Association of platelet count, fibrosis-4, and aspartate aminotransferase-to-platelet ratio index with the development and severity of esophageal varices in patients with liver cirrhosis / 临床肝胆病杂志

ABSTRACT

Objective To investigate the clinical value of platelet count (PLT), fibrosis-4 (FIB-4), and aspartate aminotransferase-to-platelet ratio index (APRI) in predicting the development and classification of esophageal varices (EVs) . Methods A retrospective analysis was performed for the clinical data of 163 patients with liver cirrhosis who visited Department of Hepatology in The First Hospital of Jilin University from January 2012 to December 2015. All patients underwent upper gastrointestinal endoscopy within one week after admission. According to PLT, the patients were divided into PLT ≤50 × 109/L group with 27 patients, PLT 50 × 109/L-≤100 × 109/L group with 84 patients, PLT 100 × 109/L-≤150 × 109/L group with 26 patients, and PLT＞ 150 × 109/L group with 26 patients. The Child-Pugh score, FIB-4, and APRI were recorded, and the association of PLT, FIB-4, and APRI with EVs was analyzed. A one-way analysis of variance was used for comparison of normally distributed continuous data between multiple groups, and the LSD-t test was used for further comparison between any two groups; the Kruskall-Wallis H test was used for comparison of non-normally distributed continuous data between multiple groups, and the Dunn-Bonferroni test was used for further comparison between any two groups. The chi-square test was used for comparison of categorical data between groups, and the Fisher's test was used for the data which did not meet the conditions of the chi-square test. Spearman rank correlation was used to investigate the correlation of PLT, FIB-4, and APRI with EVs. The receiver operating characteristic (ROC) curve and the area under the ROC curve (AUC) were used to evaluate the values of PLT, FIB-4, and APRI in the diagnosis of EVs. Results Of all 163 patients with liver cirrhosis, 114 developed EVs and 49 did not experience EVs. There was a significant difference in the incidence rate of EVs between the cirrhotic patients with different PLTs (χ2= 27. 36, P ＜ 0. 001) . Compared with those with PLT≤150 × 109/L, the cirrhotic patients with PLT＞ 150 × 109/L had a significantly lower proportion of patients with EVs (34. 6％ vs 76. 6％, P ＜ 0. 001) or severe EVs (3. 8％ vs 46. 7％, P ＜ 0. 001) . There were significant differences in FIB-4 and APRI between the patients with different PLTs (χ2= 102. 58 and 57. 02, both P ＜ 0. 001) . PLT was negatively correlated with the degree of EVs (r =-0. 491, P ＜ 0. 001), and FIB-4 and APRI were positively correlated with the degree of EVs (r = 0. 460 and 0. 325, both P ＜0. 001) . PLT had an AUC of 0. 739 and an accuracy of 75. 4％ in predicting the presence or absence of EVs, while FIB-4 had an AUC of 0. 732 and an accuracy of 71. 2％ and APRI had an AUC of 0. 651 and an accuracy of 72. 4％ . PLT had an AUC of 0. 763 and an accuracy of 69. 3％ in predicting the presence or absence of severe EVs, while FIB-4 had an AUC of 0. 742 and an accuracy of 67. 5％ and APRI had an AUC of 0. 676 and an accuracy of 66. 3％ . Conclusion Cirrhotic patients with PLT＞ 150 × 109/L have a significant reduction in the risk of EVs or severe EVs. PLT, FIB-4, and APRI are correlated with the degree of EVs in cirrhotic patients, but they cannot fully replace gastroscopy.