Post-CPB Hemodynamics and Mortality After Aortic Surgery

Post-CPB hemodynamics may help identify patients at increased early risk after aortic surgery. In a retrospective study of 691 adults, lower systolic and mean blood pressure and higher shock indices after separation from cardiopulmonary bypass were associated with 30-day mortality. These measurements describe risk during recovery from bypass; the study does not show that correcting a value to a particular threshold improves survival.

Study Design and Population

Kim and colleagues analyzed adults undergoing aortic surgery with CPB at Asan Medical Center in Seoul from March 2019 through February 2023. Among 6,138 screened cardiac surgical procedures, 691 patients met the final eligibility criteria. Thirty patients (4.3%) died within 30 days.

Procedures were grouped as aortic root/ascending-only surgery (293 patients) or surgery involving the arch and/or distal aorta (398 patients). The cohort included 92 emergency operations. Patients with incomplete intraoperative heart rate or blood pressure recordings were excluded, and no missing data were imputed. Anesthesia, vasoactive support, and perfusion management were individualized rather than governed by one uniform study protocol.

What Post-CPB Hemodynamics Meant in This Study

The investigators compared preoperative, pre-incision, pre-CPB, and post-CPB measurements. The post-CPB window extended from bypass termination to the end of anesthesia, rather than representing one reading immediately after separation or subsequent ICU measurements. Active CPB was excluded from the hemodynamic analysis because those values primarily reflect extracorporeal support.

Intraoperative measurements were recorded at five-minute intervals and averaged within each phase. Shock index (SI) was calculated as phase-mean heart rate divided by phase-mean systolic blood pressure; modified shock index (MSI) used phase-mean mean blood pressure as the denominator. These ratios of phase averages should not be assumed interchangeable with a single bedside measurement or an average of instantaneous ratios.

Post-CPB Hemodynamics: Key Findings

After Holm correction across 24 comparisons, only post-CPB systolic blood pressure (SBP), mean blood pressure (MBP), SI, and MSI remained significantly different between survivors and non-survivors. Median SBP was 110.9 versus 98.9 mmHg, and median MBP was 79.7 versus 72.7 mmHg. Earlier-phase findings and post-CPB heart rate and diastolic pressure did not remain significant after this correction.

These are relative hazard estimates, not absolute increases in the probability of death. Because each hemodynamic variable was assessed in a separate model, the results do not demonstrate that SI or MSI adds predictive information beyond blood pressure itself.

Discrimination and Exploratory Thresholds

Individual variables showed modest discrimination: AUROC was 0.723 for SI, 0.722 for MSI, 0.717 for SBP, and 0.716 for MBP. The small numerical differences do not establish that the composite indices are superior. Heart rate and diastolic pressure performed less well, with AUROC values of 0.643 and 0.618.

The ROC-derived cutoffs were SBP 99.6 mmHg, MBP 73.4 mmHg, SI 0.755, and MSI 1.208. The authors then used rounded values of 100 mmHg, 70 mmHg, 0.7, and 1.2 for survival analyses. These rounded values differ from the fitted cutoffs and were derived and evaluated within the same cohort.

These thresholds are hypothesis-generating risk markers, not validated treatment targets, automatic triggers for medication, or a universal definition of safe separation from bypass. The study did not test a threshold-guided intervention or show that manipulating heart rate or pressure changes mortality.

Depth and Duration of Hypotension

An additional analysis estimated the burden of MBP below 65 mmHg after CPB, incorporating both depth and duration. Non-survivors spent a median 22.20% of the post-CPB period below that level, compared with 4.23% among survivors. The time-weighted pressure deficit was 1.37 versus 0.11 mmHg, respectively.

The investigators estimated these measures by interpolating between recorded values, without interpolating gaps longer than 10 minutes. Findings were consistent when restricted to patients with at least 80% recording coverage. However, the analysis did not establish whether hypotension burden improves prediction beyond phase-mean pressure, and it does not validate 65 mmHg as an intervention target for this population.

Relevance for Perfusion and Cardiac Anesthesia

The report emphasizes the period between bypass termination and completion of anesthesia as a useful window for multidisciplinary risk assessment. Recorded pressures and shock indices reflect a mixture of ventricular recovery, vascular tone, volume status, operative severity, and ongoing treatment. Non-survivors also had longer median CPB times (244 versus 145 minutes) and greater vasoactive-inotropic support.

More frequent use of selected vasoactive agents in non-survivors should not be interpreted as evidence that those drugs caused harm. Sicker patients often require more support, and medication exposure and hemodynamics were measured during the same phase. The study cannot cleanly separate underlying circulatory dysfunction from treatment effects.

For perfusion teams, the findings support careful handoff and contextual review of hemodynamic trends after bypass. They do not identify a preferred pump-flow strategy, cannulation approach, drug regimen, or pressure target during active CPB.

Limitations and Disclosures

This was a retrospective, single-center study with heterogeneous operations and only 30 deaths. Non-survivors were older and more likely to undergo emergency surgery; the main adjusted models included surgical extent, CPB time, and mean vasoactive-inotropic score, leaving substantial potential for residual confounding from baseline illness and other management factors.

Ridge penalization can reduce overfitting but cannot remove confounding or substitute for external validation. Excluding incomplete recordings may introduce selection bias. Five-minute sampling and phase averaging can miss brief events and obscure changing trajectories. Fluid balance, transfusion, postoperative hemodynamics, and biochemical hypoperfusion markers were not fully incorporated into the reported analyses.

The work was supported by Korean national research and health-technology funding and Asan Institute for Life Sciences grants. The authors state that funders had no role in the study or publication decision and declare no conflicts of interest. Patient-level data and study-specific analysis code are not publicly available because of institutional restrictions.

Original Article

Kim CW, Kim WJ, Kim HS, et al. Association Between Post-Cardiopulmonary Bypass Hemodynamics and 30-Day Mortality After Aortic Surgery: A Single-Center, Retrospective Study. Journal of Clinical Medicine. 2026;15(20):7908. doi:10.3390/jcm15207908.

This independently written educational summary accompanies the unchanged source PDF. The cover is an AI-generated editorial illustration and is not a technical guide to hemodynamic management.