HEPATIC TRANSPLANTATION

Thomas E. Starzl and Thomas L. Marchioro

Reprint from Rapaport-Dausset "Human Transplantation" - Copyright 1968 Grune & Stratton, Inc.

HEPATIC TRANSPLANTATION

Thomas E. Starzl and Thomas L. Marchioro

Since the spring of 1963, a number of attempts have been made in several centers a t homotrans- plantation of the human liver. All eventually re- sulted in death of the patients. the maximum survival being five \\.eeks. These experiences and the ex- perimental background upon which the clinical trials \\.ere based have been exhaustively reviewed in ,,,,. l , ; , l s : ~ ~ . ~ ~ . ~ ~ and in a recent text." .-\nother sum-

1n;trv does not seem justified. Instead. an attempt will be made to identify

the factors which contributed to the failures. By and large these were the same whether the operations had involved total excision of the recipient's diseased liver and its replacement with a homograft (ortho- topic transplantation) or provision of a heterotopical- I\. placed homograft without recipient hepatectomy

.lllxiliary transplantation 1 . The extent to which these lrficiencies can be corrected will determine the -l'.ls~l~ility of such undertakines in the future.

Faults of Immunoeuppreesion

There is no reason to suppose that prevention of rejection is more difficult with liver than with renal or other kinds of homografts. Indeed. the converse is true. The results with orthotopic transplantation of the liver in dogs treated with azathioprine" have been as good or better than with kidney transplanta-

Supported hv USPHS grants AM 06283, HE 07735, . + I 04152. FR 00051, and FR 00064.

tion to comparably treated hosts. Several animals from a series of experiments conducted in 1964 are still alive from two and a half to three years after receipt of orthotopic homografts from nonrelated mongrel donors. Furthermore, a surprisingly large number of these dogs required azathioprine for only the early postoperative period. In Figure 1, the course is shown of an animal which was given azathioprine for the first four months. After cessation of all therapy, he has lived with completelv normal hepatic function for more than two and a half years. Biopsies after four. eight. and twelve postoperative months were all normal.

As with canine renal homotransplantation. the use of azic . y i n e alone permits really long-term sur- vival i ~ , - to 20 per cent of animals with orthotopic liver grafts. Obviously, effective clinical therapy can- not be provided with this single agent. Fortunately, the addition of prednisone. which has a synergistic action with azathioprine. has permitted kidney trans- plantation in man to b e carried out with moderate success. The same drug combination in recipients of liver homografts has proved to b e unacceptably dangerous.

In Denver. six orthotopic transplantations have been performed, with operative survival in five. Dur- ing their postoperative intervals of six to twenty- three days, all five patients developed some evidence of sepsis, including pulmonary infection in every case. The degree of immunosuppression which can often be tolerated after the less traumatic procedure

Fig. 1 . - Course of an animal which never had anv clinically evident homograft re- jection. Note rapid weight gain following cessation of therapy at four n~onths. The pronounced leukoc~tosis after withdrawal of immunosuppression was commonlv seen. This animal is still alive after two years and ten months. having received no immuno- dupprcssive therapy during the last two and a half vears of this time. (By permission of Tranuplantation. )

of renal homotransplantation was incompatible with life in the generally older. debilitated. and feeble patients \\.ho required the much more difficult opera- tion of li\,er replacement.

The same complications were encountered in the two patients who survived the operations of auxili- ~ r y liver homotransplantation at the University of Colorado. Both developed pneumonitis with pyogenic o r~anisms and died after t\venty-two and thirty- four postoperative days. Both also had infestation with ~ ~ n r ~ s u a l micl.oorganisms which usually have low p;lthogenicity. For example. the patient who lived for five weeks had invasive moniliasis of the lungs as well as most of the gastrointestinal tract. .\cute s ~ n ; ~ l l bowel ulceration caused by the fungus was responsible for severe gastrointestinal hemor- r l ~ a c e \vhich persisted for the last ten days of the patient's life.

These series of infectious complications have made it increasingly clear that the margin between desir- able immunosuppression and toxicity is too fine to permit consistent success after clinical homotrans- plantation of the liver with r : therapeutic regimens ~ l sed in the past.

Prospects for Improving Immunosuppression

The most promising new agent in recent years has been heterologous antilvmphocvte serum i ALS) or its globulin derivative (.\LG). Accounts of the experimental background of these biological prod- ucts (Chap. 32). and of their first testing in man (Chap. 31 ) are contained elsewhere in this \.nlume. In our laboratories both ALS and .4i ; 11 . been shown unequivocal l~ to prolong the i l c .. AI life of liver homografts.

.\ total of eighteen dogs were treated. nine with .\LS and nine with ALG. Those receiving the serum had intraperitoneal injections. Therapy was generally started in advance of operation and continued for only a few weeks postoperatively (Fig. 9). The material used had a relatively low hemagglutinating titer of 1 :64 to 1:%6 when tested against dog leukocvtes.

