Original Article

Modern Pathology (2006) 19, 832–838. doi:10.1038/modpathol.3800605; published online 31 March 2006

The impact of pathologist experience on liver transplant biopsy interpretation

Carla S Coffin1, Kelly W Burak1, John Hart2 and Zu-hua Gao3

  1. 1Department of Medicine, University of Calgary, Calgary, AB, Canada
  2. 2Department of Pathology, Laboratory of Surgical Pathology, The University of Chicago Hospitals, Chicago, IL, USA
  3. 3Department of Pathology and Laboratory Medicine, University of Calgary, Calgary, AB, Canada

Correspondence: Dr Z-h Gao, MD, PhD, Department of Pathology and Laboratory Medicine, Foothills Medical Center, The University of Calgary, Room C1150E, 1403-29 Street NW, Calgary, AB, Canada T2N 2T9. E-mail: zu-hua.gao@cls.ab.ca

Received 1 February 2006; Revised 7 March 2006; Accepted 10 March 2006; Published online 31 March 2006.

Top

Abstract

We studied the impact of pathologist experience on liver transplant biopsy interpretation for cases designated 'nonspecific' by pathologists at a nontransplant center. Among 102 consecutive liver transplant biopsies from 92 patients performed at the Foothills Medical Center, 30 liver biopsies from 23 patients were designated 'nonspecific' by the local pathologist. These biopsy slides were independently reviewed by an expert in liver transplant pathology at a major US transplant center. The expert pathologist was given only the information on the original requisition. In seven biopsies from five patients, there was full agreement between the external expert and the local pathologist. In 10 biopsies from six patients, the expert concurred with the initial assessment but emphasized critical negatives such as 'no evidence of rejection or recurrent hepatitis'. A discrepant diagnosis was found in 13 biopsies from 12 patients. In five biopsies from four patients, the revised diagnoses were inaccurate due to insufficient or misleading clinical information on the requisition. In eight biopsies from eight patients, the revised diagnoses were proven to be correct by clinicopathologic correlation. Our study shows that pathology expertise helped to clarify the diagnosis in about 27% of cases, which justifies the cost of obtaining a second opinion in difficult biopsies. Misinterpretation by the expert pathologist in up to 17% of biopsies highlights the importance of direct communication between hepatologist and pathologist in order to achieve a correct diagnosis. Familiarity with those cases with relatively uncommon histology, a diligent search for subtle morphologic changes, and use of standard terminology could improve the quality of liver transplant biopsy interpretation in a nontransplant center.

Keywords:

liver biopsy, transplantation, pathology, second opinion

Pathological evaluation of liver allograft biopsies plays an integral role in the management of patients following liver transplantation. In certain conditions, such as liver allograft rejection, histology is regarded as the 'gold standard' because no other reliable diagnostic marker exists. In other conditions such as cytomegalovirus (CMV) infection, the cause of graft dysfunction can be identified using other methods, but liver biopsy provides additional information. In recurrent and de novo diseases, the liver biopsy may provide prognostic information such as grade and stage and can reveal superimposed pathology or signs of biliary or vascular abnormalities.

As liver transplantation is performed only in highly specialized tertiary care centers, a significant proportion of patients are evaluated and followed elsewhere. The clinician relies on the pathological interpretation of liver biopsy tissue to diagnose or rule out graft rejection, drug toxicity, recurrence of pretransplant diseases (eg viral hepatitis), or the development of new diseases (eg CMV hepatitis). An accurate pathological report using standard terminology and grading system is essential for patient management. However, a review of the literature, as well as data from our institution suggests that 20–30% of transplant liver biopsies are reported as 'nonspecific changes'.1, 2, 3 In this study, an expert liver pathologist at a major US liver transplant center independently reviewed 30 sets of liver transplant biopsies that were previously diagnosed as 'nonspecific' by pathologists at a nontransplant center. To the best of our knowledge, this is the first study that addresses the impact of pathologist experience on liver transplant biopsy interpretation. A detailed discussion of these cases with characteristic morphology and significant clinical implications could help to improve pathologic evaluation of liver transplant biopsies in a nontransplant center.

