Establishment and validation of a diagnostic nomogram for significant histopathologic changes of hepatic injury in HBV-infected patients

Abstract

Background: Significant histopathologic changes of hepatic injury (SHCHI) play a decisive role in evaluating the condition and initiating antiviral in HBV-infected patients, especially those with normal or mildly elevated alanine transaminase levels. Considering that non-invasive methods were established through experience with chronic hepatitis C, the aim of this study was to establish and verify a nomogram based on hepatitis B for diagnosing SHCHI.Methods: Three hundred eighty-four patients who fulfilled requirements for participation were randomly assigned to training cohort (n=270) and validation cohort (n=114) according to 7:3. The selection criteria for clinical factors were based on the previous research papers. SHCHI was subgrouped as followed: grade ≥ G2 inflammation and/or stage ≥ S2 fibrosis. The predictive accuracy and discriminative ability of nomogram were determined by a concordance index (C-index), calibration curve and the area under the receiveroperating characteristic curve (AUROC). We also compared diagnostic value of nomogram with model for AST-to-PLT ratio index (APRI) score and model for Fibrosis-4 (FIB-4) score. Results: Two hundred and two patients (74.44%) and 87 patients (76.32%) were diagnosed as SHCHI, in the training and validation cohort. Logistic regression analysis illustrated that hepatitis B e antigen (HBeAg), aspartate aminotransferase (AST), γ-glutamyl transferase (GGT), and prothrombin time (PT) all independently served as risk factors for SHCHI (P<0.05) and were thus utilized to create the nomogram. The nomogram had well-fitted calibration curves and attained excellent concordance indices of 0.80 and 0.75. The sensitivity of nomogram in the diagnosis of SHCHI was 79.7%, the specificity was 68.1%. The area under the curve {AUC; 0.80 [95% confidence interval (CI): 0.74-0.86]} for diagnosing SHCHI by the nomogram was greater in comparison to that of APRI [0.78 (95% CI: 0.71-0.84)], and FIB-4 [0.76 (95% CI: 0.69-0.82)]. Patients with nomogram scores less than 119 were considered to have a lower risk of SHCHI. Conclusions: The constructed nomogram is suitable to serve as a SHCHI screening tool in chronic HBV-infected patients. But the dependability of the nomogram will necessitate further confirmation in a prospective study and further external validation is needed.