Subsequently, much stronger antisera were raised in the same horses by increasing the doses of im- munizing antigen. With this change it became pos- sible to extract high potency globulin from t ' orse serum. To date the most practical means e - ' . .>ing this has been with the technique of ammonium

\ears ILS )

< ~ t the prod-

1 1 man olume.

t % been 1.11 life

I , t , \ v i t h \

\erum .~~era l ly I ~ J for

: ~natinq ,t dog

raised of im-

ne pos- ~t. horse I doing ~ ~ o n i u m

CONTROLS

HEPATIC TRANSPL-SNTATION

Fig. 2. - A rhronologicallv surviving dog which was treated before and for twerltv dnvs after orthotopic liver transplantation with intraperitoneal antilyn~phoid serum (.ALS). Note the pronounced Iymphocvtosis late in the postperative period. The ani- mal is in excellent health after ten months. ( B y permission of Surg. Cynec. Obstet.)

ORGAN TRANSPLAS'r..\TIOS

ELECTROPHORESIS

0.40 SAT.

IMMUNOELECTROPHORESIS

0. 5 0 SAT.

Fig. 3. - Electrophoresis and immunoelectrophoresis of the horee protein obtained by precipitating four times at different saturations of ammonium sulphate. Note the progressively heterogeneous nature of the precipitate with higher ammonium sulphate concentrations. The globulin obtained with 0.4 saturation was used clinically. (By per- mission of Surg. Gynec. Obstet.)

l ~ ~ . "IC TRANSPLANTATION 219

.ulphate prec~p~tat ion. Three different products are ,lrown i l l Ficure 3, obtained with precipitation a t :3.5 satrlratlon. .-lo saturation. and .5 saturation. For

testing. the clobulin used was generally that pre- p r e d w ~ t h .4 saturation, since there was a maximum r<>tent io~~ ot ; ~ ~ ~ t ~ - \ ; . h i t e cell antibodies \vith a mini- rnrlrn ot ~ . o ~ l t a m ~ n a t i o n with extraneous horse pro- tcqn. Tile t ~ t e r of the recwnstituted globulin was I ..512 to I : 10%.

Sine :l~rlrnais received orthotopic transplantation :vhile rlntier treatment with subcutaneous injections ot the Ilorse qlobulin. The results in these animals .Ire shown 111 Fic. 4, and compared to those obtained \vith .4LS therapy or in untreated animals. A defi- 111te potet~t~:ltion of survival \vas observed both ~v i th .4LS .ind .\LC. The survival fiqures indicated

~ ~ ~ 0 - - - - ~ (9 ) 23 DAYS 3 6 DAYS

(CONTROLS 7 t 3 S.D.) F i g . 4. - Effect of ALS (serum) and ALC (plobu-

l in) upon - ~ ~ ~ . \ i \ i l l after orthotopic liver transplanta- tion in t i t ~ g a . 'l'lie mean values were computed with litnitation of .urvival credit for anv dog of seventv

rlavp. . number of the treated animals had longer -urvival than this.

were computed with a maximum survival credit of seventv days for any dog. With this statistical ceil- ing, the mean survival was approximately one monrn in the combined groups. In actuality, however, four of the eighteen treated dogs lived for at least four months. Animals receiving no therapy died in 7.0 3 (S.D.) days.

Several siqnificant observations were made in these .~nimals. One dog treated with ALS received therapy ollly prior to operation (Fig. 31. The l!.mphopenic response was exceedingly modest. Afterwards, there never was any evidence of rejection. The animal lived for more than six months and finally died as a result of a midgut volvulus.

The course of another dog is shown in Figure 2. This dog received intraperitoneal ALS prior to orthotopic transplantation for onlv the first three postoperative weeks. Here. also. the lvmphopenic response was not significant, and after cessation of therapy there was a progressive lvmphocvtosis which has lasted for several months without late deterioration of homograft function. The dog is still alive. more than ten months after homotransplanta- tion.

In the foregoing experiments, ALS or ALG was used as the sole immunosuppressive therapy. .I limited number of additional experiments have also

I

- - - I - 0 1 0 2 0 30 40 SO 60 70 80 90 I00 110 I20 I30 T I M E IN DAYS

Fig. 5. - A dog that received an orthotopic liver homograft after six intraperitoneal injections of antilvmphoid serum ( A L S ) . No postoperative therapy was ever given. Note that the lymphocyte count was little changed. The dog died o f intestinal obstruction six months postoperatively. (By permission of Surg. Gyncr. Obstet.)

ORGAN TRANSPL.ISThTION

Fig. 6.- Reconstructivn after

orthotopic liver homotrans- plantation in the do?. In-

ternal biliarv drainage is

~ i t h 3 rholecvstoduodenvs- tomv. Aorta is transplanted in

vontinuitv with the hepatic nrterv of the homograft. (Uv permi5aion of Surgerv. )

M i c anastornos 1s / y 4 \\

caval

been carried out using ALG in combination with azathioprine. or with azathioprine and a short course of prednisone. Using either schedule of multiple drug therapy. there was improved immunosuppression in- tlicatine a synerqistic effect of the different aqents. .\ppros~matelv 70 per cent of animals provided with triple tirug therapy have achieved survival of at least fort!. tla!,s. excluding operative deaths.

