Jejunal perforation: a complication of colonoscopy.

Perforation of a hollow viscus following fibreoptic endoscopy is rare. Perforation following colonoscopy has been reported to occur in the large intestine. We report a case. Small bowel perforation subsequent to colonoscopy.

Endoscopic variceal ligation: an effective method of variceal obliteration.

Endoscopic variceal ligation was performed in 20 patients with portal hypertension of varied etiology. Variceal obliteration was achieved in 18 patients (90%) while recurrence of bleeding occurred in two patients (10%). The average number of bands required was four per patient and average number of sessions required for variceal obliteration was two. We found the procedure to be cheap, safe and effective in achieving early variceal obliteration.

Can sonography replace splenoportovenography in evaluation of patients with portal hypertension?

Real-time sonography and splenoportovenography were compared in 17 patients with portal hypertension for their relative efficacy and limitations with respect to diameters of portal vessels, visualization of collaterals and demonstration of portal vein occlusion. Sonography was able to diagnose portal vein thrombosis and to differentiate an occluded portal vessel from a patent portal vessel non-visualized due to hepatofugal blood flow in the presence of intrahepatic obstruction. However, sonography had limitations in demonstrating venous structures in the presence of excessive bowel gas or fat, and did not provide the flow patterns and the complete picture of the portal vasculature in a single setting. We conclude that the two procedures are complementary to each other, and if combined in patients with portal hypertension, the portal venous system can be evaluated more thoroughly for surgical treatment.

Differential diagnosis of ascitic fluid: evaluation and comparison of various biochemical criteria with a special reference to serum ascites albumin concentration gradient and its relation to portal pressure.

Among various biochemical indices measured in 93 patients with ascites, ascitic LDH estimation was proved to be indiscreminatory, while ascites/serum LDH ratio has shown a diagnostic accuracy of 85 per cent. Ascitic total protein levels and ascites/serum total protein ratio (accuracy rates of 72 and 77% respectively) were limited, especially in differentiating the ascites due to heart failure. Serum ascites albumin gradient, showed a strong correlation to portal pressure (r, + 0.83 + 0.88), and was found to be the best diagnostic index (with an overall accuracy of 97 per cent) in distinguishing the 'transudative' from 'exudative' ascites. However, no index could discreminate the 'mixed' cases and provide the etiological diagnosis of the ascites.