Predictors of ICU admission associated with gastrointestinal endoscopy in medical inpatients: A retrospective cohort study. Issue 11 (31st July 2022)

• Record Type: Journal Article
• Authors: Sankar, Ashwin; Ladha, Karim S; Grover, Samir C; Jogendran, Rohit; Tamming, Daniel; Razak, Fahad; Verma, Amol A
• Is Part Of: Journal of gastroenterology and hepatology. Volume 37:Issue 11(2022)
• Page Start: 2074
• Page End: 2082
• Publication Date: 2022-07-31
• Subjects: critical care -- endoscopy -- gastrointestinal -- inpatient -- medicine; Gastroenterology -- Periodicals; Digestive organs -- Diseases -- Periodicals; Liver -- Diseases -- Periodicals; Gastroenterology -- Periodicals; Liver Diseases -- Periodicals; 616.33
• DOI: 10.1111/jgh.15969
• Languages: English
• ISSNs: 0815-9319

Abstract

Abstract: Background and Aim: Gastrointestinal (GI) endoscopic procedures are commonly performed in medical inpatients. Limited prior research has examined factors associated with intensive care unit (ICU) admission after GI endoscopy in medical inpatients. Methods: This retrospective cohort study was conducted using routinely‐collected clinical and administrative data from all general medicine hospitalizations at five academic hospitals in Toronto, Canada between 2010 and 2020. We describe ICU admission and death within 48 h of GI endoscopy in medical inpatients. We examined adjusted associations of patient and procedural factors with ICU admission or death using multivariable logistic regression. Results: Among 18 290 medical inpatients who underwent endoscopy, 900 (4.9%) required ICU admission or died within 48 h of endoscopy. Following risk adjustment, ICU admission or death were associated with the following procedural factors: endoscopy on the day of hospital admission (aOR 3.16 [2.38–4.21]) or 1 day after admission (aOR 1.92 [1.51–2.44]) and esophagogastroduodenoscopy (EGD) procedures; and the following patient factors: Charlson comorbidity index of two (aOR 1.38 [1.05–1.81]) or three or greater (aOR 1.84 [1.47–2.29]), older age, male sex, lower hemoglobin prior to endoscopy, increased creatinine prior to endoscopy, an admitting diagnosis of liver disease and certain medications (antiplatelet agents and corticosteroids). Conclusions: ICU admission or death after endoscopy was associated with procedural factors such as EGD and timing of endoscopy, and patient factors indicative of acute illness and greater comorbidity. These findings can contribute to improved triage and monitoring for patients requiring inpatient endoscopy.