Top

Materials and methods

Among 102 biopsies from 92 patients collected in a 3-year period (January 2001–March 2004), 30 biopsies from 23 patients were diagnosed as either 'nonspecific inflammation' or with only a microscopic description as the diagnosis. All 23 patients were seen at the Southern Alberta Liver Transplant Clinic in Calgary, Canada. This clinic follows approximately 150 patients primarily transplanted at the University of Alberta in Edmonton. Ultrasound-guided or blind needle liver biopsies (Tru-Cut 18G needles, C.R. Bard Inc, Covington, Georgia, USA) were used for sample collection. It was routine to stain eight slides in each case, including two Hematoxylin–Eosin-stained slides, and slides stained for Perls' iron, Masson's trichrome, Periodic Acid-Schiff either with or without diastase digestion, reticulin, and modified Shikata's orcein. Acute allograft rejection was graded according to the International Banff Schema.1 The primary hepatologist and local pathologist review together all liver transplant biopsy slides and exchange clinical information. Difficult biopsy slides are often sent to be reviewed by the liver transplant pathologist at the University of Alberta, where the patient received the liver transplant, however, none of these cases with 'nonspecific' pathologic diagnosis were reviewed by an external consultant pathologist.

All slides from these 23 patients were sent to an expert pathologist at the University of Chicago (JH) who did not perform any additional stains and obtained clinical information only from the original pathology requisition. The regular rate for consultation of $300 per case was not charged. Following expert consultant review, demographic information and clinical data were obtained by chart review and discussion with the primary hepatologist (KWB). The cases were then classified into: (1) full agreement; (2) agreement with clarification of terminology; and (3) having a discrepant diagnosis. There was no direct impact on patient management due to the retrospective nature of the study. The potential clinical implications of revised diagnoses were assessed by the primary hepatologist based on clinicopathologic correlation.

Top

Results

There were 23 patients (nine females and 14 males) from which 30 'nondiagnostic' liver biopsies were re-reviewed. In seven biopsies from five patients, there was full agreement between the external expert and the local pathologist. In three biopsies from one patient, the original pathologic diagnoses of 'mild nonspecific inflammation' were revised to 'liver tissue with normal histology' accepting a minimum degree of inflammation as a variant of normal after transplantation. In two biopsies from one patient, the original patholology reports were 'prominent fatty change with inflammation' and the revised diagnoses were 'steatohepatitis', using a standard terminology to replace a descriptive diagnosis. In one patient transplanted for Hepatitis B with liver failure, the CT scan reported a hypoechoic mass in the liver. The biopsy diagnosis given by the local pathologist was descriptive 'periportal fibrosis, mild focal sinusoidal dilatation and mild (grade I/IV) hemosiderosis'. The revised diagnosis was 'no evidence of mass lesion in the biopsy tissue', which clearly informed the hepatologist that either the lesion was not sampled or there was no lesion. There was in fact no lesion on follow-up CT scan. In one patient transplanted for hepatitis C, the biopsy diagnosis made by the local pathologist was 'chronic passive congestion with sinusoidal dilatation' and the revised diagnosis was 'suggest outflow obstruction' which was later confirmed by ultrasound. In two biopsies from a case transplanted for hepatitis B and one biopsy from a case transplanted for autoimmune hepatitis, the original diagnoses of 'nonspecific inflammation' were revised to 'no evidence of rejection and no evidence of recurrent hepatitis B or autoimmune hepatitis'. The revised diagnosis emphasized the critical negatives, which reassured the hepatologist and the patient.

