Can reference images improve interobserver agreement in reporting liver fibrosis?

Abstract

Staging of fibrosis in medical liver biopsies has inherent interobserver variability. There are a number of disease-specific scoring systems available. While recognising the importance of these scoring systems, there is scope to consider how concordance amongst histopathologists could be improved using a generic fibrosis staging system.Using virtual slides, we approached both specialist liver histopathologists and general histopathologists from the UK to assess the degree of fibrosis against a proposed four-tiered reporting system. Example reference images were then produced and distributed to the same responders who were asked to rate a second set of slides to assess if the use of reference images improved concordance between pathologists.The use of reference images eliminated spread across three categories (from 15% to 0%). Overall, agreement was already good; our study showed an improved agreement amongst all participants for percentage agreement (67.79% to 70.08%) and interobserver agreement improved (Fleiss' Kappa 0.55 to 0.59).

Feasibility and agreement of stiffness measurements using gradient-echo and spin-echo MR elastography sequences in unselected patients undergoing liver MRI

“…1,2 Although liver biopsy is the current gold-standard for the diagnosis and staging of liver iron overload, liver fat deposition, and fibrosis, it has limitations, most related to sampling errors and only moderate interreader agreement. [3][4][5][6] Percutaneous liver biopsy is also an invasive procedure with potential risks and unsuitable for consecutive routine follow-up assessment. 4 To address liver pathologies non-invasively, novel imaging quantification techniques have been the focus of extensive research.…”

Prognostic Role of Liver Biopsy in Patients With Severe Indeterminate Acute Hepatitis

“…To our knowledge, the reproducibility of the histological definition of multilobular necrosis in the prognostication of indeterminate hepatitis has not been formally tested. The use of reference images has been proposed to facilitate the reproducibility of histological interpretation 18 and could be applied to this setting.…”

“…Nineteen (44.2%) patients who recovered spontaneously were defined as survivors (Table 3). The survivors had longer median duration of jaundice prior to biopsy than nonsurvivors (28 [IQR, 14-49] days vs 14 [IQR, [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26] days; P ¼ .002). The median duration from the onset of jaundice to death or transplantation was 33.5 (IQR, 20.3-46.8) days.…”

Magnetic resonance elastography SE-EPI vs GRE sequences at 3T in a pediatric population with liver disease

“…Liver fibrosis is multifactorial and early diagnosis and staging are important factors contributing to a better outcome [6]. Liver biopsy is unsuitable as a routine follow-up assessment [6][7][8][9][10][11][12][13][14]. Studies have reported the risk of hospitalization related to a liver biopsy in the range from 1% to 5%, with an additional 0.57% risk of severe complications, and reported mortality rates ranging from 1:1000 to 1:10000 [15][16][17][18][19].…”