An analysis of upper GI endoscopy done for patients in surgical intensive care: high incidence of, and morbidity from reflux oesophagitis.

Objective: To investigate the role of upper gastrointestinal (GI) endoscopy in a surgical intensive care unit [ICU].

Design: Retrospective analysis.

Setting: University hospital, The Netherlands.

Subjects: 87 Male and 42 female patients, mean age 62.0 years (range 14-86).

Interventions: 198 Upper GI endoscopies.

Main outcome measures: Incidence of, indication for, and abnormalities noted at upper GI endoscopy.

Results: 52 (40%) and 18 (14%) patients underwent 82 and 27 upper GI endoscopies, respectively, for evaluation of upper GI haemorrhage and surgical anastomoses. 59 Patients (46%) underwent 89 endoscopies for placement of nasoduodenal feeding tubes (n = 86, 97%), biliary stents (n = 2, 2%) and gastrostomy cathether (n = 1, 1%). The causes of haemorrhage were: oesophagitis (n = 13, 25%), duodenal ulcer (n = 13, 25%), gastric ulcer (n = 7, 13%) and others (n = 14, 28%). In 6 cases (11%), no bleeding site was detected. As a coincidental finding, a third of all patients had oesophagitis. The incidence of haemorrhage in patients treated and not treated by mechanical ventilation was 43/1350 (3.2%) and 9/1470 (0.6%), respectively (p < 0.0001).

Conclusions: Upper GI endoscopy is a common diagnostic and therapeutic procedure in a surgical ICU. Reflux oesphagitis is often found and is clinically important. Mechanical ventilation is a risk factor for upper GI haemorrhage.