Computed Tomography Radiomics Nomogram to Predict the Intraoperative Hypertensive Crisis of Abdominal Pheochromocytoma and Paraganglioma

TL;DR

This study created a CT-based tool to predict life-threatening high blood pressure during surgery for abdominal tumors, offering a non-invasive and accurate method for risk assessment.

Contribution

A novel CT radiomics nomogram was developed and validated to predict intraoperative hypertensive crisis in PPGL patients.

Findings

The nomogram achieved high accuracy with an area under the curve of 0.91 in the training set.

The model included a history of coronary heart disease and CT radiomics score for prediction.

The radiomics nomogram showed strong clinical usefulness according to decision curve analysis.

Abstract

Patients with abdominal Pheochromocytoma and Paraganglioma (PPGL) are prone to a hypertensive crisis during surgery, which may endanger their lives. This study aimed to develop and validate a Computed Tomography (CT) radiomics nomogram for the prediction of intraoperative hypertensive crisis in patients with PPGL.

In this retrospective study, 212 patients with abdominal PPGL underwent abdominal-enhanced CT and surgical resection. Radiomic features were extracted from arterial and venous phases. Multivariable logistic regression models were developed using an internal validation and an external test set. The performance of the nomograms was determined by their discrimination, calibration, and clinical usefulness.

A total of 212 patients with PPGL were included, involving 44 with hypertensive crises. The patients were divided into training (n = 117), validation (n = 51), and test (n = 44) sets. Eighteen radiomics-relevant radiomic features were selected. A history of coronary heart disease and the CT radiomics score were included in the prediction model, which achieved an area under the curve of 0.91 [95% Confidence Interval (CI) 0.85-0.97] in the training set, 0.93 (95% CI 0.84-0.99) in the validation set, and 0.85 (95% CI 0.72-0.97) in the test set. The decision curve analysis demonstrated the radiomics nomogram to be clinically useful.

Our study has developed and validated a CT radiomics nomogram that has demonstrated remarkable potential in predicting intraoperative hypertensive crisis in patients with abdominal pheochromocytoma and paraganglioma. This non-invasive, straightforward approach has exhibited high accuracy, ease of use, and predictive power.

Figures