Estimated Pulse Wave Velocity, Early Postoperative Neutrophil to Lymphocyte Ratio, and Delirium After Cardiac Surgery: A Retrospective Cohort Study Using the Medical Information Mart for Intensive Care IV Database.

OBJECTIVES

To evaluate the associations of estimated pulse wave velocity (ePWV) and early postoperative neutrophil to lymphocyte ratio (NLR) with postoperative delirium (POD) after cardiac surgery and their incremental model value.

DESIGN

Retrospective cohort study.

SETTING

Cardiac surgical intensive care unit (ICU) cohort from the Medical Information Mart for Intensive Care IV database.

PARTICIPANTS

Adults undergoing open cardiac surgery with available ePWV, postoperative 0- to 24-hour NLR, and evaluable Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) assessments.

INTERVENTIONS: None.

MEASUREMENTS AND MAIN

RESULTS

Estimated pulse wave velocity was calculated from age and early postoperative mean arterial pressure, and NLR was calculated from measurements obtained within 24 hours after ICU admission. The primary outcome was strict CAM-ICU-detected POD, excluding assessments during deep sedation. The multivariable logistic regression model adjusted for age, mean arterial pressure, baseline clinical covariates, and perioperative covariates, with multiple imputation for missing data. Among 5,684 patients, 1,001 (17.6%) developed POD. Each 1-standard deviation increase in ePWV was associated with POD (odds ratio [OR], 1.46; 95% confidence interval [CI], 1.06-2.02; p = 0.021), as was each doubling of NLR (OR, 1.15; 95% CI, 1.06-1.25; p = 0.001). Findings were generally consistent across sensitivity analyses. Adding both markers to the reference model increased the apparent area under the curve from 0.790 to 0.791 (change, 0.0016; 95% CI, -0.0006 to 0.0037).

CONCLUSIONS

Higher ePWV and early postoperative NLR were associated with strict CAM-ICU-detected POD. However, ePWV represents an age- and blood pressure-derived composite index, and the incremental discrimination provided by both markers beyond clinical and perioperative factors was limited.