الفهرس 
COVEROnly 14 pages are availabe for public view
 |  | from | 172 |  |  |
| | | from | 172 | | |
Abstract
Endoscopy is the gold standard procedure for diagnosing EV. However it is not cost-effective to screen all cirrhotic patients by endoscopy for detection of varices. Also, many cirrhotic patients are non-compliant, refuse repeated screening endoscopy.
The aim of this study was to assess insulin resistence (IR) as a non- invasive predictor of esophageal varices in patients with liver cirrhosis.
The present cross-sectional study was conducted on 80 patients with liver cirrhosis.
All studied patients were subjected to complete history taking, thorough clinical examination, laboratory investigations, abdominal ultrasonography and diagnostic upper GIT endoscopy.
According to the result of upper GIT endoscopy, patients were classified into two groups:
• Group I: included 60 patients (75%) with evidence of esophageal varices (EV).
• Group II: included 20 patients (25%) with no evidence of esophageal varices.
Group I was further classified into:
• Group Ia: included 11 patients (18.3%) with small EV (EV grades I, I-II, II).
• Group Ib: included 49 patients (81.7%) with large EV (EV grades II-III, III, III-IV)
On analyzing laboratory data, patients with varices had lower serum albumin, more elevated serum bilirubin and lower platelets count than those without varices with a highly significant difference between the two groups. On the contrary, serum ALT and AST levels and prothrombin time showed no significant difference.
The frequency of varices in the patients of our study increases with the increase of their Child score.
Regarding abdominal ultrasonographic findings in our study, there was a highly significant increase in portal vein diameter in patients with varices in comparison to those without varices.
The platelet count/ spleen diameter (mm) ratio in the present study was highly significantly decreased in patients with varices in comparison to patients without.
In our study, it was found that patients with varices had higher HOMA-IR than those without varices with highly significant difference between the two groups. Also patients with large varices had higher HOMA-IR than those with small varices with highly significant difference between the two subgroups.
In our study, ROC curve analysis identified HOMA-IR score of greater than 3.85 (sensitivity, 95%; specificity, 95%) as the best cutoff for predicting the presence of EV. Above this value we can, non invasively, predict presence of esophageal varices.
In our study, the presence or absence of DM showed no significant difference between patients with varices and those without.
In our study, we tested the performance of HOMA-IR, as a non invasive predictor of esophageal varices, in non diabetic patients, we plotted ROC Curve, the CUT-OFF value was 3.5 (sensitivity, 95%; specificity, 93).
By multivariate analysis, we found the most predictable variables for diagnosis of EV to be HOMA-IR, S.Albumin and Splenic size (cm).