Background <p>Hypotension is an important complication in the procedural sedation for the gastrointestinal endoscopy despite the use of remimazolam, which has superior hemodynamic stability compared to propofol. Prophylactic vasopressor therapy may preserve cerebral perfusion and enhance the outcome of recovery. This study evaluated the association between continuous infusion of phenylephrine and hemodynamic stability and recovery of cognition in patients undergoing endoscopy under remimazolam sedation.</p> Methods <p>This retrospective cohort study analyzed clinical records of 108 adult patients (American Society of Anaesthesiologists I-III, age 18–75 years) who underwent elective gastroscopy and/or colonoscopy at Pangang Group General Hospital between January 2022 and December 2024. Patients were classified into two groups based on whether they received prophylactic phenylephrine infusion (30&#xa0;µg/min) throughout the procedure (phenylephrine group, <i>n</i> = 54) or standard care without prophylactic vasopressor support (control group, <i>n</i> = 54). All patients received sufentanil 0.1&#xa0;µg/kg and remimazolam 0.15&#xa0;mg/kg for sedation induction. Primary outcomes were hemodynamic parameters and the incidence of hypotension (systolic blood pressure &lt; 90 mmHg or more than 20% decrease from baseline). Secondary outcomes were recovery time, orientation recovery time and adverse events. Hemodynamic variables were measured at baseline and at several intervals after induction. Statistical analyses were performed using independent samples t-tests, repeated measures ANOVA and multiple regression models with adjustments for baseline imbalances.</p> Results <p>The phenylephrine group showed significantly better hemodynamic stability (hypotension, defined as systolic blood pressure &lt; 90 mmHg or a decrease &gt; 20% from baseline, occurred in only 1.8% of patients compared with 61.8% of control group patients; relative risk 0.03, 95% CI: 0.00-0.20, <i>P</i> &lt; 0.001; number needed to treat 1.67). Systolic blood pressure was 12–14 mmHg higher in the phenylephrine group at all post induction time points (all <i>P</i> &lt; 0.001). Maximum blood pressure reduction was significantly reduced (9.94 ± 11.51% vs. 22.47 ± 7.75%, <i>P</i> &lt; 0.001). Recovery time was significantly reduced in the phenylephrine group (8.48 ± 1.63 vs. 11.67 ± 2.21&#xa0;min, mean difference 3.19&#xa0;min, 95% CI: 2.45–3.92, <i>P</i> &lt; 0.001, Cohen’s d = 1.64), a difference of 27.3%. Orientation recovery time was also statistically significantly accelerated (9.11 ± 1.55 vs. 14.63 ± 3.70&#xa0;min, mean difference 5.52&#xa0;min, 95% CI: 4.45–6.59, <i>P</i> &lt; 0.001, Cohen’s d = 1.94), indicating a 37.7% improvement. Severe hypotension (systolic blood pressure &lt; 80 mmHg) occurred in 5.5 vs. 40.0% of patients (<i>P</i> &lt; 0.001). Gender-stratified analyses and multiple regression analyses adjusting for baseline imbalances confirmed the robustness of the observed associations in every subgroup. Adverse events were significantly less in the phenylephrine group (9.3% vs. 74.1%, <i>P</i> &lt; 0.001), with no serious adverse events in either group.</p> Conclusions <p>Continuous phenylephrine infusion during remimazolam sedation was associated with better hemodynamic stability and appeared to be related to a lower incidence of hypotension and faster recovery in patients undergoing gastrointestinal endoscopy. These results suggest the possible benefit of prophylactic vasopressor support to both increase the safety and efficiency of procedural sedation. However, the retrospective nature of this study precludes causal inference. Further prospective randomized studies in various patient populations and clinical settings are warranted in order to confirm these observations and to determine optimal dosing strategies.</p> Trial registration <p>Not applicable. This retrospective study did not require clinical trial number.</p>