Transient elastography (TE) is the reference method to obtain liver stiffness measurements (LSM), but no results are obtained in 3.1% and unreliable in 15.8%. We assessed the applicability and diagnostic accuracy of TE re-evaluation using M and XL probes. From March 2011 to April 2012 868 LSM were performed with the M probe by trained operators (50-500 studies) (LSM1). Measurements were categorized as inadequate (no values or ratio <60% and/or IQR/LSM >30%) or adequate. Inadequate LSM1 were re-evaluated by experienced operators (>500 explorations) (LSM2) and inadequate LSM2 using XL probe (LSMXL). Inadequate LSM1 were obtained in 187 (21.5%) patients, IQR/LSM >30% in 97 (51%), ratio <60% in 24 (13%) and TE failed to obtain a measurement in 67 (36%). LSM2 achieved adequate registers in 123 (70%) of 175 registers previously considered as inadequate. Independent variables (OR, 95%CI) related to inadequate LSM1 were body mass index (1.11, 1.04-1.18), abdominal circumference (1.03, 1.01-1.06) and age (1.03, 1.01-1.04) and to inadequate LSM2 were skin-capsule distance (1.21, 1.09-1.34) and abdominal circumference (1.05, 1.01-1.10). The diagnostic accuracy (AUROC) to identify significant fibrosis improved from 0.89 (LSM1) to 0.91 (LSM2) (P = 0.046) in 334 patients with liver biopsy or clinically significant portal hypertension. A third evaluation (LSMXL) obtained adequate registers in 41 (93%) of 44 patients with inadequate LSM2. Operator experience increases the applicability and diagnostic accuracy of TE. The XL probe may be recommended for patients with inadequate values obtained by experienced operators using the M probe. http://clinicaltrials.gov (NCT01900808).
Keywords: fibrosis; liver; noninvasive; stiffness.
© 2014 John Wiley & Sons Ltd.