temporarily support the host, maturation of the fetal graft will occur in response to functional demand. Elevations of serum bile acids were seen consistently in CBD ligated recipients even though in some cases graft excretion rates of bile acids were high. Possible explanations include one or more of the following: (1) Host CBD obstruction resulted in regurgitation of a portion of the bile acid pool, normally sequestered in the host liver, resulting in an increased bile acid load for the fetal graft; (2) Shunting of bile acids into the systemic circulation occurred through a patent ductus venosus with subsequent inadequate clearance of bile acids from the systemic circulation by the fetal graft; (3) Alteration of the normal enterohepatic circulation (ie, returning all bile in a bolus once a day) may have saturated clearance from the portal circulation (usually >90% on a single pass) resulting in a physiologic shunt of bile acids to the systemic circulation;

564

FLAKE ET AL

and/or (4) Small size and relative functional immaturity of the graft prevented clearance of a supraphysiologic load of bile acids. There is abundant experimental and clinical evidence documenting the functional immaturity of the fetal liver. 17'~8 Although there is considerable species variation, all mammals studied have deficiencies in hepatic synthetic and excretory function at birth. Placental exchange substitutes for many of the livers metabolic and homeostatic functions in utero allowing fetal liver development to occur in an environment of low functional demand. 19 After birth, the human liver undergoes a rapid maturational process. The factors responsible for inducing the maturation process in the perinatal period are unknown. Is it a genetically predetermined event or some factor in the postnatal environment? Is it simply a matter of functional demand? Our functional data suggests the latter. In recipients with CBD ligation, ie, an environment of high functional demand, the fetal liver underwent rapid maturation of excretory function. In recipients with host liver excretory function intact, ie, an environment of low functional demand, fetal liver excretory function remained immature for prolonged periods after transplantation. Interpretation of our functional data is complicated by several factors. First, the influence of chronic rejection and cholangitis on hepatocyte function is unknown but is presumably detrimental. Histologically, our grafts remained relatively normal for the first seven days after transplantation, and we have confined much of our data analysis to that interval. It is encouraging that these grafts continued to function, in some cases quite well, after histologic evidence of rejection and/or cholangitis appeared. Second, the host liver was not completely defunctionalized in CBD ligated recipients. The ability of the host liver to conjugate bile acids and bilirubin remained intact. Therefore, the adaptation of the fetal liver graft may simply reflect transport of an increasing load of conjugated bile acids and bilirubin rather than increasing synthesis. Finally, our alteration of the normal enterohepatic circulation, with replacement of bile in a single bolus each day was distinctly unphysiologic and may

have influenced function of the graft, and the host liver, in a variety of ways.2~ In this study, we chose to drain bile externally so that excretory function of the fetal graft could be studied. This undoubtedly decreased recipient survival because of the high rate of cholangitis associated with chronic catheterization and immunosuppression. Achievement of long-term survival in this model will require improvement of the immunosuppression regimen and development of a method of internal biliary drainage. Do our findings in the lamb model apply to human transplantation? In many respects fetal liver transplantation should be easier in the human than in the lamb. Clinical immunosuppression regimens are better defined and more effective. Human anatomy is more favorable with a longer portal vein segment and a more accessible vena cava. Physiologically, auxiliary fetal liver transplantation should be applicable to most of the current indications for liver transplantation. The most frequent indication for pediatric liver transplantation is biliary atresia. Children with biliary atresia maintain parenchymal liver function until the final stages of their disease. They have a primary excretory deficit that may be ideally treated by auxiliary transplantation. The auxiliary graft would have time to adapt to the functional requirements of the host and could slowly assume hepatic function as the host liver failed. Application of auxiliary fetal liver transplantation to more fulminant cases of hepatic failure may be less effective and will depend upon the adaptational capacity of human fetal livers. We conclude from these studies that (1) intraabdominal auxiliary transplantation of the fetal lamb liver is technically feasible and physiologically sound; (2) the auxiliary fetal liver graft is capable of rapid adaptation and can assume a significant portion of host excretory function; and (3) excretory function of the fetal graft is proportional to the functional demands of the host. Auxiliary transplantation of the fetal liver is a promising alternative to conventional methods of liver transplantation.

REFERENCES

1. Schmid R, BerwickDM, Combes B, et al: Liver Transplantation. National Institute of Health Consensus DevelopmentConference Summary, Vol 4, No. 7, 1983 2. Starzl TE, Iwatsuki S, Shaw BW: Analysisof liver transplantation. Hepatology4:475-495, 1984 3. Gartner JC, Zatelli BJ, Starzl TE: Orthotopicliver transplantation. Two year experiencewith 47 patients. Pediatrics 74:140-145, 1984 4. BahnsonHT, Starzl TE, Hakala TR, et al: Developmentand organization of a multiple organ transplant program. Ann Surg 203:620-625, 1986

5. Flake AW, LaBerge JM, Adzick NS, et al: Auxiliary transplantation of the fetal liver I: Development of a sheep model. J Pediatr Surg 21:515-520, 1986 6. Harrison MR, Jester JA, Ross NA: Correctionof congenital diaphragmatic hernia in utero I. The model: Intrathoracic balloon produces fatal pulmonaryhypoplasia.Surgery 88:174-182, 1980 7. Osuga T, Mitamura K, Mashige F, et al: Clin Chem Acta 75:81-90, 1977 8. Billing BH, Haslan R, Wald N: Methodology in bilirubin determinations. Ann Clin Biochem8:21-30, 1970 9. Naeye RL, Blanc WA: Organ and bodygrowth in anencepha-

AUXILIARY TRANSPLANTATION OF THE FETAL LIVER

ly. A quantitative and morphology study. Arch Pathol 91:140-147, 1971 10. Milunsky A, Alpert E, Neff RK, et al: Prenatal diagnosis of neural tube defects. IV. Maternal serum alpha-feto protein screening. Obstet Gynecol 55:60-66, 1980 11. Elwood JM, Elwood JH: International variation in the prevalence at birth of anencephalus in relation to maternal factors. Int J Epidemiol 11:132-137, 1982 12. Harrison MR: The anencephalic newborn as organ donor. Commentary. Hastings Center Report 16:21-22, 1986 13. Fletcher JC, Robertson JR, Harrison MR: Primates and anencephalics as sources for pediatric organ transplants. Fetal Therapy 1:150-164, 1986 14. Fortner JG, Yeh SDJ, Kim DK, et al: The case for and technique for heterotopic liver grafting. Trans Proc XI:269-275, 1975

565

15. Halgrimson CG, Marchioro TL, Faris TD, et al: Auxiliary liver transplantation: Effect of host liver portocaval shunt. Arch Surg 93:107-118, 1966 16. Flake AW, Harrison MR: Unpublished observations, May 1986 17. Lester R, Jackson BT, Smatlwood RA, et al: Fetal and neonatal hepatic fuction II. Birth Defects: Original Article Series. Vol XII, No. 2:307-315, 1976 18. Gartner LM, Lee KS, Vaisman S, et al: Development of bilirubin transport and metabolism in the newborn monkey. J Pediatr 90:513-531, 1977 19. Watkins JB: Placental transport: Bile acid conjugation and sulfation in the fetus. J Pediatr Gastroenterol Nutr 2:365-373, 1983 20. Strange RC: Hepatic bile flow. Physiol Rev 64:1055-1102, 1984