Diagnostic value of alcoholic liver disease (ALD)/nonalcoholic fatty liver disease (NAFLD) index combined with γ-glutamyl transferase in differentiating ALD and NAFLD

Abstract

Background/aims: This study aimed to verify the reliability of the alcoholic liver disease (ALD)/nonalcoholic fatty liver disease (NAFLD) index (ANI) for distinguishing ALD in patients with hepatic steatosis from NAFLD, and to investigate whether ANI combined with γ-glutamyl transferase (GGT) would enhance the accuracy of diagnosis in China.

Methods: A hundred thirty-nine cases of fatty liver disease (FLD) were divided into two groups of ALD and NAFLD. The ANI was calculated with an online calculator. All indicators and ANI values were analyzed using statistical methods.

Results: ANI was significantly higher in patients with ALD than in those with NAFLD (7.11 ± 5.77 vs. -3.09 ± 3.89, p < 0.001). With a cut-off value of -0.22, the sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC) of diagnosed ALD cases was 87.1%, 92.5%, and 0.934 (95% confidence interval [CI], 0.879 to 0.969), respectively. The corresponding values for aspartate aminotransferase (AST)/alanine transaminase (ALT), mean corpuscular volume (MCV), and GGT were 75.29%, 72.94%, and 0.826 (95% CI, 0.752 to 0.885); 94.34%, 83.02%, and 0.814 (95% CI, 0.739 to 0.875) and 80.23%, 79.25%, and 0.815 (95% CI, 0.740 to 0.876), respectively. ANI AUROC was significantly higher than the AST/ALT, MCV, or GGT AUROCs (all p < 0.001), moreover, ANI showed better diagnostic performance. The combination of ANI and GGT showed a better AUROC than ANI alone (0.976 vs. 0.934, p = 0.016). The difference in AUROCs between AST/ALT, MCV, and GGT was not statistically significant (all p > 0.05).

Conclusions: ANI can help distinguish ALD from NAFLD with high accuracy; when ANI was combined with GGT, its effectiveness improved further.

Keywords: ALD/NAFLD index; Diagnosis, differential; Gamma-glutamyltransferase; Liver diseases, alcoholic; Non-alcoholic fatty liver disease.

Figure 1.

(A) Nonalcoholic fatty liver disease (NAFLD) characterized by macrovesicular fatty changes and the hepatic lobular infiltrates are composed predominantly of lymphocytes and macrophages, while (B) a clinicopathologic lesion of nearly pure microvesicular steatosis and neutrophilic infiltration in the hepatic lobules in alcoholic liver disease, are rarely detected in livers of patients with NAFLD (H&E, ×200).

Figure 2.

ANI values of patients in alcoholic liver disease (ALD) and nonalcoholic fatty liver disease (NAFLD) groups. The difference of ANI values between ALD and NAFLD groups was statistically significant (t = –11.86, p < 0.001). ANI, ALD/NAFLD index.

Figure 3.

ANI values of patients in alcoholic liver disease (ALD) and nonalcoholic fatty liver disease (NAFLD) subgroups. Comparison of area under the receiver operating characteristic curve no statistically significant difference between nonalcoholic simple fatty liver (NAFL) and nonalcoholic steatohepatitis (NASH; Z = –0.48, p = 0.961), and so was between alcoholic fatty liver (AFL) and alcoholic steatohepatitis (ASH; Z = –1.23, p = 0.218). ANI, ALD/NAFLD index.

Figure 4.

The receiver operating characteristic curve of ANI, alanine aminotransferase/aspartate aminotransferase (AST/ALT), mean corpuscular volume (MCV), γ-glutamyl transpeptidase (GGT), and ANI combined with GGT for diagnosis of alcoholic liver disease (ALD). ANI, ALD/nonalcoholic fatty liver disease (NAFLD) index.

Figure 5.

The distribution of alcoholic liver disease (ALD) and nonalcoholic fatty liver disease (NAFLD) patients, when the cut-off of ANI toke –0.22. ANI, ALD/NAFLD index.