.\ previously uncommon technical complication was seen with increased frequencv with the better immunosuppression of such regimens. Since the be- ginning of our investigations of orthotopic trans- plantation. the procedure used has been that shown in Fig. 6. In the donor, a segment of the aorta was removed in continuity with the hepatic artery. It was then possible to anastomose the end of the homografted aorta to the side of the recipient's mid- .tbdominal aorta. Although this method of arterializa- tion has hemodynamic disadvantages, clotting in the arterial svstem was rare.

iVith the improved earlv function resulting from the new programs of immunosuppression, a verv high incidence of thrombosis of the aortic graft was en- countered, killing more than two-thirds of the ani- mals. \lore recentlv, an alternative procedure has been used. A right nephrectomv was performed in the recipient. The celiac axis of the homograft was then attached by an endl to lend anastomosis to the right renal artery. The incidence of intra-arterial thrombosis was therebv reduced to approximately 15 per cent.

Only one attempt a t human liver homotransplanta- tion has been made with the use of ALC. This was in a 28 vear old man with a hepatoma and represented the sixth case of orthotopic transplantation in Denver. The homograft was obtained from a 73-year old pa- tient who died of a cerebrovascular accident. The donor was accepted in spite of his advanced age be- cause an unusual degree of antigen compatabilitv with the recipient was demonstrated by van Rood's

om the \. high :';is en- :le ani- :re has .ned in ,I:' \vas to the irterial

,rnatelv

Ienver. )Id pa-

it. The ige be- . ta bility

Rood's

HEPATIC TRANSPLANTATION 22 1

, I I ~ Amos' tissue-typing r ( ::ns which were working :unimals, emphasize the important principle that re-

111 Denver ' ~ t that time. ,. period from death to jection is a phenomenon which tends to be spon-

revascularization was 99 minutes. Although this in- taneously reversible. It also emphasizes the caution terval is acceptable in canine experiments, there had that is necessary in attributing benefit to other ther- been serlorls injury to the homograft. l l u c h of the apeutic maneuvers carried out a t this critical time. damace to the donor liver had apparently occurred Sluch more effort has been made to determine in the pre-rnortem stages. The recipient patient did l~istocompatibilitv antigens in man than in dogs. Un- not achieve coot1 earlv homograft function and died fortunately, the location, number. and nature of hrl- seven cla1.s /titer ot hepatic insufficiency.

In spite of this disappointing experience, the evi- tlence is strong, ns reviewed in Chapter 31. that .\LC will make possible improved immunosuppres- sion for homotransplantation of human tissues and organs. This clinical evidence has been obtained with the study ot recipients of renal homografts. but there is no reason to think that the experience is not trans- ferable to problems of the liver.

Histocompatibility

The desirability of improving immunosuppressive therapy is qenerullv acknowledged. .In alternative approach worlld be to reduce the need for immuno- suppression by application of various techniques of human histocompatabilitv analysis. Experience with both canine iund human organ homotransplantation have shown beyond doubt that certain nonrelated members in either outbred population can provide tissues tvhich elicit a relatively mild immunologic reaction. and conversely that certain members are favored recipients. In dogs, for example. 1.3 to 20 per cent ot recipients which receive organs from randomly selected mongrel donors have a very be- nign postoperative course under therapy with aza- thioprine illone. often never manifesting overt evi- dence of homograft rejection. An example of such a fortunate animal has been previously alluded to ( Fig. 1 1 . .It the other end of the spectrum. involv- ing approximatelv one-third of the animals, rejection is onlv delayed for a few days or weeks. In these clogs. relection once begun (Fig. 7 ) is inexorable. The histoloqic findings in the homografts from these llnfavorable experiments a r r indistinguishable from those in the untreated anim'i,

Between the two extre11ic.r are to be found the results in approximately one-half the experiments. In these recipient animals, rejection is seen. often to a severe degree. but it is more or less reversible. It is important to emphasize that no supplementary ther- apy was instituted in these animals at the time of their rejection crises. Treatment with azathioprine was continued in approximately the same doses as i~cfore with ultimate improvement in liver function. The latter observations, made in more than forty

rnan histocompatability antigens are inco~npletelv understood. Such information may come from the investigations of human isoimmune antisera discussed in Chapter 24. Such sera have been obtained from patients who were accidentally or deliberately sen- sitized to homologous whitecell antigens or other tissues. The agglutination or cvtolysis of test lym- phocytes by these antisera imply the presence of the same or similar antigens as those which originally sensitized the donor; tailure of such reactions implies the absence of the antigens.

There are indications from a number of groups that the antigens detected by these serologic methods ;Ire a t least related to histocompatibility factors. The most convincing evidence has come from the patho- logic studies of Porter,'* who examined biopsies ob- tained from more than forty patients approximatelv two years after renal homotransplantation. There was a high degree of correlation between the quality of preservation of these chronically tolerated homo- grafts and the completeness of antigen matching be- tween the recipients and their respective donors. These findings suggested that the long-term fate of organ homografts was influenced by the degree of donor-recipient white-cell antigen conformity.

The foregoing studies supported the concept that improvement in whole organ transplantation could be achieved by prospective antigen matching. Un- fortunately, efforts to identify the biologically more suitable donors by this technique have proved some- what disappointinq.13 An increased survival was ob- tained in a series of patients provided with renal homografts from nonrelated donors at the University of Colorado, but the one-year mortality was still ap- proximately 50 per cent. The results with intra- familial renal homotransplantation were not improved at all. It has become increasingly clear that the high early mortality in such cases t a n only be slightly modified by the application of antigen matching techniques, and that the most urgent necd is for changed immunosuppressive protocols as discusscd earlier.

