Development and validation of a predictive model for postoperative delirium in patients undergoing cardiac surgery - Report - MDSpire

Development and validation of a predictive model for postoperative delirium in patients undergoing cardiac surgery

  • By

  • Lin Cui

  • Jinhong Zhang

  • Xiaoling Sun

  • Jiangling Xia

  • Hongyu Xu

  • June 4, 2026

  • 0 min

Share

Clinical Report: Creation and assessment of a predictive nomogram for POD

Overview

This study identifies key risk factors for postoperative delirium (POD) in cardiac surgery patients and develops a validated predictive nomogram. The model demonstrates excellent discriminative ability and strong clinical utility.

Background

Postoperative delirium (POD) is a common and serious complication following cardiac surgery, affecting cognitive function and increasing healthcare costs. Accurate risk assessment is crucial for timely interventions to mitigate its impact. This study addresses the need for a comprehensive predictive model that incorporates preoperative, intraoperative, and postoperative factors.

Data Highlights

MetricValue
POD Incidence39.4%
Training Cohort Size507
Validation Cohort Size217

Key Findings

  • POD occurred in 285 out of 724 patients (39.4%).
  • Key predictors of POD included emergency surgery, age, Sequential Organ Failure Assessment score, postoperative shock, and blood lactate and glucose levels.
  • The nomogram showed high area under the receiver operating characteristic curve (AUROC) indicating excellent discriminative ability.
  • Strong calibration agreement was observed in both training and validation datasets.
  • Decision curve analysis (DCA) and clinical impact curve (CIC) results indicated strong clinical utility of the nomogram.

Clinical Implications

The validated nomogram provides a practical tool for clinicians to predict the risk of POD in cardiac surgery patients. Understanding the identified risk factors can guide preoperative assessments and interventions.

Conclusion

This study successfully develops a predictive nomogram for POD, highlighting its potential as a valuable tool in clinical practice for cardiac surgery patients.

Related Resources & Content

  1. BMC Psychiatry, Springer, 2025 -- Machine Learning-Based prediction models for postoperative delirium: a systematic review and Meta-Analysis
  2. npj Digital Medicine, 2025 -- Development of a deep learning-based prediction model for postoperative delirium using intraoperative electroencephalogram in adults
  3. Intensive Care Medicine, 2015 -- International Collaboration on the Development and Validation of an Early Prediction Model for Delirium in ICU Patients
  4. Frontiers in Psychiatry, 2026 -- Prediction model for postoperative delirium risk in elderly hypertensive patients: machine learning-based development and validation
  5. Update of the European Society of Anaesthesiology and Intensive Care Medicine evidence-based and consensus-based guideline on postoperative delirium in adult patients - PubMed, 2024
  6. The Effects of Dexmedetomidine on Postoperative Delirium in Adult Cardiac Surgical Patients: A Bayesian Meta‐Analysis and Trial Sequential Analysis - PMC
  7. Full article: Preoperative prediction models for postoperative delirium in cardiac surgery patients – a scoping review, 2025
  8. Update of the European Society of Anaesthesiology and Intensive Care Medicine evidence-based and consensus-based guideline on postoperative delirium in adult patients - PubMed
  9. The Effects of Dexmedetomidine on Postoperative Delirium in Adult Cardiac Surgical Patients: A Bayesian Meta‐Analysis and Trial Sequential Analysis - PMC
  10. Full article: Preoperative prediction models for postoperative delirium in cardiac surgery patients – a scoping review

Original Source(s)

Related Content