Analysis pipeline for: "Beyond Mean Arterial Pressure: A Retrospective Cohort Analysis of Pressure–Flow Dissociation and Postoperative AKI in Major Aortic Surgery"

This archive contains the de-identified analytic dataset and the complete statistical analysis pipeline supporting the manuscript “Beyond Mean Arterial Pressure: A Retrospective Cohort Analysis of Pressure–Flow Dissociation and Postoperative Acute Kidney Injury in Major Aortic Surgery” (Gezela, Żurański, Szrama, Kulas, Sobczyński; prepared for submission to Journal of Clinical Medicine, 2026). The study is a cardiac-index–focused secondary analysis of 100 consecutive patients undergoing elective major aortic surgery (retrospective, single-centre, Poznań). Continuous cardiac index (CI) and mean arterial pressure (MAP) signals were extracted from HemoSphere monitoring exports at approximately 20-second intervals. The primary metric was the burden of low CI (< 2.2 L/min/m2) during normotensive periods (MAP ≥ 65 mmHg); the primary outcome was acute kidney injury (AKI) by KDIGO criteria. Key findings (n = 99 evaluable, 34 AKI events): pressure–flow dissociation was frequent (low CI during normotension in 94% of patients), but neither the burden of low CI (adjusted OR 0.88 per 10–percentage-point increase, 95% CI 0.67–1.12, p = 0.33) nor the choice of monitoring strategy (HPI vs FloTrac; adjusted OR 0.57, 95% CI 0.20–1.53, p = 0.27) was associated with AKI. Sensitivity, time-to-event, and heart rate × CI interaction analyses yielded uniformly null findings. These results suggest a boundary condition on the generalisability of recent cardiac-surgery CI–AKI associations. Archive contents: De-identified per-patient dataset (100 patients × 71 variables; sequential IDs P001–P100; no dates, no HemoSphere filenames, no identifiers). R and Python analysis scripts to reproduce every table and figure in the manuscript and supplementary material. Main-text figures (PDF + PNG). sessionInfo.txt with pinned R 4.6.0 and Python 3.14.6 versions. Full README.md with reproduction instructions and complete file inventory. Access mode: Restricted Access. Access to the dataset requires approval by the corresponding authors. Requests for legitimate academic research, peer review, or replication purposes are reviewed under the Poznań University of Medical Sciences Bioethics Committee approval (protocol 409/25, 28 May 2025). Ethics: the study was conducted in accordance with the Declaration of Helsinki. The retrospective design carried an IRB waiver of informed consent. Parent publication (same cohort, MAP-focused analysis): Szrama J, Gezela M, Żurański Ł, Kulas K, Gajda M, Smuszkiewicz P, Sobczyński P. Hypotension Prediction Index Software Compared with Standard Advanced Haemodynamic Monitoring in Patients Undergoing Major Aortic Surgery: A Retrospective Study. J Clin Med 2025;14:8791. doi:10.3390/jcm14248791. Full technical documentation, reproduction instructions, software requirements, and complete file inventory: see README.md inside the archive.