Improving haemodynamics in acute gastrointestinal bleeding: Ketamine for endoscopic sedation in active gastrointestinal bleeding in critically Ill patients

Background We aimed to investigate the safety and efficacy of ketamine in sedation for active gastrointestinal bleeding (GIB). Early endoscopy for GIB likely improves clinical outcomes, is often limited acutely by unfavourable haemodynamics, which make safe and effective sedation challenging. Ketamine has favourable haemodynamic properties that make its use desirable in patients with active GIB. Methods We performed a retrospective cohort study from a large military tertiary referral centre of critically ill patients who received ketamine (n = 15) and matched 3:1 to patients who received benzodiazepine/opiate based moderate sedation (n = 45). Data ed included procedural indication, demographics, medication doses, pre‐procedural labs, transfusions, vital signs and adverse events. Results The overall patient cohorts had clinically significant GIB (most commonly upper) with abnormal haemodynamics and substantial blood transfusion requirements. Pre‐endoscopy Rockall score and Charlson Comorbidity Index were higher in patients receiving ketamine. The median ketamine dose was 1 mg/kg. The median midazolam dose was 5 mg in controls vs 2 mg (P = .002) in the ketamine group. Fentanyl doses were also higher in the control group with a median of 100 mcg (range 25‐200) compared to 75 (range 0‐200) for ketamine (P = .02). There was one adverse event in the control group and none in the ketamine group. Mean arterial pressure (MAP) dropped significantly during endoscopy in the moderate sedation group and increased in the ketamine group (P