Endoscopic ultrasonography classification of gastric varices in patients with cirrhosis. Comparison with endoscopic findings.

Endoscopic ultrasonography was performed in 80 cirrhotic patients and 50 control subjects. The aim of the study was to describe and classify the vascular abnormalities of the gastric wall as observed on echoendoscopy in cirrhotic patients and to compare the data with endoscopy. The size of gastric varices was classified into 3 grades: grade 0 (none), grade 1 (small or non-confluent varices < 5 mm), grade 2 (large or confluent varices > or = 5 mm). The abnormalities of the gastric wall were classified into 3 grades: 0 (none), grade 1 (thickening and brilliance of the third hyperechogenic layer with or without fine internal anechogenic structures), grade 2 (visible vessels in the third layer which deform the entire wall, with penetrating varices). Endoscopy provides better visualization of esophageal varices than echoendoscopy but does not detect gastric varices and the first signs of portal hypertension as well. In conclusion echoendoscopy should be a routine examination for screening portal hypertension in cirrhotic patients and could be used in therapeutic follow-up.

[Reversal of portal blood flow in cirrhosis. Clinical, endoscopic and ultrasound endoscopic correlations in 72 patients]. / Inversion du débit sanguin portal au cours de la cirrhose. Corrélations cliniques, endoscopiques et échoendoscopiques chez 72 patients.

From January to December 1991, the portal venous system was evaluated by Doppler ultrasonography in 72 patients with liver cirrhosis. The objectives of this study were to evaluate the prevalence of spontaneous reversal of blood flow in the portal vein and to assess the relationship between Doppler ultrasound investigation and clinical, biochemical, endoscopic (70 patients), and endosonographic (44 patients) features. Reversed flow was quite frequent (alternating: 17%, permanent: 22%) and its prevalence did not differ in relation to age, sex, serum gammaglobulin concentration and Child-Pugh class. In patients with reversed portal venous flow, the prevalence of hepatic encephalopathy was higher (39% vs 13.5%, P < 0.05), but the prevalence of esophageal or gastric varices was not related to that pattern. Endosonography detected gastric wall abnormalities in a higher proportion of patients with reversed portal flow than in patients without it (P < 0.05). This study suggests that reversal of flow in the portal vein could play a role in the development of the gastric wall abnormalities in liver cirrhosis, which are detected by endosonography but not by endoscopy.

[Myocardial infarction and rectocolonic polyps]. / Infarctus du myocarde et polypes rectocoliques.

The prevalence of polyps in the rectum and sigmoid colon was estimated in 103 patients with recent myocardial infarction and 200 controls. All patients were asymptomatic and older than 35 years. A flexible proctosigmoidoscopy was performed. One or more adenomatous polyps were found in 19.6 p. 100 of patients with myocardial infarction and in 16.2 p. 100 of controls (difference not statistically significant). In males, the odds ratio for adenomatous polyps was 0.92 (confidence limits, 0.43-1.93). Acceptability of the flexible rectosigmoidoscopy was excellent in controls and poor in patients with myocardial infarction (1.0 p. 100 and 26.4 p. 100 of patients respectively refused this procedure). Tolerance, bowel preparation and the length of the explored rectosigmoid were not different. There was no statistically significant difference in the site, number or size of polyps. This study shows a high prevalence of polyps in patients with myocardial infarction and control groups although not statistically different. Consequently, a screening procedure for polyps is not indicated in patients with myocardial infarction.

[Endoscopic excision of malignant colorectal polyps. Study of a series of 82 cases]. / Exérèse endoscopique des polypes malins rectocoliques. Etude d'une série de 82 cas.

The management of malignant colonic polyps removed by endoscopic polypectomy is a controversial subject. We reported a series of 81 patients with 82 malignant polyps removed by endoscopic polypectomy between 1977 and 1984. 15 polyps contained carcinoma in situ and were treated by endoscopic polypectomy alone. 36 polyps contained superficial cancer; 35 were treated by EP alone; 1 was treated by endoscopic polypectomy and colectomy. 26 polyps contained invasive carcinoma and 2 were classified as polypoid adenocarcinomas. 19 had clear resection margins at polypectomy and seven had involved resection margins: 17 were treated by endoscopic polypectomy alone, 9 were treated by endoscopic polypectomy and colectomy. The patients were followed with colonoscopy. Follow-up has been 6 to 74 months (mean 32 months). The patients whose polyps were treated by endoscopic polypectomy alone, had had no evidence of recurrent tumor at the polypectomy site. Of the 12 patients undergoing colectomy, 4 had residual tumor at the polypectomy site. No patients had involved lymph nodes. There was no evidence of recurrence in any of these cases. Polyps containing contained carcinoma in situ and superficial cancer, polyps containing invasive carcinoma and clear resection margins can be treated with endoscopic polypectomy alone. Polyps with invasive carcinoma and involved resection margins should undergo segmental colonic resection. This approach must be weighed against the age of the patient, the medical status and the morbidity and mortality of the surgical procedure.

[Value of ultrasonography in the diagnosis of cirrhoses. Prospective study of 128 patients]. / Valeur de l'ultrasonographie dans le diagnostic des cirrhoses. Etude prospective de 128 patients.

This prospective study was carried out in order to assess the accuracy of ultrasound in the diagnosis of cirrhosis. One hundred and twenty eight alcoholic patients were included. A careful ultrasonographic examination of the liver was performed before liver biopsy (100 patients). In 15 cases, liver histology was normal, steatosis and/or fibrosis, cirrhosis were diagnosed in 13 and 72 cases respectively. Ultrasonic patterns were classified by the same examiner, according to several criteria: volume, irregular outline, coarse and fine bright echo pattern, attenuation of the ultrasound beam, splenomegaly, ascite, portal hypertension. Cirrhosis was diagnosed in 58 out of 72 patients (80.5%). Specificity was 78.5%, positive and negative predictive values were 90.6% and 61% respectively, and global efficacy was 80%. Irregular outline (0.66), hepatomegaly (0.66) and attenuation of the ultrasound beam (0.64) were the best signs. In case of fine bright echo pattern, the diagnosis of cirrhosis would be missed. The results suggest that ultrasonography is a good test for screening alcoholics for cirrhosis. Therefore, it is useful when liver biopsy is contra indicated or refused or when liver is not detected at the clinical examination.