Organ Preservation

More effective means of either immunosuppres- sion or histocompatability analysis will not insure the

ORGAN TRANSPLASThTION

I- I

SG P T (S-F Units -----------A 70

ALKALINE PHOSPHATASE 10

(Bodansky Units)

-5 0 5 10 15 2 0 25 30 35 PREiOP I POSZ- OP TIME IN DAYS

Fig. 7. - Example of inesorable rejection despite immunosuppressive therapy. Serum bilirubin was the most useful measurement for following the course after homo- transplantation. since the other abnormalities of liver function depicted can also be caused by azathioprine. ( B y permission of Surgery.)

HEPATIC TRANSPLANTATION 223

SUCC.e5S ~t homotransplantation of the liver. The sen- ties were probably due to anoxic injury rather than sitivitv ot this orcan to anoxla has created a serious to rejection. Although the injury was partially rever- problem ill virtually every clinical trial to date. In- sible in many instances, the quality of subsequent variablv. there has been early postoperative ~vidence homograft function ranged from mediocre to poor,

iscllemic lniurv to hepatic parench\.:,..l \, th early contributing to the ultimately unfavorable outcome. high increases in SGOT, SGPT. and LL)H Fig. 8). The need for haste has led to the development of \vithin .A tew clays such patients have developed compromise clinical procedures, including staging. jaundice. These varlous early biochemical abnormali- .it a first operation, all structures entering and leav-

ing the diseased liver were skeletonized (Fig. 9 ) . With the subsequent availability of a suitable donor - SGOT (S-F UNITS) candidate. it was then possible to rapidly reopen the

. SGPT (S-F UNITS) incision and to quickly remove the patient's diseased - LDH (BB UNITS) liver. This approach is not without disadvantages. Recipients who are candidates for such operations are generally suffering from a terminal illness and can ill afford two lengthy and traumatic procedures. .Cloreover, the timing of the second stage is unpre- dictable since it is never known in advance when a donor may be expected to arrive. In one of the cases treated in Denver, there was a delay of two weeks between the first and second operation.

Methods were needed which would permit con- servation of hepatic tissue for ten hours or longer. Until recently, the simplest and most effective way of ameliorating the effects of ischemia were by sim- ple perfusion of the homografts with cold electrolyte solution (Fig. 10). In dogs, a liver so treated could be transplanted successfully after an ischemic period of two hours with a high expectation of success.

Fig. 8. - Serious injury to human liver homograft (Case 2 Colorado Series). Time from death to re-

Beyond this time, however, there was an exponential-

~~cularization was 152 minutes. during 98 minutes lv increasing number of failures due to poor initial

:i which extracorporeal perfusion was carried out. homograft function. Yote the high but reversible increases in serum en. Methods of continuously perfusing the cadaver ryme values. (By permission of Surg. Cymec. Obstet.) liver permitted only a slight extension of the accept-

able postmortem time, in spite of the fact that ele- gant svstems of perfusing dog, pig, and cow livers for other purposes had been previously de- ~cr ibed.*-~.~ ' The first perfusion system which was used for preservation of liver homografts emploved a heart-lung machine which was attached via trans- femoral cannulas to the great vessels of the recently

~ ~ e v l o u ~ deceased cadaver.lO By in(.orporating a heat ex- changer in the circuit. both . - !rfusion and hvpother- mia could be used simultaneously (Fig. 11). In dogs. this system permitted functional kidneys to be con- sistently obtained as long as twelve hours after death. but there was little if any demonstrable protection of the canine liver homograft. Alternatively, Mikaeloff and Kestens, using a normothermic system of isolated organ perfusion, were able to extend the acceptable

Fia. 9.- Skeletonization of diseased recipient liver postmortem time six at a first stage. Subsequently, the patient can be The foregoing studies, as well as more recent in- rapidly re-explored and the liver removed. (By per- vestigations by Brown and McDermott and their as- t~lirrion of Sorp. Gvnee. Obstet.) s~c ia t e s ,~ provided little hope that immediately func-

ORGAN TRANSF

( Bottle ) l blood 1

Fig. 10. - Method of further cooling a liver homo- graft just before its removal. Donor animals are operated on with total bodv hypothermia of 29 to 31 C. Cold lactated Ringer's solution is infused thr rgh the portal vein at the same time the donor animal is exsanguinated. ( B y permission of Sure. Gynee. Obstet. )

tioning and life-sustaining livers could be preserved even for intermediate intervals. The results recently obtained bv Dr. Lawrence Brettschneider, working in our laboratories. were therefore both welcome and somewhat surprising. Five cadaver canine livers were stored using a modification of the technique developed for kidneys by Ackerman and Bamard1-a combination of low-flow hemodilution perfusion, hypothermia of 4" C., and hyperbaric oxvgenation at 3 to 4 atmospheres. The p H was carefullv controlled at 7 .3 .