A discrepant diagnosis was found in 13 biopsies from 12 patients. In five biopsies from four patients (cases 9, 10, 11a, 11b, 12), the revised diagnoses were incorrect, primarily due to insufficient or misleading clinical information provided on the requisition. In eight biopsies from eight patients, the revised diagnoses were proven to be correct by clinicopathologic correlation. These cases include: early recurrent hepatitis C (case 1), fibrosing cholestatic hepatitis (case 2), evolving chronic rejection (cases 3 and 4), nodular regenerative hyperplasia (cases 5 and 6), de novo autoimmune hepatitis (case 7) and alcoholic cholestatic hepatitis (case 8). Results of the individual cases are presented in Table 1.


Top

Discussion

In liver transplant patients biopsied for graft dysfunction, sufficient clinical information and detailed analysis of good quality representative biopsy material usually facilitates a specific diagnosis. However, up to 40% of protocol biopsies in patients with normal or nearly normal graft function will show a variable degree of chronic hepatitis.2, 3, 4 It is important to recognize that a nonspecific category may be appropriate in certain clinical situations as it prevents 'pigeon-holing' of biopsies and unnecessary committed therapy. If the changes are only reported as mild, this provides reassurance and may allow the clinician more time to observe and repeat biopsy if necessary. In this study, there was full agreement between local and expert pathologist on the diagnosis of 'nonspecific inflammation' in seven biopsies from five patients. In 10/30 biopsies there was no diagnostic disagreement but the expert consultant recommended clarifying the terminology. This emphasizes the need for standard diagnostic terminology and quality improvement in pathology reporting.

In 5/30 (17%) cases, the revised diagnosis from the expert pathologist was found to be incorrect by clinicopathologic correlation. Although factors such as sampling error and tissue processing could theoretically affect the accuracy of pathologic interpretation,5, 6 insufficient or misleading clinical information was the major cause of misinterpretation. For example, the consultant pathologist had given an erroneous diagnosis of recurrent viral hepatitis when unaware of the patient's negative viral serology. This data further highlights the importance of direct communication between hepatologist and pathologist in order to achieve a correct diagnosis.

In 8/30 (27%) cases, the revised diagnosis from the expert pathologist was proved to be correct by clinicopathologic correlation, which is likely due to the expert's increased experience in liver transplant pathology and ability to appreciate subtle pathological changes. Familiarity with the morphology and clinical implications of these discrepant diagnoses could help to improve pathologic evaluation of liver transplant biopsies in a nontransplant center.

Recurrent Hepatitis C

Hepatitis C recurrence is nearly universal after transplantation. Recurrent hepatitis C leads to chronic hepatitis and liver cirrhosis in a significant proportion of patients and lowers graft and patient survival.7 The difficulty for the local pathologist in diagnosing recurrent hepatitis C arises in two scenarios. The first scenario is early recurrent hepatitis C when there are only small numbers of lymphocytes in the portal tracts and lobules with occasional acidophil bodies (case 1, Figure 1a). The second scenario is a severe fibrosing cholestatic variant seen in 5% of recurrent hepatitis C.8, 9 Pathologists unfamiliar with fibrosing cholestatic hepatitis could mistakenly interpret the relative paucity of portal infiltrates and the prominence of ductular reaction in association with lobular cholestasis as an abnormality of the bile ducts (case 2, Figure 1b). Accurate diagnosis of recurrent hepatitis C, especially the fibrosing cholestatic hepatitis variant, is important for appropriate reduction of immunosuppression and for predicting the prognosis.

