This study aimed to evaluate the prognostic significance of the cellular proliferation marker Ki-67 in the liver biopsies of patients with fulminant hepatic failure.

Under IRB approval, the surgical pathology database at TJUH was searched for biopsies with the keywords “confluent”, “submassive”, or “massive hepatic necrosis”. The identified liver biopsies and medical records were reviewed, and all patients with confirmed confluent hepatic necrosis and fulminant hepatic failure were selected. The biopsies were stained for Ki-67 positivity by standard immunohistochemical procedures. The biopsies were read as either proliferative (numerous Ki-67 positive hepatocytes) or non-proliferative (few if any positive cells). From the medical records, demographic and laboratory data and clinical outcome (survival with or without liver transplantation) were obtained.

Thirty patients constituted the study. Of these, 13 patients were eventually transplanted, 13 survived with medical management alone, and 4 died. All the biopsies were easily interpreted as either proliferative (15) or non-proliferative (15). With respect to death, a non-proliferative biopsy had a positive predictive value of 50%. Conversely, a proliferative biopsy had a positive predictive value for survival of 69%. The overall diagnostic accuracy (total correct predictions/total tests) was 77% in non-transplanted patients. Amongst all patients, a non-proliferative biopsy had a positive predictive value for death or transplantation of 53%. A proliferative biopsy had a positive predictive value for survival 73%. For all patients, the overall diagnostic accuracy was 63%. When a non-proliferative biopsy was taken together with a MELD score greater than 32 and the King's College Criteria, a composite score requiring at least 2 of these 3 criteria resulted in a positive predictive value of 88% for death or transplantation. One or less of the same 3 criteria had a positive predictive value for survival without transplantation of 77%. The overall diagnostic accuracy of our composite score (2 or 3 criteria) was 83%, a result more accurate than any of the criteria individually.

Evaluation of liver biopsies for Ki-67 expression provides a useful adjunct to established criteria in determining the prognosis of patients with fulminant liver failure. Larger studies should be performed to determine the true import of Ki-67 activity in such patients.

« Back to 2015 American Transplant Congress