.ifter ten to fifteen hours, the livers were removed from the chamber and used for orthotopic trans- plantation to nonrelated recipients. Excellent func- tion was evident. The animals awoke and had a postoperative convalescence indistinguishable from that of animals receiving grafts under optimal con- ditions. All five dogs lived for at least three days. These de~:,lopments have made it clear that a method n l th immediate clinical applicability is al-

#

'LAXTATION

~.caatIv ; ~ \ ; l i i : i l ) l c ~ . .I . l lr . certainty of being able to hold i l I I \ ( Y Iior~locl.atts tor mnnv hours 11 solved

0 1 t 1 1 ( , ~iloqt (liffi't~lt . I I I ~ practical p; .ems ot , i ~ ~ i c ; ~ i I ~ o ~ l ~ o t ~ . ; u ~ s p l ; ~ ~ ~ t : ~ t i ~ ~ ~ i .

Tet111~0rary Hepatic ? t i ,port

Experience \vith clinical liver trn~~splantation has sliown t11;it temporarily life-s~istaining tunction can be expected even from verv seriouslv damaged liomo- grafts. There is not an all-or-none hepatic response to ischemic iniurv :IS is often the case with cadaver kidneys, which mav pass through a totally nonfunc- tional state of acute tubular necrosis.

Nevertheless. an important impediment to suc- cessful llepatic transplantation is the fact that an artificial liver for temporary assistance during a peri- otl of homograft recovery is not iivailable. The de- \.elopment ot such an artificial liver in the immediate future appears unlikely clue to the complex and in- completely understood function of this organ.

The brilliant investiqations of Eiseman. Liem, and Raffucci.; extended bv N o ~ m a n et 31.'; and other in- ves t iga tors .~~~" have explored the alternative possibil- ity of u t i l i z in~ extracorporeal homologous or heterol-

.ogous livers temporarily revascularized in parallel with the recipient's own circulation. The principal tlisadvantages of this approach are the short-term benefits which can be expected and the relatively complicated instrumentation which is required.

Sevel.theless. this form ot therapy, either for pre- operari\.e resuscitaL:rln of the critically i : : y t i e n t or as a tc :nporarv asslsi p r o c e d ~ ~ r e during recovery from ischemic i n j ~ ~ r y or severe rejection. is the onlv means now available for the post-transplant period. Eiseman and Norman have reported definite but brief benefit in patients dying of chronic liver disease. It could be ar:r,,pd that the method had not received a fair t: I . . . .ince the nature of the original illness precluded e\ ual recovery.

ri further application of this concept has been tested in our laboratory and in two patients." In the animals, total hepatectomv was carried out. A num- ber of hours later, homografts from nonrelated ani- mals were revascularized in the cervical area, ob- taining the arterial inflow from the carotid artery and directing the hepatic venous outflow through the external jugular vein. Such heterotopicallv placed grafts secreted bile and permitted a doubling of the expected survival. Detoxification of the barbiturate anesthetic was proven by the frequent necessity for re-anesthesization.

Two patients dying of acute or subacute hepatic failure have been treated with this method a 16

, ~ n d in- n e d :I sBm. and a

~ t h e r In- 9

!>osslbil- *. lleterol-

parallel ~rlnclpal art-term tzlativelv ( I

itlent or

\ means

Fig. 11. - Technique of extracorporeal canine cadaver perfusion. The aorta and inferior vena cava are cannulated via the femoral vessels as soon as possible after death. Heparin and procaine are added to the glucose or electrolvte solution used to prime the extracorporeal circuit. Antieoapulation of the corpse occurs with the first surge of the pump. The heat exchanger permits rapid cooling. An identical technique has been used for humans. (By permission of Surgery.)

ORGAN TRANSPLANTATION

Fig. 12. - Technique of extracorporeal liver homotransplantation used in two patients. The superficial femoral artery was anastomosed to the hepatic artew, and the hepatic venous outflow was directed into the femoral system. The liver was left in situ for three and a half days. ( B y permission of Amer. J. Surg.)

HEPATIC TRANSPLANTATION 207

\car old bov, and a 7 year old girl. The livers were :,t:tilied trom a cadaveric donor in the first case. and blood-type compatible chimpanzee in the second.

Thev \\-ere revasculnrized in the groin (Fig. 13). There was a prompt fall in the hyperbilirubinemia in I)oth patients b11t with little evidence of protein ,!.r~thesls. The estrncorporeal livers were left in place tor three :l11(1 ;I half and one davs.

Yeither ot the patients aroused from the pre- f \isking I~epatic coma. although this appeared to be

chtened in the case of the vouns girl, who sub- ,t.clr~entlv awoke completely several davs after the l i ~ ~ e r was removed. The first patient died three and n half days after insertion of the graft, and the second cllild died of a massive gastrointestinal hemorrhnee fourteen davs after its removal.

The clillical benefit in these two cases was dis- ,~ppointinc. It seems likely that assist procedures \\.111 be necessary for a longer period if benefit is to Ile derivecl either before or in the earlv period after i iomotransplantation.

