Non-Invasive Diagnostic Test for Advanced Fibrosis in Adolescents With Non-Alcoholic Fatty Liver Disease

Abstract

IntroductionNon-alcoholic fatty liver disease (NAFLD) is a multifaceted disease that includes a wide spectrum of liver damage. The presence and the degree of fibrosis are considered important factors for the prognosis of NAFLD and in predicting the risk of developing cirrhosis. Our aim was to evaluate the usefulness of four fibrosis scores (aspartate aminotransferase/Platelet Index [APRI], FIB-4, NAFLD Fibrosis Score [NFS], and Hepamet) in predicting different degrees of fibrosis among children with biopsy-proven NAFLD.MethodsAbout 286 adolescents [mean age 14.3 years ± 2.5; 154 (53.6%) males], referred between January 2014 and December 2019, with biopsy-proven NAFLD were enrolled.ResultsAbout 173 (60.4%) patients presented fibrosis at histological analysis. In particular: 140 (49.3%) patients had F = 1, 31 (10.8%), had F = 2 and 2 (0.66%) had F = 3. APRI (AUROC 0.619, 95% CI 0.556–0.679) and Hepamet (AUROC 0.778, 95% CI 0.722–0.828) scores had significant (p < 0.001) accuracy to distinguish subjects with fibrosis; while NFS and FIB-4 had not. APRI had a positive predictive value (PPV) of 62.77% (95% CI 57.96–67.35) and an negative predictive value (NPV) of 52.01% (95% CI 46.54–57.43); Hepamet a PPV of 63.24% (95% CI 59.95–66.41) and an NPV of 61.29% (52.9–69.01).ConclusionsOur study showed that Hepamet and APRI perform better than NFS and FIB-4 for identifying fibrosis in patients with NAFLD, but do not have PPVs so high to be considered diagnostic. Therefore, they cannot be employed, in children, for a certain diagnosis of fibrosis or its progression and cannot replace liver biopsy as the gold diagnostic standard. It is, therefore, necessary to continue to research and develop new markers of exclusive fibrosis.

Noninvasive scores are poorly predictive of histological fibrosis in paediatric fatty liver disease

“…For PNFI, the data included in our cohort revealed the inability to replicate the performance recently reported by Mosca et al 27 Of note, ALT levels in the study of Mosca et al (median 42 IU/L; IQR 21–58 IU/L) were lower in patients without fibrosis as compared to the present aggregated cohort (median: 69 IU/L; IQR: 51–114 IU/L), suggesting different patient profiles in the F0 group. Also, the PNFI score for F0 was lower in the study of Mosca et al, (median 4.6, IQR 1.1–7) as compared with our result (median: 9.8 IQR 7.0‐10.0).…”

“…For the detection of moderate fibrosis, the highest AUROC of 0.697 was afforded by APRI, which was comparable to previous studies 24,27 . Yet, diagnostic utility of APRI was rendered insufficient by low test sensitivity.…”

“…42,43 For PNFI, the data included in our cohort revealed the inability to replicate the performance recently reported by Mosca et al 27 For the detection of moderate fibrosis, the highest AUROC of 0.697 was afforded by APRI, which was comparable to previous studies. 24,27 Yet, diagnostic utility of APRI was rendered insufficient by low test sensitivity. Here, the highest sensitivity (67%) out of all significantly associated tests was reached by the sole use of ALT.…”

“…More recently, the paediatric NAFLD fibrosis index (PNFI) and the paediatric NAFLD fibrosis score (PNFS) were developed 22,23 . Previous validation studies in independent paediatric NAFLD cohorts led to inconclusive results regarding accuracy of the aforementioned fibrosis score systems, yet were limited due to relatively small study size or unequal distribution of fibrosis stages 24–27 . Therefore, the aim of this study was to apply published noninvasive fibrosis scores to a large European paediatric patient cohort with biopsy‐proven NAFLD to assess their diagnostic accuracy in ruling in, or ruling out specific degrees of fibrosis (Table S1), and therefore evaluate their applicability in defined scenarios of clinical practice.…”

Difficulties in diagnosis of alcoholic fatty liver disease and metabolically associated fatty liver disease

“…Для оцінки ймовірності наявності та визначення ступеню стеатозу чи (або) фіброзу печінки широко використовуються ще інші розрахункові коефіцієнти, що поєднує різні сироваткові біомаркери та метаболічні параметри [10]. Наприклад показник наявності жиру в печінці (FLI) або шкала наявності жиру в печінці (NAFLD liver fat score), існують також показники для ідентифікації фіброзу печінки: NAFLD liver fibrosis score, APRI та FIB4.…”