Non-invasive fibrosis algorithms are clinically useful for excluding cirrhosis in prisoners living with hepatitis C

Timothy Papaluca, Anne Craigie, Lucy McDonald, Amy Edwards, Michael MacIsaac, Jacinta A. Holmes, Matthew Jarman, Tanya Lee, Hannah Huang, Andrew Chan, Mark Lai, Vijaya Sundararajan, Joseph S. Doyle, Margaret Hellard, Mark Stoove, Jessica Howell, Paul Desmond, David Iser, Alexander J. Thompson

Research output: Contribution to journalArticleResearchpeer-review

6 Citations (Scopus)

Abstract

BACKGROUND AND AIMS: Prison-based HCV treatment rates remain low due to multiple barriers, including accessing transient elastography for cirrhosis determination. The AST-to-platelet ratio index (APRI) and FIB-4 scores have excellent negative predictive value (NPV) in hospital cohorts to exclude cirrhosis. We investigated their performance in a large cohort of prisoners with HCV infection. METHODS: This was a retrospective cohort study of participants assessed by a prison-based hepatitis program. The sensitivity, specificity, NPV and positive predictive value (PPV) of APRI and FIB-4 for cirrhosis were then analysed, with transient elastography as the reference standard. The utility of age thresholds as a trigger for transient elastography was also explored. RESULTS: Data from 1007 prisoners were included. The median age was 41, 89% were male, and 12% had cirrhosis. An APRI cut-off of 1.0 and FIB-4 cut-off of 1.45 had NPVs for cirrhosis of 96.1% and 96.6%, respectively, and if used to triage prisoners for transient elastography, could reduce the need for this investigation by 71%. The PPVs of APRI and FIB-4 for cirrhosis at these cut-offs were low. Age ≤35 years alone had a NPV for cirrhosis of 96.5%. In those >35 years, the APRI cut-off of 1.0 alone had a high NPV >95%. CONCLUSION: APRI and FIB-4 scores can reliably exclude cirrhosis in prisoners and reduce requirement for transient elastography. This finding will simplify the cascade of care for prisoners living with hepatitis C.

Original languageEnglish
Article numbere0242101
Number of pages11
JournalPLoS ONE
Volume15
Issue number11
DOIs
Publication statusPublished - 2020

Cite this