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

(a) Early recurrent hepatitis C. A portal tract with mild lymphocytic infiltration and an acidophil body in the adjacent parenchyma (hematoxylin–eosin, magnification times 400). (b) Fibrosing cholestatic variant of recurrent hepatitis C. There is less dense portal mononuclear infiltrate than is often seen in chronic viral hepatitis. The portal area shows markedly fibrosis with fibrous strands radiate into the lobular parenchyma. Ductular proliferation in association with lobular cholestasis is evident (hematoxylin–eosin, magnification times 400). (c) Evolving chronic rejection. An interlobular bile duct shows nuclear pleomorphism, hyperchromasia and distorted polarity, producing a dysplastic appearance (hematoxylin–eosin, magnification times 400). (d) Evolving chronic rejection. Pericentral ballooning degeneration of hepatocytes, hepatocyte drop out and sinusoidal inflammation (hematoxylin–eosin, magnification times 100). (e) Nodular regenerative hyperplasia. Expansile nodule composed of thick liver cell plates intervened by compressed atrophic liver cell plates (reticulin, magnification times 100). (f) De novo autoimmune hepatitis. Infiltration of lymphocytes and plasmacytes in the portal tract, interface and lobular parenchyma (hematoxylin–eosin, times magnification 400). (g) Alcoholic cholestatic hepatitis. Feathery degeneration of hepatocytes and ductular proliferation in earlier biopsies. Some hepatocytes showed microvesicular steatosis. Bile plug is present in a canalicular space. (hematoxylin–eosin, magnification times 400). (h) Alcoholic cholestatic hepatitis. Macrovesicular steatosis and sinusoidal fibrosis in subsequent biopsies (Masson's trichrome, magnification times 400).

Full figure and legend (1,235K)

Evolving Chronic Rejection

In a review of 870 consecutive primary liver transplant recipients who survived 15 days or more following liver transplantation, the incidence of graft failure from chronic rejection was 2.4%.10 The diagnostic histologic features of chronic rejection are loss of bile ducts and obliterative arteriopathy affecting large- or medium-sized arteries.11 The fact that these arteries are not usually biopsied, and that bile duct loss is not always apparent, makes it difficult for a pathologist at a non-liver transplant center to confidently diagnose chronic rejection. In a patient presenting with progressive jaundice, the following histological findings should make evolving chronic rejection a serious consideration: (1) 'dysplastic' or atrophic bile ducts featuring lymphocytic infiltration, nuclear pleomorphism, disordered polarity and focal attenuation/disruption of biliary epithelium (case 3, Figure 1c), (2) pericentral ballooning degeneration of hepatocytes, hepatocyte drop out and intimal or sinusoidal inflammation (case 4, Figure 1d), (3) cholestasis in the absence of ductular proliferation.11 An early diagnosis of chronic rejection is imperative so that salvage therapy can be initiated. It is easier to diagnosis advanced classical chronic rejection but the clinical value of this is limited.

Nodular Regenerative Hyperplasia

Nodular regenerative hyperplasia is usually associated with autoimmune disease, medications and a variety of disorders. It is a common finding in late post-transplant biopsies.12, 13, 14 Some cases appear to be related to drug toxicity (particularly Azathioprine) or a vascular abnormality (eg portal venous insufficiency). However, in many cases no obvious cause can be found. Microscopically, nodular regenerative hyperplasia is characterized by diffuse nodularity of the liver with nodules formed by thickened liver cell plates (2–3 cells thick) with intervening atrophic parenchyma. There is no evidence of fibrosis or inflammation. The expansile architecture is best appreciated in reticulin preparations (case 5, Figure 1e). Although the clinical impact of identifying nodular regenerative hyperplasia in liver allografts may be limited, it may represent the cause of otherwise unexplained liver enzyme elevation. Awareness and identification of this entity could prevent unnecessary further investigations.

De Novo Autoimmune Hepatitis

De novo autoimmune hepatitis has been reported in patients transplanted for other reasons in both pediatric and adult populations.15, 16, 17 All reported cases have had high serum concentrations of immunoglobulin G and high titers of autoantibodies. Liver biopsies from these patients show portal and periportal hepatitis with lymphocytes and plasma cells, bridging collapse, and perivenular cell necrosis without changes typical of acute or chronic rejection (case 7, Figure 1f). The diagnosis requires the incorporation of serology and morphological findings and the disease usually responds to increased immunosuppression. In our case, lack of knowledge about the patient's original liver disease, serum IgG levels, and clinical follow-up makes it impossible to exclude the differential diagnosis of viral hepatitis, adverse drug reaction or recurrence of possible original autoimmune hepatitis.

