Retrospective analysis of endoscopy and colonoscopy procedures performed on patients in the Internal Medicine Intensive Care Unit

This retrospective study examined the clinical and demographic data of patients admitted to the internal medicine intensive care unit (ICU) who underwent endoscopy and colonoscopy. A total of 158 patients were included from January 1, 2013, to December 30, 2023. The mean age of the patients was 69.8 years with a standard deviation of ±15.0 years. Among the participants, 60.1% were male and 39.9% were female. Comorbidities among the patients included hypertension (51.9%), diabetes mellitus (33.5%), coronary artery disease (28.5%), congestive heart failure (15.2%), chronic kidney disease (12.7%), chronic obstructive pulmonary disease (12.7%) and cirrhosis (11.4 %). Of the patients, 61 (38.6%) were admitted to the ICU due to gastrointestinal bleeding (GIS), 40 (25.3%) due to pneumonia, 27 (17%) due to septicemia, 11 (7.0%) due to cerebrovascular accident, 7 (4.4%) due to postoperative follow-up, and 12 (10.6%) due to other reasons. The mean Glasgow Coma Scale (GCS) score at admission was 9.0 (±4.1), the mean Sequential Organ Failure Assessment (SOFA) score was 9.2 (±3.7), and the mean Acute Physiology and Chronic Health Evaluation II (APACHE II) score was 32.7 (±9.8). Of the patients included in the study, 132 (83.5%) underwent endoscopy, 11 (7%) underwent colonoscopy, and 15 (9.5%) underwent both procedures. The most common indication for these procedures was gastrointestinal system (GIS) bleeding, accounting for 77.2% of cases. Other indications included percutaneous enteral gastrostomy (13.9%), gastrointestinal obstruction (7.6%), tracheoesophageal fistula (0.6%), and ingestion of corrosive substances (0.6%). Clinical findings were detected in 134 patients (85%), which included melena (35.8%), hematochezia (12.7%), hematemesis (9.7%), coffee ground vomiting (9.7%), blood in the nasogastric tube (8.2%), and other clinical findings (21.6%). The most common findings in endoscopy and colonoscopy procedures were gastric ulcer, erosive gastritis and duodenal ulcer. Endoscopic and colonoscopic therapeutic procedures were performed in 83 patients (52%), including cauterization (28.5%), band ligation (7%), sclerotherapy (4.4%), stent placement (3.8%), and other therapeutic procedures (8.9%). The 28-day mortality rate for patients was 47.5%, the 3-month mortality rate was 53.2%, and the 1-year mortality rate was 56.5%. The admission GCS score was significantly lower in patients who died in the ICU compared to those who survived (survived: 11.2±4.0; deceased: 7.6±3.4; p=0.016). Additionally, the admission APACHE II score (survived: 27.8±9.0; deceased: 39.7±5.5; p<0.001) and the SOFA score calculated at admission (survived: 8.0±3.4; deceased: 10.9±3.1; p<0.001) were significantly higher in patients who did not survive. Among those who died in the ICU, the rates of needing invasive mechanical ventilation (survived: 38.3%; deceased: 92.2%; p<0.001), developing acute kidney injury (survived: 38.3%; deceased: 73.4%; p<0.001), requiring hemodialysis (survived: 11.7%; deceased: 42.2%; p<0.001), intubation due to bleeding (survived: 6.4%; deceased: 23.4%; p=0.002), and requiring vasopressor support (survived: 63.0%; deceased: 95.0%; p<0.001) were significantly higher compared to those who survived. However, no significant differences were found regarding the number of blood transfusions (survived: 3.6±2.2; deceased: 3.2±2.0; p=0.289) or the rates of requiring a blood transfusion (survived: 73.4%; deceased: 85.9%; p=0.06). The types of endoscopy and colonoscopy procedures performed in patients who died in the ICU (p=0.243), clinical findings during the procedures (p=0.921), indications for the procedures (p=0.570), findings detected in endoscopy and colonoscopy (p>0.05), types of therapeutic procedures performed (p=0.776), and the status of repeated procedures in the ICU (p=0.346) were found to be similar to those in patients who survived.