Ultrasonically guided insertion of a peritoneo-gastric shunt in patients with malignant ascites.

PURPOSE: A new method for internal drainage of malignant ascites is presented in 5 patients with symptomatic malignant ascites. MATERIAL AND METHODS: US-guided percutaneous gastrostomy and paracentesis were performed using the Seldinger technique. A 2.5-mm Cope-loop catheter was inserted in the fluid-filled stomach. In the lower abdomen the proximal part of a Denver peritoneo-venous shunt was introduced after dilation up to 4.8 mm. The pump chamber was sutured to the skin. The distal part of the Denver shunt was cut a few cm from the pump chamber and connected to the gastrostomy catheter. When pumping, ascites is shunted to the stomach lumen. RESULTS: The insertion presented no complications, and all shunt systems initially functioned well. However, the shunts had to be removed within the first 2 weeks because of mechanical problems such as clotting, leakage, and peritoneal septum formation. No infections were reported. CONCLUSION: The peritoneo-gastric shunt may present a therapeutic alternative in selected patients, but the mechanical problems have first to be solved.

Gastric and duodenal wall thickening on abdominal ultrasonography. Positive predictive value.

In a retrospective study, 79 patients with gastric (n = 59) or duodenal (n = 20) wall thickening on conventional abdominal ultrasonograms were included. To reduce bias, patients with a microscopic diagnosis of upper GI neoplasia present at the time of scanning were excluded. The final diagnosis was based upon endoscopy, operation, upper GI series, or autopsy. Among the 59 cases of gastric wall thickening, 33 (56%) proved to be gastric cancer, one (2%) was benign neoplasia, and 17 (29%) showed nonneoplastic pathology (ulcer, gastritis, fibrosis). In eight cases (13%) no gastric pathology was present in the final diagnosis, thus producing false-positive sonographic diagnoses. Among the 20 cases of duodenal wall thickening, five (25%) proved to be duodenal cancer and seven (35%) showed nonneoplastic duodenal pathology. We concluded that gastric or duodenal wall thickening shown on abdominal sonography is a significant finding indicating upper GI pathology (malignant or nonmalignant) in a high percentage of the gastric (86%) and duodenal (60%) cases.

Cost-benefit analysis of prophylaxis against deep vein thrombosis in surgery.

A cost-benefit analysis was performed based on the incidence of postoperative thromboembolic complications, the cost of prophylaxis, and the cost of treating the complications. It was determined that prophylaxis against thromboembolic complications in both general surgery and total hip replacement, with either standard heparin or low molecular weight heparin, saves lives as well as reduces health care costs.

Fatal pulmonary embolism associated with surgery. An autopsy study.

The records of all autopsies performed at two major Danish hospitals in 1986 were reviewed in order to analyze cases of fatal pulmonary embolism. There were 2,609 hospital deaths and 1,603 post mortem examinations. Pulmonary embolism was the primary cause of 74 deaths, 16 of which were postoperative. The median age of these ten men and six women was 72 years. In nine of the 16 cases the prognosis would have been favorable had embolism not occurred. Only three of the 16 had received thromboembolic prophylaxis. Four of the deaths from embolism occurred less than 24 hours after surgery, five within 7 days and seven between postoperative days 7 and 30. The estimated incidence of fatal pulmonary embolism following surgery was 1.2-1.3 per thousand.

[Presence of pulmonary emboli in Danish autopsy material]. / Forekomsten af lungeemboli i et dansk obduktionsmateriale.

Two hundred and ten patients with verified pulmonary emboli (LE) at autopsy were assessed retrospectively. These patients comprised 13% of all those submitted to autopsy. LE was the primary cause of death, a contributory cause of death or an incidental finding in 35%, 31% and 34%, respectively, of the cases. Two thirds of the lethal cases were not recognized prior to autopsy. LE was preceded by medical conditions and operation with subsequent immobilization in 55% and 22% of the cases, respectively. Out of these, it is estimated that 17% of the medical patients and 56% of the surgical patients would have had a good prognosis if LE had not occurred. If foreign results can be applied to Danish conditions, approximately 1,400 fatal cases of LE should occur in Denmark per annum. This review confirms that the actual Danish figures are of this magnitude, at least, as approximately 1,500 lethal cases must be assumed to occur per annum solely among patients dying in hospital. In 1986, a total of 278 cases of LE were stated as the primary cause of death on the death certificates. Despite the limitations of a retrospective investigation, it is concluded that the vital statistics of the Danish Board of Health underestimate the genuine number of cases. The range of indications for prophylaxis of thrombosis is possibly too narrow. LE should be considered in the differential diagnosis more frequently, particularly in elderly immobilized patients.