Alcoholic Cholestatic Hepatitis

It has been estimated that about 10–30% of alcoholic patients return to drinking following transplantation.18, 19 The diagnosis is difficult when uncommon histology such as cholestasis is present, or only microvescicular steatosis is evident (case 8, Figure 1g). The characteristic changes of macrovesicular steatosis, steatohepatitis and pericentral sinusoidal fibrosis usually manifest in subsequent biopsies when the disease is more advanced (case 8, Figure 1h). Awareness of uncommon presentations and a diligent search for subtle morphological signs of alcohol-related liver damage could prevent repeated admissions/liver biopsies and allow early treatment.

Previous studies have stressed the importance of secondary re-review of medical and surgical biopsies.20 In one study of 178 consecutive nontransplant liver biopsy slides provided to hepatopathology consultants, the authors identified a major discrepant diagnosis in 28% of specimens.21 In the year 2000, a consensus conference of expert pathologists stated that medical biopsies such as liver are so frequently misinterpreted or interpreted in a clinically nonmeaningful manner that expert reevaluations are essential.22 This is especially important in the interpretation of liver allograft biopsies. As this was a retrospective study, we were unable to determine if the revised diagnosis affected patient outcome. We are unable to comment on the potential problem of timeliness of an external review and how feasible it would be to wait for a second opinion. However, the issue of waiting could be resolved by on-line consultation with the increasing use of telepathology.23

In summary, we have found that pathology expertise could help to clarify the diagnosis of about 27% cases that were designated 'nonspecific' by the local pathologist. Therefore, a significant proportion of post-transplant liver biopsies which are regarded as 'nonspecific' by a general pathologist may benefit from review by an expert liver pathologist, especially when patient have clinical evidence of significant graft dysfunction. We also found that up to 17% cases were incorrectly interpreted by the expert pathologist due to insufficient or misleading clinical information on the requisition, which further highlights the importance of direct communication between hepatologist and pathologist in order to achieve a correct diagnosis. Although this was only a single institution study with a small number of cases, it has identified several case scenarios with characteristic morphology and important clinical implications. Familiarity with these relatively uncommon histological findings, a diligent search for subtle morphologic changes and the use of standard terminology could improve the quality of liver transplant biopsy interpretation in a nontransplant center.

