Temporal trends in intraoperative hypotension exposure in major non-cardiac surgery: a retrospective cohort analysis

Abstract

Abstract Background Intraoperative hypotension is a common occurrence during general anaesthesia that has consistently been associated with perioperative organ injury. While international consensus guidelines increasingly advocate for maintaining absolute mean arterial pressure thresholds to mitigate this risk, it remains unclear whether these safety recommendations have translated into measurable changes in routine clinical practice. This study aimed to evaluate temporal trends in the burden of hypotension exposure among adults undergoing major non-cardiac surgery. Methods We conducted a retrospective single-centre cohort study of adults undergoing major non-cardiac surgery at a tertiary academic centre between May 2019 and December 2024. The primary outcome was the time-weighted average mean arterial pressure < 65 mm Hg (TWA MAP < 65 mm Hg), a composite metric integrating severity and duration of hypotensive exposure. Secondary outcomes included the incidence, frequency, and cumulative duration of hypotensive episodes. Temporal trends were assessed using Mann–Kendall tests, and multivariable mixed-effects regression models were constructed to quantify changes over time while adjusting for confounding variables, including patient demographics, comorbidities, and surgical specialty. Results A total of 5,872 patients were included in the final analysis. The primary outcome decreased by 41%, falling from a mean TWA MAP < 65 mm Hg of 0.92 mm Hg in 2019 to 0.54 mm Hg in 2024. Time-series analysis confirmed a significant monotonic downward trend (Mann–Kendall tau = -0.55, p < 0.001). Multivariable mixed-effects regression demonstrated that the year of surgery was an independent predictor of reduced hypotension exposure, with an adjusted geometric mean ratio of 0.90 (95% CI 0.88–0.93; p < 0.001). Consistent reductions were observed across all secondary outcomes, including the frequency and total duration of episodes. These trends remained robust in sensitivity analyses restricted to patients with continuous invasive blood pressure monitoring. Conclusion There was a sustained reduction in intraoperative hypotension exposure over the five-year study period. However, as a purely descriptive study, these data document a temporal phenomenon but cannot identify its drivers. We lack data on specific clinical interventions, such as vasopressor use, fluid administration, or the adoption of advanced monitoring technologies. Further research is required to explicitly investigate the clinical and systemic causes of this intraoperative hypotension exposure trend, as well as its impact on postoperative outcomes.