In Press, 1995, Transplantation Proceedings copyright by Appleton and Lange.

Chronic Liver Transplant Rejection: Definition and Diagnosis

DGD Wight

Department of Histopathology, Box 235, Addenbrooke's Hospital, Cambridge, CB2 2QQ, UK


Chronic liver transplant rejection was described for the first time in four of Starzl's first 27 patients reported from Denver [1]. Many of the small branches and some of the large branches of the hepatic artery were completely occluded by intimal thickening which included many foam cells. Porter called this process "late rejection with vascular change" and noted that it was similar to that seen in human renal grafts. The arterial narrowing was accompanied in all instances by fibrosis and cholestasis, and in two patients by cirrhosis, and occurred 68 to 186 days after transplantation. Similar changes were seen in two of the first 26 patients reported from the United Kingdom [2]. A later report from Denver, reporting the first 93 patients [3], used the term "chronic rejection" to describe a very similar syndrome, although only three livers were then considered to be affected. No mention was made of bile duct damage in any Denver case and, indeed, one of the cases illustrated in the earlier report [1] (OT9) clearly shows a normal small bile duct. Nevertheless, it is of interest to note that one liver (case not identified), from an anencephalic monster, never cleared bilirubin and when it was removed 85 days later it had "the histopathological findings of intrahepatic biliary atresia". It is possible that this was the first, but unrecognized, description of the syndrome that was clearly defined in the next review from the Cambridge/King's College Hospital group, covering the first 64 patients [4]. One of their patients showed complete absence of small bile ducts (Figure 1), a condition that was later called "vanishing bile duct syndrome" by Portmann [5, 6]. This term has since gained wide acceptance [7-19] and is frequently used interchangeably with "chronic rejection". Ludwig [20] prefers "ductopenic rejection" since the lesion can be found as early as three weeks after transplantation, and the term chronic is therefore, he argues, inappropriate. A recent study from Cambridge [21] suggests that, although in the majority of cases both vascular lesions and ductopenia are found together, in a significant minority of cases each can occur independently. The present study extends the latter series and assesses the role of biopsy in the diagnosis of chronic rejection.



In Cambridge, between 1968 and 1993, 145 patients required a second or subsequent transplant, determined mainly on clinical grounds supported by a variety of investigations, including prior biopsy. 56 livers removed from 51 patients showed either bile duct loss (VBD) or foam cell endovasculitis (FCE). Standard blocks, a minimum of eight from each liver, included samples of hilar and septal regions, right and left lobes. Bile duct loss was assessed by counting 100 portal tracts in random blocks from the right and left lobes and recorded as the percentage of portal tracts with no bile duct. Foam cell endovasculitis was present if arterial lesions were seen in any block, and was scored as causing narrowing or total occlusion. The vessels involved were divided into three sizes, hilar (>0.05cm), medium-sized (0.01 to 0.04 cm) and small (<0.01 cm). Other features identified in the early series listed above [1-6], namely, perivenular dropout, cholestasis (both on a semiquantitative scale from 0 to 3) and fibrosis (presence or absence) were also assessed. The 224 prior biopsies (mean 4 per case) were also assessed for the presence of FCE (positive if present in any vessel) and VBD (considered positive if more than 50% of portal tracts contained no bile duct) and correlated with the subsequent hepatectomy appearances.



38 (68%) livers showed both VBD and FCE, 11 (20%) FCE alone (Table 1) and 7 (12%) VBD alone (Table 2).

FCE was found most often in medium-sized or hilar vessels, of a diameter not normally sampled in needle biopsies (Figure 2). In the 45 livers which showed VBD, this was not an all or none phenomenon. Only 18 (32%) showed 100% bile duct loss, with diminishing proportions of livers showing 75-100%, 50-75%, and four livers less than 50% loss (Figure 3). Perivenular dropout, cholestasis and fibrosis were common but did not correlate well with the other findings. For example, one liver with no VBD but extensive FCE, showed no dropout (Table 1), whilst another, with 100% VBD but no FCE showed neither dropout nor cholestasis (Table 2). Median graft survival was below average in those patients with FCE alone and above average in those with VBD alone (Figure 4).

Of the grafts in which both lesions were present, one or more of the prior biopsies was diagnostic in 71%. When only VBD was present, biopsy was positive in 6/7 (86%), whilst biopsy was positive in only 5 of the 11 livers (45%) with FCE only. Overall, biopsy was positive in 68% (38/56) of the livers subsequently removed for chronic rejection, (Figure 5).



