Value of noninvasive liver fibrosis markers in predicting high-risk gastroesophageal varices in patients with liver cirrhosis

ObjectiveTo investigate the clinical value of routine liver fibrosis markers in the noninvasive diagnosis of high-risk gastroesophageal varices (HRGOV) in patients with liver cirrhosis. MethodsA total of 165 patients with liver cirrhosis who underwent gastroscopy and routine laboratory examinations in Nanfang Hospital, Southern Medical University, from January to December 2018 were enrolled, and the presence or absence of HRGOV was determined based on the gold standard of gastroscopy with reference to Child-Pugh score. The independent samples t-test or the Mann-Whitney U test was used for comparison of continuous data between two groups, and the chi-square test was used for comparison of categorical data between two groups. The area under the receiver operating characteristic curve (AUC) was calculated and compared, and likelihood ratio was used to confirm diagnosis and exclude cut-off values. ResultsOf all 165 patients, 83 (50.3%) patients were diagnosed with HRGOV. There were significant differences between the HRGOV group and the non-HRGOV group in albumin (t=-5.118, P＜0.001), platelet count (Z=-5.409，P＜0.001), proportion of male patients (χ2=3.883, P=0.049), prothrombin time (Z=-4.433, P＜0.001), prothrombin time activity (Z=-4.447, P＜0.001), international normalized ratio (Z=-4.426, P＜0.001), white blood cell count (Z=-2.371, P=0.018), spleen thickness (Z=-6.296, P＜0.001), aspartate aminotransferase-to-platelet ratio index (APRI) (Z=-3.409, P=0.001), fibrosis index based on four factors (FIB-4) (Z=-4.494, P＜0.001), gamma-glutamyl transpeptidase-to-platelet ratio (GPR) (Z=-2.377, P=0.017), red blood cell distribution width-to-platelet ratio (RPR) (Z=-5.345, P＜0.001), Easy Liver Fibrosis Test (eLIFT)(t=3.170, P=0.002), gastroscopic grade (χ2=131.714, P＜0001), and Child-Pugh class (χ2=30.663, P＜0.001). Spleen thickness on ultrasound, platelet count, RPR, FIB-4, APRI, eLIFT, and GPR had an AUC of 0.785, 0.747, 0.744, 0.703, 0.652, 0.622, and 0.606, respectively, in the diagnosis of HRGOV, and there was no significant difference between the models with an AUC of ＞0.7 (P＞0.05). With spleen thickness ＜29.5 mm for exclusion and spleen thickness ≥53.0 mm for confirmed diagnosis, 51 patients (30.9%) had no need to undergo gastroscopy, with a diagnostic accuracy of 94.1%; with RPR ＜0.067 for exclusion and RPR ≥0.480 for confirmed diagnosis, 19 patients (115%) had no need to undergo gastroscopy, with a diagnostic accuracy of 89.5%. Spleen thickness followed by RPR was used to confirm the presence or absence of HRGOV in 63 patients (38.2%) and achieved a diagnostic accuracy of 92.1%. ConclusionSpleen thickness, RPR, platelet count, and FIB-4 had moderate efficiency in the diagnosis of HRGOV, among which spleen thickness has the highest diagnostic efficiency, followed by RPR, platelet count, and FIB-4. Spleen thickness followed by RPR can be used for the screening for HRGOV in liver cirrhosis patients, helping some patients avoid gastroscopy.