Stress-induced gastroduodenal lesions and total parenteral nutrition in critically ill patients: frequency, complications, and the value of prophylactic treatment. A prospective, randomized study.

Abstract

The objective of this study was to assess the frequency, complications, and value of prophylactic treatment of stress‐induced gastroduodenal lesions in critically ill patients requiring total parenteral nutrition (TPN) and prolonged mechanical ventilation. Multidisciplinary intensive care unit (ICU) patients were prospectively randomized to treatment with TPN alone, TPN with sucralfate, or TPN with ranitidine. Ninety‐seven patients were evaluated. Prospectively randomized patients with expected mechanical ventilation for a minimum of 6 days who were metabolically stressed, hemodynamically stable, had normal hepatic and renal function, and required TPN were included. Exclusion criteria were history of gastroduodenal ulcer in the past 12 months before admission, previous operations of the upper gastrointestinal tract, diagnosis of active gastrointestinal hemorrhage, renal or hepatic failure, and administration of antacids, H2‐receptor blocking agents, and sucralfate 48 hours before entering the study. Patients were removed from the study if they were weaned from mechanical ventilation before the sixth day (n = 10), had suffered an episode of acute gastrointestinal hemorrhage (n = 5), developed stress‐induced gastroduodenal ulcers (n = 2), had chronic duodenal ulcers (n = 2), had gastric cancer (n = 1), or could tolerate enteral feedings (n = 1). Ninety‐seven patients were evaluated. Seventy‐three patients completed the study (53 men and 20 women), and the mean age was 38 ± 16 years. To stratify the severity of illness, the APACHE II score and the catabolic index score were obtained on admission to the study. Gastroduodenal mucosal damage was determined by endoscopy performed on day 3 after admission to the ICU and every 7 days thereafter. Endoscopic mucosal appearance was classified using a modification of the Pleura and Johnson1 classification: (1) normal mucosa or erythema, (2) nonhemorrhagic erosions/petechial changes, (3) stress gastroduodenal ulcers without bleeding, and (4) stress gastroduodenal ulcers with endoscopic signs of bleeding.Thirty patients received TPN alone. Twenty‐four patients received TPN and sucralfate (1 g by nasogastric tube every 4 hours) and 19 patients received TPN and ranitidine (50 mg IV every 6 hours). The three groups were matched for severity of illness (APACHE II score) and demographic characteristics. Patients were monitored for clinical evidence of acute upper gastrointestinal hemorrhage and underwent urgent endoscopy if symptoms or signs of gastrointestinal hemorrhage developed. Coagulation studies, hemodynamic status, and active intervention measures were recorded. Endoscopic examinations (n = 101) were performed in 64 patients (9 patients did not undergo endoscopy). Evidence of gastroduodenal mucosal abnormalities (categories 2, 3, or 4) was noted in 19 of 64 patients (overall occurrence rate of 29.6%). The overall frequency for stress ulcerations was 15.6% and 6.2% for stress‐induced hemorrhage. There were no deaths due to stress‐induced hemorrhage. The results demonstrate no statistically significant differences in the frequency of stress‐induced lesions and stress‐induced hemorrhage among the three groups of patients. The authors concluded that prophylaxis with sucralfate or ranitidine in patients receiving TPN offers no additional advantage in a group of ventilated ICU patients receiving TPN in terms of risk for developing stress‐induced gastroduodenal ulcerations and hemorrhage.