Diagnostic accuracy of transient elastography in diagnosing clinically significant portal hypertension in patients with chronic liver disease: a systematic review and meta-analysis

Purpose Liver stiffness measurement using transient elastography (TE-LSM) is a promising noninvasive alternative to hepatic venous pressure gradient (HVPG) for diagnosing clinically significant portal hypertension (CSPH). However, previous studies have yielded conflicting results. We evaluated the correlation between TE-LSM and HVPG and the performance of TE-LSM in diagnosing CSPH (HVPG >= 10 mmHg). Methods We conducted a systematic review and meta-analysis by searching PubMed and Scopus databases for relevant literature evaluating the clinical usefulness of TE for diagnosing CSPH in patients with chronic liver disease. Results Twenty-six studies (4337 patients with valid TE and HVPG) met our inclusion criteria. The median correlation coefficient of TE with HVPG was 0.70 (range 0.36-0.86). The weighted mean of optimal cut-off of liver stiffness value for diagnosing CSPH was 22.8 kPa (95% CI 22.7-23.0 kPa). The summary sensitivity and specificity were 79% (95% CI 74-84%) and 88% (95% CI 84-91%), respectively. The area under the hierarchical summary receiver operating characteristic (HSROC) curve was 0.91 (95% CI 0.88-0.93) according to the bivariate model. One limitation of the study was significant heterogeneity in the results of summary sensitivity and specificity (I-2 83 and 74%, respectively). The heterogeneity could be explained by the variable liver stiffness cut-offs used in studies. The meta-regression plot revealed that as the optimal cut-off increased, the sensitivity decreased, the specificity increased, and vice versa. Conclusions Liver stiffness measurement using TE correlates well with HVPG, and a liver stiffness cut-off value of 22.8 kPa shows a high accuracy for diagnosing CSPH. Thus, use of TE should be integrated into clinical practice for noninvasive diagnosis of CSPH.