Auxiliary Liver Transplantation

\lost of the foregoing remarks have been based on esperience with orthotopic transplantation. In this situa.tion in which the sole liver tissue is represented in the homograft. there can be no question of the qource of either earlv or late hepatic function. Con- sequently. the most incisive information concern-

ing the cli1:~litv of organ preservation. the efficier~r of immunosuppression. ' t' tiistopathologic events rejection. iuid ,riel . other questions can be be obtained I\ . + I is o\ :imental preparation.

Severtheless. .Ludieb . I auxiliarv liver trunsplan- tation l~ilve unfolcled picture tullv as interesting as. .lnd quite different from. that \vhich Iias evolved from studies of the orthotopic operation. The follow- ing remarks will review the evidence on certain con- troversial observations which have been made in dogs which were provided with a second ubt~ormallv placed liver.

This procedure. as first described bv CVelch.:!' in- volved placement of the homograft in the right para- vertebral gutter (Fig. 13A) with rearterialization from the terminal aorta or iliac artery. Inflow to the portal vein was provided from the distal transected inferior vena cnva. Subsequentlv. it was found in :~nimals treated with azathioprine that such homo- grafts were profoux~dlv afflicted with atrophy." The shrinkage usually began within two or three weeks. and reached an advanced stage within forty-five to sistv davs ( Fig. 14 ) .

It was soon found that the homograft atrophy could be prevented by changing the method of revascularization. If nonhepatic splanchnic blood were diverted into the auxiliary liver i Fig. 13B), its shrinkage was averted and the atrophy now af- fected the host's own liver. TI:< findings sr~ggested that there \vas competition bet\r.een the two coexist-

Fig. 13. - Auxiliarv lker transplantation. A. method of

Welch. hote that portal 5rnous inflow is from the inferior vena

cava. The homograft undergoes rapid atroph*. B. rnodifiration

o f Welch method in whirh non- llrpatie 3planchnic flow is c l i - ~ e r t e d through the homograft. With this preparation. the honio- graft retains its size and the ani- mal's own li5er shrinks. It is usuallv more convenient to bring the hepatic arterv behind rather than in front of the portal 5ein as depicted. (By permission of Surg. Gynec. Obstet.)

ORGAN TRANSPLANTATION

Fig. 14. - Results with auxiliary liver transplantation when revascularized bv Welch method (Fig. 1 3 A ) . Note marked atrophy of the homograft ( r igh t ) and no change in the animal's own liver (left). General morphology of the homotransplant is quite rrcognizable. The two specimens were obtained f o r t ~ f i v e davs after auxiliarv trans- plantation.

Fi*. 15. - Technique of partial portacaval transposition. One part of the liver rvreives svstemic venous blood via the inferior vena cava while the other portion is *upplied with splanchnic venoms blood. (By permission of Surgew.)

(-e~!raucy qlro~ jo s3!u!y3 s!rls!pad jo uo!ss!uc~ad ig) -~alln% ~s~qa~ra~ereci

aqS!d aqa u! ra*!l .i~e!l!xne aql jo uo!lez!relnsseaan -:) .aPeqrromaq laaa!rea JO 10~1uo.b

roj iasssaaau aq illensn [I!M s!q~ 'pdUI~0j~ad LZ! S~~OUJOI~EUE ~EABJEI~O~ aPe1fi is~y .\ 'E!sOqJJ!J s.Jauuae7 fEu!ucral ql!w s~ua!~ed u! ralua3 IsJ!pa&q operolo-~ jo i~!sraa!u~ aql le aa!nl pasn uaaq seq qqqm uo!~aius~dsue~~ ra.t!l irs!l!xne fo poqaa1i - '91 .;I!.g

230 OHGAN TRASSPLAST.-\TI~\

i l l < livers tor some substlate or substrates present i ~ ~ c l r ~ t l i ~ ~ l r \ 1 1 1 r l c ~ ! ) I - ( . \ ~olrslv ur~published ol~~prvat ions, i l l selective cotrcentrntion iri portal blood. T h e orcan 11.1s I ) ( Y . I I ( ~ t c s t i . I K I I ~ S ilope that antigen r i h i n g or which hat1 first exposure to this intestinal effluent h e p ; ~ t ~ c \rrp~>ort p r o ( w l ~ r r e ~ w ~ i i e of im. . tance in operated at ;I metitbolic a t lvan t ; t~e .~ l permittitle . ~ t I(.irst c~i~rl\ . srlrvl-. , I ;~ f te r hver trans-

The above experiments sr~ffered from the imper- - plantatio~l \c,rlns less \veil Countied. fections that the homograft \vus rrnder immunoloeic ,ittacli \vl~ereus t l ~ e ~~utoloqous liver was not. and tililt tlre host was bei~iq treated wit11 ;~zathioprine. References

. I 11ep;~tosic tlrt~q. Tlre purer experiment shown in Fiqure 1.5 was theretore des ig~led .~ ' In this prepara- tion. a traction of the animal's own liver was supplied wit11 systemic blood by means of an anastomosis of the inferior vena cava to one of the main portal I~ranches wllile the other portion was perfused with splanchnic venous inflow. Blood flow \vns equivalent in the two hepatic components. Atrophy invariably ensued in the ab~iormally vnscularized fraction with compensatory hypertrophy in the other part. The observations could not be supported in a preliminary ~.eport by IVelborn. L~uiier. and Foster." but were sr~bsequently confirmed in principle by the studies of Price and his as~ociates.~: '

' ,ese studies indicate, of course, only that the tle.:L~ibed competitive relationship can result in "starvation" of the less favorably disposed organ. Re- cent studies have shown that if the abilitv of the host liver to compete is reduced by a variety of methods, the atrophy of a heterotopically placed homograft can b e partially or even nearlv com- pletelv avoided. Such procedures include autol- ogous hepatectomy.'~ ' r a t i o n of the common duct,+l.;, I(;.;. arid consr.-.,.tion of a host Eck fis- tula,t;...s.~. Such observatior~s. have clinical applica-

bilitv since the livers of patients being considered for this kind of therapy have failed and utould presum- 'iblv be incapable of aggressive metabolite extraction.

