Prediction model of ICU readmission in Chinese patients with acute type A aortic dissection: a retrospective study

Area of Science:

• Cardiovascular Surgery
• Critical Care Medicine
• Medical Prediction Modeling

Background:

• Intensive care unit (ICU) readmission for acute type A aortic dissection (ATAAD) patients is a significant clinical challenge.
• Readmission is associated with increased mortality and healthcare costs.

Purpose of the Study:

• To develop and validate a nomogram-based prediction model for ICU readmission in ATAAD patients.
• To identify independent risk factors associated with ICU readmission in this population.

Main Methods:

• Retrospective analysis of 846 ATAAD patients (May 2014 - October 2021).
• Logistic regression to identify independent risk factors for ICU readmission.
• Model validation using Hosmer-Lemeshow test, calibration curves, and Area Under the Receiver Operating Characteristic Curve (AUC).
• Clinical utility assessed via Decision Curve Analysis (DCA).

Main Results:

• 6.7% of ATAAD patients experienced ICU readmission.
• Predictors of ICU readmission included age ≥ 65 years, BMI ≥ 28 kg/m², tracheotomy, continuous renal replacement therapy (CRRT), and longer initial ICU stay.
• The prediction model achieved an AUC of 0.837 and demonstrated good fit (H-L test P=0.519) and clinical utility (DCA).

Conclusions:

• A validated nomogram-based prediction model can effectively assess ICU readmission risk in ATAAD patients.
• Early identification of high-risk patients facilitates timely clinical intervention.
• The model offers practical utility for clinicians managing ATAAD patients post-ICU discharge.