Accuracy and concordance of two-dimensional shear-wave elastography using transient elastography as the reference in chronic viral hepatitis and HIV infection in Rio de Janeiro, Brazil

Eur J Gastroenterol Hepatol. 2023 May 1;35(5):583-590. doi: 10.1097/MEG.0000000000002537. Epub 2023 Mar 18.

Abstract

Objectives: Evaluate the accuracy and agreement of two-dimensional shear-wave elastography (2D-SWE) LOGIQ-S8 with transient elastography in patients from Rio de Janeiro, Brazil.

Method: This retrospective study compared liver stiffness measurements (LSMs) using transient elastography (M and XL probes) and 2D-SWE GE-LOGIQ-S8 performed by a single experienced operator on the same day in 348 consecutive individuals with viral hepatitis or HIV infection. Suggestive and highly suggestive compensated-advanced chronic liver disease (c-ACLD) were defined by transient elastography-LSM ≥10 kPa and ≥15 kPa, respectively. Agreement between techniques and accuracy of 2D-SWE using transient elastography-M probe as the reference was assessed. Optimal cut-offs for 2D-SWE were identified using the maximal Youden index.

Results: Three hundred five patients [61.3% male, median age = 51 [interquartile range (IQR), 42-62] years, 24% with hepatitis C virus (HCV) ± HIV; 17% with hepatitis B virus (HBV) ± HIV; 31% were HIV mono-infected and 28% had HCV ± HIV post-sustained virological response] were included. The overall correlation (Spearman's ρ ) was moderate between 2D-SWE and transient elastography-M ( ρ = 0.639) and weak between 2D-SWE and transient elastography-XL ( ρ = 0.566). Agreements were strong ( ρ > 0.800) in people with HCV or HBV mono-infection, and poor in HIV mono-infected ( ρ > 0.400). Accuracy of 2D-SWE for transient elastography-M ≥ 10 kPa [area under the receiver operating characteristic (AUROC) = 0.91 (95% confidence interval [CI], 0.86-0.96); optimal cut-off = 6.4 kPa, sensitivity = 84% (95% CI, 72-92), specificity = 89% (95% CI, 84-92)] and for transient elastography-M ≥ 15 kPa [AUROC = 0.93 (95% CI, 0.88-0.98); optimal cut-off = 7.1 kPa; sensitivity = 91% (95% CI, 75-98), specificity = 89% (95% CI, 85-93)] were excellent.

Conclusion: 2D-SWE LOGIQ-S8 system had a good agreement with transient elastography and an excellent accuracy to identify individuals at high risk for c-ACLD.

MeSH terms

  • Brazil / epidemiology
  • Elasticity Imaging Techniques* / methods
  • Female
  • HIV Infections* / complications
  • Hepatitis B virus
  • Hepatitis B*
  • Hepatitis C*
  • Hepatitis, Chronic
  • Humans
  • Liver / diagnostic imaging
  • Liver Cirrhosis / diagnostic imaging
  • Liver Diseases*
  • Male
  • Middle Aged
  • Retrospective Studies