Occult upper gastrointestinal mucosal abnormalities in critically ill patients.

BACKGROUND: The objectives of this study were to estimate the frequency of occult upper gastrointestinal abnormalities, presence of gastric acid as a contributing factor, and associations with clinical outcomes. METHODS: Data were extracted for study participants at a single centre who had an endoscopy performed purely for research purposes and in whom treating physicians were not suspecting gastrointestinal bleeding. Endoscopic data were independently adjudicated by two gastroenterologists who rated the likelihood that observed pathological abnormalities were related to gastric acid secretion using a 3-point ordinal scale (unlikely, possible or probable). RESULTS: Endoscopy reports were extracted for 74 patients [age 52 (37, 65) years] undergoing endoscopy on day 5 [3, 9] of ICU admission. Abnormalities were found in 25 (34%) subjects: gastritis/erosions in 10 (14%), nasogastric tube trauma in 8 (11%), oesophagitis in 4 (5%) and non-bleeding duodenal ulceration in 3 (4%). The contribution of acid secretion to observed pathology was rated 'probable' in six subjects (rater #1) and five subjects (rater #2). Prior to endoscopy, 39 (53%) patients were receiving acid-suppressive therapy. The use of acid-suppressive therapy was not associated with the presence of an endoscopic abnormality (present 15/25 (60%) vs. absent 24/49 (49%); P = 0.46). Haemoglobin concentrations, packed red cells transfused and mortality were not associated with mucosal abnormalities (P = 0.83, P > 0.9 and P > 0.9 respectively). CONCLUSIONS: Occult mucosal abnormalities were observed in one-third of subjects. The presence of mucosal abnormalities appeared to be independent of prior acid-suppressive therapy and was not associated with reduced haemoglobin concentrations, increased transfusion requirements, or mortality.