Magnitude dependent discordance in liver stiffness measurements using elastography point quantification with transient elastography as the reference test

Abstract

Objectives: To investigate diagnostic performance of point shear wave elastography by elastography point quantification (ElastPQ) for non-invasive assessment of liver fibrosis in patients with chronic liver diseases (CLD).

Methods: Liver stiffness measurement (LSM) by transient elastography (TE) and ElastPQ was performed in patients with CLD and healthy volunteers. The stage of liver fibrosis was defined by TE which served as the reference. We compared two methods by using correlation, area under the receiver operating characteristics curve (AUC) analysis, Bland and Altman plot and Passing-Bablok regression.

Results: A total of 185 subjects (20 healthy volunteers and 165 patients with CLD (128 non-alcoholic fatty liver disease), 83 (44.9%) females, median age 53 years, BMI 27.3 kg/m²) were evaluated. There were 24.3%, 13.5% and 11.4% patients in ≥ F2, ≥ F3 and F4 stage, respectively. The best performing cutoff LSM values by ElastPQ were 5.5 kPa for F ≥ 2 (AUC = 0.96), 8.1 kPa for F ≥ 3 (AUC = 0.98) and 9.9 kPa for F4 (AUC = 0.98). Mean (SD) difference between TE and ElastPQ measurements was 0.98 (3.27) kPa (95% CI 0.51-1.45, range 4.99-21.60 kPa). Two methods correlated significantly (r = 0.86; p < 0.001), yet Bland and Altman plot demonstrated difference between measurements, especially with TE values > 10 kPa. Passing and Bablok regression analysis yielded significant constant and proportional difference between ElastPQ and TE.

Conclusion: ElastPQ is reliable method for assessment of liver fibrosis but LSM values are not interchangeable with TE, especially above 10 kPa. Diagnostic performance of ElastPQ for sub-classification of patients with compensated advanced chronic liver disease should therefore be furtherly investigated.

Key points: • ElastPQ appears to be reliable method for assessment of liver fibrosis, with data presented here mostly applicable to NAFLD. • LSM values produced by TE and ElastPQ are NOT interchangeable-in values < 10 kPa, they are similar, but in values > 10 kPa, they appear to be increasingly and significantly different. • Diagnostic performance of ElastPQ for sub-classification of patients with compensated advanced chronic liver disease should be furtherly investigated.

Keywords: Diagnostic performance; Liver fibrosis; Liver stiffness measurement; Non-alcoholic fatty liver disease.