Advertisement

Prediction of liver-related mortality in a community setting

  • Author Footnotes
    # contributed equally
    Carolin V. Schneider
    Footnotes
    # contributed equally
    Affiliations
    Department of Internal Medicine III, Gastroenterology, Metabolic diseases and Intensive Care, University Hospital RWTH Aachen, Aachen, Germany
    Search for articles by this author
  • Author Footnotes
    # contributed equally
    Stefan Gross
    Footnotes
    # contributed equally
    Affiliations
    Department of Internal Medicine B, Cardiology, University Medicine Greifswald, Greifswald, Germany

    DZHK (German Center for Cardiovascular Research), Partner Site Greifswald, Germany
    Search for articles by this author
  • Pavel Strnad
    Correspondence
    Corresponding author. Pavel Strnad, MD, Department of Internal Medicine III, RWTH Aachen, Pauwelsstr. 30, 52074 Aachen, Germany, , Tel: 00492418035324
    Affiliations
    Department of Internal Medicine III, Gastroenterology, Metabolic diseases and Intensive Care, University Hospital RWTH Aachen, Aachen, Germany
    Search for articles by this author
  • Author Footnotes
    # contributed equally
Published:November 09, 2022DOI:https://doi.org/10.1016/j.jhep.2022.11.002

      Linked Article

      We would like to congratulate Innes et al. for their study assessing the usefulness of established risk scores as a predictor of development of liver cirrhosis-related complications.
      • Innes H.
      • Morling J.R.
      • Buch S.
      • Hamill V.
      • Stickel F.
      • Guha I.N.
      Performance of Routine Risk Scores for Predicting Cirrhosis-Related Morbidity in the Community.
      Because of their everyday relevance, the study also raises several questions. While development of liver cirrhosis-related complications constitutes an indisputable hard end-point, the occurrence of liver-related death as a primary cause of death should also be considered. Given that UK Biobank (UKB) is linked to national death registries, the survival status and the causes of death are readily available. In our dataset (Total N = 502,511 exclusion criteria: non-European or unknown ethnicity, liver transplantation before study beginning, diagnosis of viral hepatitis C and B (ICD10: B16-B19), or self-reported risky alcohol consumption (>60 g/>40 g alcohol/d for men/women, missing values in age at baseline, sex, vital status, follow-up time, and variables to calculate APRI, FIB-4, CIRRUS, ALBI-FIB4, ALBI, PALBI, and NAR scores), this information was retrieved for 391,631 participants with a median follow-up period of 11.3 (IQR 10.6; 12.0) years until September 2022. All analyses were done using R version 4.1.3 with packages tidyverse, survival, survminer, and riskRegression. This study was approved by the UKB under application number 47527.
      During this period of time, 782 liver-related deaths were recorded and this event was approximately half as common than occurrence of liver cirrhosis-related complications observed by Innes et al. When considering liver-related death as a time-to-event variable, CIRRUS score constituted a reasonable predictor while APRI and FIB-4 scores performed less well (Figure 1A: time-dependent AUROC based on age/sex adjusted Cox models). The better short-term performance of CIRRUS as well as ALBI score is not surprising since they recognize advanced liver disease while APRI and FIB-4 are used to stratify liver fibrosis in general and lack liver synthesis parameters.
      • Innes H.
      • Morling J.R.
      • Buch S.
      • Hamill V.
      • Stickel F.
      • Guha I.N.
      Performance of Routine Risk Scores for Predicting Cirrhosis-Related Morbidity in the Community.
      • Johnson P.J.
      • Berhane S.
      • Kagebayashi C.
      • Satomura S.
      • Teng M.
      • Reeves H.L.
      • et al.
      Assessment of Liver Function in Patients with Hepatocellular Carcinoma: A New Evidence-Based Approach-the ALBI Grade.
      • Hydes T.
      • Moore M.
      • Stuart B.
      • Kim M.
      • Su F.
      • Newell C.
      • et al.
      Can Routine Blood Tests Be Modelled to Detect Advanced Liver Disease in the Community: Model Derivation and Validation Using UK Primary and Secondary Care Data.
      Figure thumbnail gr1
      Figure 1Liver-related death screening capabilities of different scores in the population-based UKB.
      A) Time-dependent Area under the curve (AUROC) with 95% confidence interval per score on age/sex adjusted Cox models
      B) Death within follow-up time stratified by score and by percentile of risk
      C) 10-year death within follow-up time stratified by score and by percentile of risk
      Abbreviations: UKB, UK Biobank; APRI, AST-to-platelet-ratio; Fib-4, Fibrosis-4Index; CIRRUS, CIRRhosis Using Standard tests; ALBI-FIB-4, albumin-bilirubin fibrosis-4 index; ALBI, Albumin-Bilirubin score; PALBI, Platelets- Albumin-Bilirubin score; NAR, neutrophil-to-albumin ratio.
      Given that both emergence of liver cirrhosis-related complications and liver death constituted very rare events in the UKB, we evaluated the usefulness of available scores for the routine patient triage. For this, we studied three different scenarios, i.e., selection of 20%, 10% and 5% of participants with the highest values to determine, how many liver-related deaths would remain undetected when these triage algorithms would be employed. We separately assessed the whole dataset (Figure 1B) and a subset of subjects in whom 10-year follow-up data were available (Figure 1C). Notably, both analyses yielded nearly identical results. In all tested scenarios, APRI score
      • Wai C.-T.
      • Greenson J.K.
      • Fontana R.J.
      • Kalbfleisch J.D.
      • Marrero J.A.
      • Conjeevaram H.S.
      • Lok A.S.-F.
      A Simple Noninvasive Index Can Predict Both Significant Fibrosis and Cirrhosis in Patients with Chronic Hepatitis C.
      recognized the highest number of subjects, who suffered a liver-related death. At the 20% selection level, APRI and FIB-4
      • Sterling R.K.
      • Lissen E.
      • Clumeck N.
      • Sola R.
      • Correa M.C.
      • Montaner J.;S.
      • et al.
      APRICOT Clinical Investigators
      Development of a Simple Noninvasive Index to Predict Significant Fibrosis in Patients with HIV/HCV Coinfection.
      performed similarly and detected 60.4%, and 59.2% respectively of participants who developed liver-related death during the follow-up period, while the remaining scores were inferior. At the 10 and 5% levels, APRI was superior to the others but recognized only 45.0-51.7% of the subjects we sought to identify. Collectively, our analyses confirm the data from Innes et al. suggesting that APRI, FIB-4 and CIRRUS are the best predictors in the routine setting, but their usefulness depends on the exact aim. More importantly, in the UK Biobank, none of these scores were useful for everyday hepatologic triage of the participants. In line with our data, SEAL program employed a combination of elevated AST/ALT levels and APRI to identify cirrhotics in the routine care but reported only a moderately higher increase in detection rates.
      • Labenz C.
      • Arslanow A.
      • Nguyen-Tat M.
      • Nagel M.
      • Wörns M.-A.
      • Reichert M.
      • et al.
      Structured Early Detection of Asymptomatic Liver Cirrhosis: Results of the Population-Based Liver Screening Program SEAL.
      These data clearly indicate that further studies are needed to enable the detection of liver-sick patients in the routine care. Needless to say, the performed analysis has several important caveats. UKB is skewed towards White, high-income individuals
      • Bycroft C.
      • Freeman C.
      • Petkova D.
      • Band G.
      • Elliott L.T.
      • Sharp K.
      • et al.
      The UK Biobank Resource with Deep Phenotyping and Genomic Data.
      and the ICD-10 codes that were used for assessment of liver related death carry a risk of misclassification.

