Optimizing the Yield of Abnormal Preoperative Chest Radiographs in Elective Non-cardiothoracic Surgery: Development...

Area of Science:

• Medical Imaging
• Clinical Prediction Models
• Preoperative Assessment

Background:

• Current guidelines for preoperative chest radiographs lack clarity on individual patient benefit.
• Variability in recommendations necessitates a tool to predict abnormal findings.

Purpose of the Study:

• To develop and validate a prediction score for abnormal preoperative chest radiographs in adults undergoing elective non-cardiothoracic surgery.
• To assess the impact of abnormal chest radiographs on perioperative management.

Main Methods:

• Prospective observational study of 703 adult patients undergoing elective non-cardiothoracic surgery.
• Development of a risk prediction score with external validation using 411 patients.
• Receiver operating curve analysis for discriminative performance assessment.

Main Results:

• Abnormal preoperative chest radiographs were identified in 19.5% of patients.
• Significant predictors included age, pulmonary disease, cardiac disease, and diabetes.
• The risk score demonstrated good performance (C-statistic 0.739) and external validation (C-statistic 0.731).

Conclusions:

• Age, pulmonary disease, cardiac disease, and diabetes are key predictors of abnormal preoperative chest radiographs.
• The developed risk score is effective in identifying patients at higher risk.
• This tool may aid in optimizing preoperative assessment and patient selection.