The results confirm the opinion of the international working party [22], that chronic rejection should be defined as FCE and/or >50% bile duct loss. Although the terms vanishing bile duct syndrome or ductopenic rejection are appealing, neither is appropriate in 20% of cases and therefore the non-committal term chronic rejection is to be preferred, despite the early onset in some cases [8]. This is in line with the findings in other organ grafts [23] where it is recognized that the syndrome of chronic rejection consists of FCE plus direct damage to specialized organ tissues, for example the pancreatic acini, pulmonary bronchioles, renal tubules and hepatic bile ducts.

The survival of grafts without arteriopathy is greater than when both lesions are present (Figure 4), and shorter than average when only arteriopathy is present, suggesting that FCE is the more important in determining survival. Furthermore, significant fibrosis was seen mainly in the longer surviving grafts (Tables 1 & 2), suggesting that this is mainly a function of the length of graft survival.

Needle biopsy can be expected to diagnose chronic rejection in about 68% of cases, but other parameters, such as high resolution arteriography, are necessary in the remaining patients. Following a number of reports where bile duct loss has apparently reversed [12], it has been suggested that a biopsy diagnosis of chronic rejection would be unsafe if less than 50% of bile ducts are without a bile duct. A retrospective study of this nature, where the end-point was hepatectomy for graft failure, clearly does not include any cases which have failed to progress. Nevertheless, four of our patients with less than 50% VBD did progress to graft failure with otherwise typical chronic rejection (Figure 3). Furthermore, the variability of the proportion of duct loss seen in the hepatectomy specimens in this series, all of which were removed for graft failure, suggests that any attempt to quantitate or grade chronic rejection on prior biopsies will be doomed to failure.



  1. Porter KA: in Starzl, T.E. (ed.): Experiences in Hepatic Transplantation. Philadelphia, W B Saunders, 1969, p422


  2. Williams R, Smith M, Shilkin KB, et al: Gastroenterology 64:1026, 1973


  3. Starzl TE, Porter KA, Putnam CW, et al: Surg Gynecol Obstet 142:487, 1976


  4. Calne RY, McMaster P, Portmann B, et al: Ann Surg 186:282, 1977


  5. Portmann B, Williams R: in Williams, R. & Cantoni, L. (ed.): Recenti progressi in epatologia. Milan, Casa Editirice Ambrosiana, 1979, p369


  6. Portmann B, Neuberger JM, Williams R: in Calne, R.Y. (ed.): Liver Transplantation. London, Grune & Stratton, 1983, p279


  7. Donaldson PT, Alexander GJM, O'Grady J, et al: Lancet 1:945, 1987


  8. Ludwig J, Wiesner RH, Batts KP, et al: Hepatology 7:476, 1987


  9. O'Grady J, Alexander GJM, Sutherland S, et al: Lancet 2:302, 1988


  10. Pirsch JD, Kalayoglu M, Hafez GR, et al: Transplantation 49:1015, 1990


  11. Wiesner RH, Ludwig J, van Hoek B, et al: Hepatology 14:721, 1991


  12. Hübscher SG, Buckels JA, Elias E, et al: Transplantation 51:1004, 1991


  13. Noack KB, Wiesner RH, Batts K, et al: Transplant Proc , 1991


  14. Arnold JC, Portmann BC, O'Grady JG, et al: Hepatology 16:285, 1992


  15. Desmet VJ: Prog Liver Dis 10:89, 1992


  16. van Hoek B, Wiesner RH, Krom RAF, et al: Semin Liver Dis 12:41, 1992


  17. Rubin R, Munoz SJ: Hum Pathol 24:643, 1993


  18. Dhillon AP, Burroughs AK, Hudson M, et al: Hepatology 19:106, 1994


  19. Gómez R, Colina F, Moreno E, et al: J Hepatol 21:441, 1994


  20. Ludwig J: Mayo Clin. Proc. 64:676 , 1989


  21. Deligeorgi-Politi H, Wight DGD, Calne RY: Transplant Int. 7:442, 1994


  22. Ludwig J: Am. J. Gastroenterol. 89:S177, 1994


  23. Sibley RK: Clin Transplant 8:293, 1994






Figure 1
Photomicrograph of the liver biopsy in which VBD was first described. 4, 5 The patient was a 56 year old man who received a graft for alcoholic cirrhosis, complicated by hepatocellular carcinoma, on 16th June 1976. He developed progressive cholestasis and on the 44th day was found to have a complete absence of interlobular ducts. The section shows a significant-sized artery, but no accompanying portal tract.

Figure 2
Foam cell endovasculitis
FCE - Distribution of lesions.

Figure 3
Portal tract bile duct loss
VBD - Distribution of lesions.

Figure 4
Chronic Rejection
Relative survival of the 11 livers without VBD, and the 7 livers without arteriopathy, compared to that of the whole series of 56 livers.

Figure 5
Chronic Rejection - Summary


Return to Banff Presentations


Last Modified: June 26, 1996 12:40:59 PM