The kind of compromise clinical procedure which has evolved from these animal investigations is shown in Figure 16. Two patients, both suffering trom Laennec's cirrhosis. have received such opera- tions in Denver from one to four days after emer- gelicy portacaval shunt for v~lriceal hemorrhage. As recorrntecl earlier, both recipients died of septic com- plications after twenty-two and thirtv-four days, re- spectively. However, there was unequivocal evidence of homograft function. There was no evidence of homograft atrophy in either instance.

Summary

Successful homotmnsplantation of the human liver depends primarily upon further refinements in im- mrrnos~~ppressive therapy and in organ preservation. Recent substantial nrorrcss toward both objectives,

1. .icker111;111. J . I\. \V.. '~nd Barnard. C. S.: Srtccehh- f r t l storage 01 kidneys. Brit. J. Surg. 33:323, 1966.

2. Bihler. 1). D.. Contion. R. E.. and Syhrls. L. \I.: Effect of heteroloqous liver perfusion on dogs in esperr- mental hepatic insufficiency. Surg. Forum. 16:286. 1065.

.3. Brown. H.. Patel, J.. Blair. D. IV.. Barsaniian, E. Sl., and \lcDermott, \V. \'.. Jr.: .\ method for cold preservation of the liver. J. Sttrq. Res. 5:71. 1967. 1. Eiseman. B., Knipe, P.. SlcCall, H., and Orloff. 51.:

Iwlated liver periusion tor redt~clnq 1)lood clmnionia. .\rch. Srtrg. 83:363, 1961.

'3. Eiseman. B.. Liem. D. S.. and Rntfucci. F.: Heter- ologous liver pc.rfr~sion in treatment of hepatic failure. .inn. Surg. l62:329, 196.5.

G. Faris, T. D., Dickhaus. -1. J. , llarchioro, T. L., and Starzl, T. E.: Liver radioisotope scannlng in auxil- ictrv hepatic homoerafts. Srtrr.. Gynec. Oilstet. 123: 1261. 1966.

7 . Hacihar;~. P.. .uid .il~solon. 6 B.: Experimental studies on homolocrous heterotop~c ..,er transplantation. Srtrg. Gynec. Ohstet. 119: 1297. 1964. S. Halgrimhon. C. (;.. \l.~rchioro. T. L.. Fnris, T. D.,

I'orter, K. .i.. Peter\. G. 1.. and Starzl, T. E.: Clinical and r\-perltnentiri .~rrsrliarv hepatic hor~~otransplantation \vith ernphi~~is on ho\t portacaval shunt. .iSlA Arch. Srrrg. $13: 107. 1966.

9. Kestens. P. J. , .\usten. \V. C., and SlcDermott, IV. \'., Jr.: Biocl~emical and phvsiologic studies on the vial)le extmcorporeul liver. Surq. Fon~rn. 10:225, 1959.

10. Slarchioro, T. L., Huntley, R. T., LVaddell, \V. R., and Starzl, T. E.: The rrse of extracorporeal perfusion for obtaining postmortem grafts. Surgery. 34:900, 1963.

11. llarchioro. T. L., Porter, K. A,, Dickinson. T. C.. Faris, T. D.. and Starzl, T. E.: Phvsioloeic requirements for ;~rrxiliary liver homotransplantation. Surg. Gynec. Obstet. 1 9 : 17, 1965.

1.7. Slarchioro. T. L., Porter, K. A., Illinworth, B. I.. Faris, T. D., Herrmann, T. J., Sudweeks. A., and Starzl, T. E.: The sprv.*+,- influence of nonhepatic splanchnic venous blood Ho, I I ,In the liver. Sure. Forum. 16:280, 1965.

13. Slurchioro: T. L., Terasaki, P. I., Hutchison. D. E., Brettschneider, L.. Cerilli, G. J., Groth. C. G., and Starzl. T. E.: Renal transplantation at the University of Colorado. Transplantation. 6: July, 1967.

14. Itartinis, A. J., Goldsworthv, P. D., Jones. T. W., Nyht~s, L. \I., DeVito, R. V., Volwiler, W., and Harkins, H. N.: Studies of hepatic physioloqy in the isolated per- fused calf liver. Srlrg. Forum. 9:489, 1958.

HEPATIC TRANSPLANTATION "1

. 4 (.. 3-,.-'

Iirnonia.

23:1261, .

.rimental

.lntation.

\. T. D., Clinical

I .lntation i .\rch.

Dermott, r on the 19.59.

