A prospective study was performed on 358 patients to examine the diagnosis, management, and natural history of fish bone ingestion. All patients admitted with the complaint had a thorough oral examination. Flexible endoscopy under local pharyngeal anesthesia would be performed on patients with negative findings. Of 117 fish bones encountered, 103 were removed (direct removal, 21; endoscopic removal, 82) and 12 were inadvertently dislodged. One was missed and the other one necessitated removal with rigid laryngoesophagoscopy under general anesthesia. Morbidity (1%) occurred in patients with triangular bones in the hypopharynx, resulting in one mucosal tear and two lengthy procedures. Mean hospital stay was 7 hours. Prediction of the presence of fish bones by symptoms and radiograph was poor. The location of symptoms, however, was useful in guiding the endoscopist to the site of lodgment. Of patients who refused endoscopy, only one (2.8%) developed retropharyngeal abscess. As compared to those who received endoscopy, 31.8% had fish bones detected. As the yield of fish bone detected was also inversely related to the duration of symptoms, we strongly suspect that most of the unremoved fish bones would be dislodged and passed. However, because of the serious potential complication from fish bone ingestion, we believe that a combination of oral examination followed by flexible endoscopy is indicated in all patients. When triangular bones in the hypopharynx are encountered, rigid laryngoesophagoscopy should be considered. This protocol had safely and effectively dealt with the present series of patients.

To determine the efficacy of ranitidine in preventing clinically acute overt gastroduodenal (GD) complications (bleeding and/or perforation) after neurosurgery, 101 patients with nontraumatic cerebral disease considered at high risk of developing postoperative GD complications were randomized in a standard double-blind manner to receive either ranitidine (50 mg every 6 hours) or placebo medication preoperatively. Postoperative serial GD endoscopy was used to document the occurrence of complications: an overt symptomatic complication was defined as bleeding requiring blood transfusion and/or surgery. Fifty-two patients received ranitidine and 49 received a placebo preoperatively; 30 developed overt GD bleeding; nine of these received ranitidine and 21 received a placebo. Ranitidine significantly reduced the incidence of bleeding (p < 0.05). Multivariate logistic regression analysis revealed three factors of independent significance in predicting overt GD bleeding: use of a placebo drug, a gastric pH of less than 4, and a high daily volume of gastric output. The authors conclude that ranitidine is useful in preventing postoperative GD complications in high-risk neurosurgical patients.

No abstract