Top

References

  1. Demetris A, Batts K, Dhillon A, et al. Banff schema for grading liver allograft rejection: an international consensus document. Hepatology 1997;25:658–663. | PubMed | ISI |
  2. Nakhleh R, Schwarzenberg S, Bloomer J, et al. The pathology of liver allografts surviving longer than one year. Hepatology 1990;11:465–470. | PubMed | ISI | ChemPort |
  3. Pappo O, Ramos H, Starzl T, et al. Structural integrity and identification of causes of liver allograft dysfunction occurring more than 5 years after transplantation. Am J Surg Pathol 1995;19:192–206. | PubMed | ISI | ChemPort |
  4. Rosenthal P, Emond J, Heyman M, et al. Pathological changes in yearly protocol liver biopsy specimens from healthy pediatric liver recipients. Liver Transpl Surg 1997;3:559–562. | Article | PubMed | ChemPort |
  5. Colloredo G, Guido M, Sonzogni A, et al. Impact of liver biopsy size on histological classification of chronic viral hepatitis: The smaller the sample, the milder the disease. J Hepatol 2003;39:239–244. | Article | PubMed | ISI |
  6. Ottevanger P, Therasse P, Van de Velde C, et al. Quality assurance in clinical trials. Crit Rev Oncol Hematol 2003;47:213–235. | PubMed | ISI | ChemPort |
  7. Pessoa M, Terrault N, Ferrell L, et al. Hepatitis after liver transplantation: the role of known and unknown viruses. Liver Transpl Surg 1998;4:461–468. | Article | PubMed | ISI | ChemPort |
  8. Schluger L, Sheiner P, Thung S, et al. Severe recurrent cholestatic hepatitis c following orthotopic liver transplantation. Hepatology 1996;23:971–976. | Article | PubMed | ISI | ChemPort |
  9. Taga S, Washington M, Terrault N, et al. Cholestatic hepatitis c in liver allografts. Liver Transpl Surg 1998;4:304–310. | PubMed | ISI | ChemPort |
  10. Wiesner R, Batts K, Krom R. Evolving concepts in the diagnosis, pathogenesis and treatment of chronic hepatic allograft rejection. Liver Transpl Surg 1999;5:388–400. | Article | PubMed | ISI | ChemPort |
  11. Demetris A, Quan S, Sun H, et al. Liver allograft rejection: an overview of morphologic findings. Am J Surg Pathol 1990;14:49–63. | PubMed | ISI |
  12. Gane E, Portmann B, Saxena R, et al. Nodular regenerative hyperplasia of the liver graft after liver transplantation. Hepatology 1994;20:88–94. | Article | PubMed | ISI | ChemPort |
  13. Sebagh M, Farges O, Samuel D, et al. Nodular regenerative hyperplasia of the liver following orthotopic liver transplantation. Transplant proc 1995;27:2510–2511. | PubMed | ISI | ChemPort |
  14. Wanless I. Micronodular transformation (nodular regenerative hyperplasia) of the liver: A report of 64 cases among 2500 autopsies and a new classification of benign hepatic nodules. Hepatology 1990;11:787–797. | PubMed | ISI | ChemPort |
  15. Gupta P, Hart J, Millis J, et al. De novo hepatitis with autoimmune antibodies and atypical histology: a rare cause of late graft dysfunction after pediatric liver transplanatation. Transplantation 2001;71:664–668. | PubMed | ISI | ChemPort |
  16. Jones D, James O, Portmann B, et al. Development of autoimmune hepatitis following liver transplantation for primary billiary cirrhosis. Hepatology 2000;32:868–870. | PubMed |
  17. Kerkar N, Hadzic N, Davies ET, et al. De-novo autoimmune hepatitis after liver transplantation. The Lancet 1998;351:409–413. | Article | ChemPort |
  18. Gerhardt T, Goldstein R, Urschel H, et al. Alcohol use following liver transplantation for alcoholic cirrhosis. Transplantation 1996;62:1060–1063. | PubMed | ISI | ChemPort |
  19. Lim J, Keeffe E. Liver transplantation for alcoholic liver disease: current concept and length of sobriety. Liver Transpl 2004;10:S31–S38. | Article | PubMed | ISI |
  20. Tsung JSH. Institutional pathology consultation. Am J Surg Pathol 2004;28:399–402. | PubMed | ISI |
  21. Bejarano PA, Koehler A, Sherman KE. Second opinion pathology in liver biopsy interpretation. Am J Gastroenterol 2001;96:3158–3164. | Article | PubMed | ISI | ChemPort |
  22. Tomaszewski J, Bear H, Connally J, et al. Consensus conference on second opinion in diagnostic anatomic pathology. Who what, and when. Am J Clin Pathol 2000;114:329–335. | PubMed | ISI | ChemPort |
  23. Ito H, Adachi H, Taniyama K, et al. Telepathology is available for transplantation-pathology: experience in Japan using an integrated, low cost, and high quality system. Mod Pathol 1994;7:801–805. | PubMed | ISI | ChemPort |
Top

Acknowledgements

We thank Dr Wanda Lester for her assistance in editing the manuscript.

Extra navigation

.

naturejobs

More pathology jobs

natureproducts


ADVERTISEMENT