I. W. R., ?erfusion

I. 1963. 11, T. C.. lirements

Cynec.

lth, B. I., ~d Starzl. ,)lanchnic

t . 16:280,

LS. SlikileloB. P., Kestens, P. J., and Dureau, C.: . I . ~ ; I ~ S ~ ) ~ : I I I ~ ; L ~ I ~ I I orthotopique du Foie chez le chien apres co~~wrvatlori < I ( ' I'orgnne par perfusion. hlem. .\cad. (:11ir. 91 :71 1. lS)(j.5.

16. \lit(). \ I . . .ickrovd. F . W., Covelli. V. H.. Eyskens. I < . . K . I ~ ~ ~ ; I I I I ; I . I . . .uld SlcDermott, W. V., Jr.: Partial Ilt,tcrotol)~c I~\ctr I~ounocrnft in dogs utilizing portal ar- t c r ~ ~ t l l z i i t ~ o ~ ~ . . i n n . Sure. 165:10. 1967.

17. S I ~ ~ I I I ~ ~ ~ I . I . C.. Si~mvis, C. A.. Brown. 11. E.. and \ r t t \ , r : Iinn~unochemical observations 1 1 clinical porclite li\,er perfusions. Surgery. 60:179, ; [Kj6.

13. Portrr. Ii. .i.. Rendall, J. 51.. Stolinski, C.. Tera- ..rki. P. I . . \larchioro, T. 1 ,., and Starzl, T. E.: Light ,111d clectron nllcrozraphic shidy of biopsies from 33 I I I I I I ~ ; I ~ rc.ni11 allografts and an isograft I?; to ?!i years .~fter transplantation. Ann. S. Y. Acad. Sci. 129:615, 1 986.

10. I'ricca, J . H.. Jr.. \.oorhees. .\. B., Jr.. and Britton, 11. C . . T h r role of portal blood in receneration and t l ~ r ~ c t ~ o n ot con~pletely revascularized partial hepatic .~utocr;~tt>. Srlrzery. 62: July, 1967.

30. Starzi. T. E.. Slarchioro, T. L.. and Faris, T. D.: '1 . i tc .r tr,~nsl)l,~nt;~tion." Editorial. .Ann. Intern. \led. O4:47O. 1966.

11. St.~rzl. T. E.. .\larchioro, T. L.. Faris, T. D., hlc- C,irdlc. 11. J . . .ind Itvasaki. Y.: .\venues of future re- \ c ;~ rc l~ in I~omotransplantation of the liver: With partic- c11,1r r<.ference to Ilepatlc supportive procedures, anti- I!inl)l~ocyte wrrlnl and tissue typ~nq. hmer. J. Surg. 112:391. 1966.

22. Starzl. T . E.. Slarchioro. T. L., and Porter. 6. A.: Progress 111 homotransplantation of the liver. I n Welch, C . i Etl. I : .\cl\.ances In S~~rt lery . Chicago, Illinois, Year

Book Sledical Publishers, Inc., 1966, pp. 295-370. 23. Starzl. T. E., Marchioro, T. L., Porter. K. .I..

Tuylor. P. D., Fi~ris. T. D., Herrmann, T. J., HLd. C. J., and Waddell, LV. R.: Factors detemlining short- and long-term sr~rvival after orthotopic liver homotransplnn- tation m the dog. Snmery. 38:131, 1965.

24. Stnrzl. T. E., hlarchioro, T. L., Rowlands. D. T.. Jr., Kirkpatrick, C. H., Wilson, LV. E. C., Rifkind. I)..

, ~ n d \Vaddell, \V. R.: lmmunosuppression after experi- lnental iund clinicl~l homotransplantation of the liver. .inn. S~lrg. 160:411, 1964.

23. Stuart. F. P., Torres, E., Hester, W. J., Damnvn. G. J., clnd Noore, F. D.: Orthotopic autotransplantation and allotransplantation of the liver: Functional and htructural patterns in the dog. Ann. Surg. 165:325, 1967.

26. Thomford, X. R., Shorter, R. G., and Hallenbeck. G. A.: Homotransplantation of the canine liver: Sur- vival and histology with and without azathiopr~ne. .\rch. Surg. 90:527. 1963.

27. Tretbar. L. L.. Be\ E. G., and Herrmann. R. E.: Homotransplantatioi . an atlxiliarv doc liver into the pelvis: Effect of portacaval shunt in the p r e - vention of liver atrophy. Sure. Fonlm. 16:219. 1963.

28. van der Heyde, \I. S.. Schalm. L.. . ~ n d \'ink. 11.: The role of functional competition in ausiliarv liver trilnsplantation. Transplantation. 3:78, 1967.

29. Vang, J. O., Zemel, R., and Drapanas, T.: Citrate nletabolism by the isolated perfused liver. .\rch. Sr~rg. 93: 142, 1966.

30. Welch, C. S.: A note on the transplantation of the whole liver in dogs. Transplant. Bull. 2:34, 1953.

31. Welbom, . \I . B., Jr., Lanier, V. C., Jr., and Foster, J. H.: Hepatotrophic influence of portal venous blood. Surg. Forum. 17:381, 19' ,

,In, D. E.. G., and

.versity of I I