Diagnostic accuracy of AGILE3+ score for advanced fibrosis in patients with non-alcoholic fatty liver disease: A systematic review and meta-analysis
Citation
Dalbeni, A., Lombardi, R., Henrique, M., Zoncapè, M., Pennisi, G., Petta, S., Tateishi, R., Keklikkiran, C., Colecchia, A., Sacerdoti, D., Mantovani, A., & Ravaioli, F. (2023). Diagnostic accuracy of AGILE3+ score for advanced fibrosis in patients with non-alcoholic fatty liver disease: A systematic review and meta-analysis. Hepatology (Baltimore, Md.), 10.1097/HEP.0000000000000694. Advance online publication. https://doi.org/10.1097/HEP.0000000000000694Abstract
Background and aims: A simple non-invasive score, the Agile3+ score, combining liver stiffness measurement (LSM), aspartate aminotransferase/alanine aminotransferase ratio, platelet count, diabetes status, sex, and age, has been proposed for identification of advanced fibrosis in patients with suspected non-alcoholic fatty liver disease (NAFLD). We performed a systematic review and meta-analysis of observational studies to evaluate the diagnostic accuracy of the Agile 3+ score in identifying patients with NAFLD and advanced fibrosis. Recently, an International consensus changed the nomenclature of NAFLD into metabolic-associated steatotic liver disease (MASLD), so currently, the two terms are interchangeable.
Methods: We systematically searched MEDLINE, Ovid Embase, Scopus, and Cochrane Library electronic databases for full-text published articles in any language from the inception to the 24th of April 2023. We included original articles reporting data on the sensitivity and specificity of the Agile 3+ score, according to previously described rule-out (≤0.451) and rule-in (≥0.679) cut-offs.
Results: We included 6 observational studies (total 6955 participants) with biopsy-proven NAFLD (mean age 53 [SE 4] years, mean BMI 30.9 [SE 2.3] Kg/m2, 54.0% men, prevalence of diabetes 59.6%). The pooled prevalence of advanced fibrosis (≥F3) was 42.1%. By the rule-out cut-off, the overall sensitivity and specificity were 88% (95%CI 81-93%; I2=89.2%) and 65% (95%CI 54-75%; I2=97.6%), respectively. By the rule-in cut-off, the overall sensitivity and specificity were 68% (95%CI 57-78%; I2=91.1%) and 87% (95%CI 80-92%; I2=96.7%), respectively. Meta-regression analyses reported that the diagnostic accuracy was partly mediated by age (p<0.01), BMI (p<0.01), and, although not statistically significant, sex (p=0.06).
Conclusion: Our systematic review and meta-analysis suggest that Agile3+ accurately diagnoses NAFLD with advanced fibrosis and can identify patients eligible for biopsy and emerging pharmacotherapies.
