BACKGROUND Occult hepatitis B virus infection is defined as detectable HBV-DNA in liver of HBsAg-negative individuals, with or without detectable serum HBV-DNA. In deceased liver donors, results of tissue analysis cannot be obtained prior to allocation for liver transplantation. OBJECTIVE we investigated prevalence and predictability of occult hepatitis B using blood markers of viral exposure/infection in deceased liver donors. METHODS In 50 consecutive HBsAg-negative/anti-HBc-positive and 20 age-matched HBsAg-negative/anti-HBc-negative donors, a nested-PCR assay was employed in liver biopsies for diagnosis of occult hepatitis B according to Taormina criteria. All donors were characterized for plasma HBV-DNA and serum anti-HBs/anti-HBe. RESULTS In liver tissue, occult hepatitis B was present in 30/50 anti-HBc-positive (60%) and in 0/20 anti-HBc-negative donors (p<0.0001). All anti-HBc-positive donors with detectable HBV-DNA in plasma (n=5) or anti-HBs>1,000 mIU/mL (n=5) eventually showed occult infection, i.e, 10/30 occult hepatitis B-positive donors which could have been identified prior to transplantation. In the remaining 40 anti-HBc-positive donors, probability of occult infection was 62% for anti-HBe-positive and/or anti-HBs ≥ 58 mIU/mL; 29% for anti-HBe-negative and anti-HBs<58 mIU/mL. CONCLUSIONS In deceased donors, combining anti-HBc with other blood markers of hepatitis B exposure/infection allows to predict occult hepatitis B with certainty and speed in one third of cases. These findings might help refine the allocation of livers from anti-HBc-positive donors.