[LOW SPECIFICITY OF PLATELET TO SPLEEN RATIO FOR NONINVASIVE PREDIC- TION AND CHARACTERIZATION OF ESOPHAGEAL VARICES IN PATIENTS WITH ALCOHOLIC LIVER CIRRHOSIS].

Diagnosis of esophageal varices (EV) is based upon endoscopic examination, which is a rather unpleasant method that carries a certain risk of complications. For that reason, efforts have been made to develop noninvasive methods for characterization of EV. The aim of this study was to explore the value of platelet count to spleen size ratio (PSR) for noninvasive prediction and characterization of EV in patients with alcoholic liver cirrhosis (ALC). One hundred and seventeen patients (20 females and 97 males, mean age 60.7) with ALC were included in our research. All patients underwent endoscopic examination upon which the EV were classified as small (< 5 mm), large (> 5 mm), or absent. Spleen size (bipolar diameter in mm) was assessed by ultrasound. Platelet count to spleen diameter ratio was calculated and the values obtained were compared to the presence, size and risk of bleeding from EV as defined by endoscopy. No significant difference in PSR could be found between patients without and with EV (1.341 ± 0.725 vs. 1.053 ± 0.636, respectively; p = 0.06). The PSR was significantly different between the patients with small and large EV (1.103 ± 0.689 vs. 0.876 ± 0.314; p < 0.05) with a cut-off value of 1.141 (sensitivity 94.7%, specificity 38.2%, AUROC = 0.656; p = 0.042). The value of PSR below 1.182 pointed to patients at risk from variceal bleeding with 91.7% sensitivity and 38.5% specificity (AUROC = 0.625, p = 0.035). Based on our results, it is not possible to recommend the use of PSR as the exclusive noninvasive indicator for the presence, size and bleeding risk from EV due to its low specificity for these categories in patients with ALC.