      Author contribution

      Acquisition of data: C.V.S, P.S; Analysis and interpretation of data: all authors; Drafting of the manuscript: all authors; Critical revision of the manuscript for important intellectual content: all authors; Figures and tables: S.G.; Statistical analysis: S.G.; Obtained funding: none; Study supervision: P.S.

      Declaration of interest

      The authors declare no competing interests.

      Financial support or funding

      None.

      Acknowledgements

      All authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. All authors agreed to submit the manuscript, read, and approved the final draft and take full responsibility of its content, including the accuracy of the data and its statistical analysis.

      References

        • Innes H.
        • Morling J.R.
        • Buch S.
        • Hamill V.
        • Stickel F.
        • Guha I.N.
        Performance of Routine Risk Scores for Predicting Cirrhosis-Related Morbidity in the Community.
        J. Hepatol. 2022; 77: 365-376https://doi.org/10.1016/j.jhep.2022.02.022
        • Johnson P.J.
        • Berhane S.
        • Kagebayashi C.
        • Satomura S.
        • Teng M.
        • Reeves H.L.
        • et al.
        Assessment of Liver Function in Patients with Hepatocellular Carcinoma: A New Evidence-Based Approach-the ALBI Grade.
        J. Clin. Oncol. Off. J. Am. Soc. Clin. Oncol. 2015; 33: 550-558https://doi.org/10.1200/JCO.2014.57.9151
        • Hydes T.
        • Moore M.
        • Stuart B.
        • Kim M.
        • Su F.
        • Newell C.
        • et al.
        Can Routine Blood Tests Be Modelled to Detect Advanced Liver Disease in the Community: Model Derivation and Validation Using UK Primary and Secondary Care Data.
        BMJ Open. 2021; 11e044952https://doi.org/10.1136/bmjopen-2020-044952
        • Wai C.-T.
        • Greenson J.K.
        • Fontana R.J.
        • Kalbfleisch J.D.
        • Marrero J.A.
        • Conjeevaram H.S.
        • Lok A.S.-F.
        A Simple Noninvasive Index Can Predict Both Significant Fibrosis and Cirrhosis in Patients with Chronic Hepatitis C.
        Hepatol. Baltim. Md. 2003; 38: 518-526https://doi.org/10.1053/jhep.2003.50346
        • Sterling R.K.
        • Lissen E.
        • Clumeck N.
        • Sola R.
        • Correa M.C.
        • Montaner J.;S.
        • et al.
        • APRICOT Clinical Investigators
        Development of a Simple Noninvasive Index to Predict Significant Fibrosis in Patients with HIV/HCV Coinfection.
        Hepatol. Baltim. Md. 2006; 43: 1317-1325https://doi.org/10.1002/hep.21178
        • Labenz C.
        • Arslanow A.
        • Nguyen-Tat M.
        • Nagel M.
        • Wörns M.-A.
        • Reichert M.
        • et al.
        Structured Early Detection of Asymptomatic Liver Cirrhosis: Results of the Population-Based Liver Screening Program SEAL.
        J. Hepatol. 2022; 77: 695-701https://doi.org/10.1016/j.jhep.2022.04.009
        • Bycroft C.
        • Freeman C.
        • Petkova D.
        • Band G.
        • Elliott L.T.
        • Sharp K.
        • et al.
        The UK Biobank Resource with Deep Phenotyping and Genomic Data.
        Nature. 2018; 562: 203-209https://doi.org/10.1038/s41586-018-0579-z