ABSTRACT
Objective: To study the predictive power of noninvasive investigative parameters (clinical, biochemical, radiological) for detection of esophageal varices in patients with portal hypertension (PHT) as compared to invasive parameters (upper gastrointestinal endoscopy). Materials and methods: Fifty patients with PHT, between May 2008 to September 2010, were studied. Those who had decompensated liver diseases, HIV, hepatocellular carcinoma, metastasis in liver, parentral drug addiction, chronic febrile illness, H/O treatment taken for PHT in the form of surgery or endoscopic bending or sclerotherapy were excluded. Detailed clinical history was taken and physical examination was done. All patients underwent the required hematological, biochemical, radiological, endoscopic and histopathological investigations. Results: Platelet count/splenic size showed a significant correlation between presence or absence and grade of esophageal varices (p < 0.00015). If a cut-off value of 1,000/cu mm is taken, then 87.5% (35/40) patients with esophageal varices have ratio <1,000 while 20% (2/10) of patients with ratio <1,000 did not have any varices. It was also observed that lower the ratio, higher the grade of varices. Conclusion: Asymptomatic esophageal varices, which is quite common, can be easily diagnosed with invasive endoscopy or otherwise can be suspected with noninvasive predictors like platelet/spleen size ratio in our country, where financial constraint is a major problem for investigations like endoscopy.