Probabilistic Prediction of Gastrointestinal Ischemia after Cardiothoracic Surgery

Area of Science:

• Cardiovascular Surgery
• Gastrointestinal Medicine
• Radiology

Background:

• Gastrointestinal ischemia (GIisch) diagnosis is challenging post-cardiothoracic surgery.
• Computed tomography angiography (CTA) has limitations, including high false-negative and false-positive rates.
• Improved diagnostic methods are needed for GIisch in this patient population.

Purpose of the Study:

• To evaluate if combining readily available variables with CTA findings improves GIisch diagnosis.
• To develop and validate a GIisch prediction score for post-cardiothoracic surgery patients.

Main Methods:

• Retrospective study of intensive care patients post-cardiothoracic surgery.
• GIisch confirmed by surgical/endoscopic findings.
• Developed a GIisch prediction score using the Spiegelhalter-Knill-Jones system in a training cohort and validated it.

Main Results:

• The developed score incorporated cardiopulmonary bypass, intraoperative mean arterial pressure, AST levels, lactate increase, and CTA findings (bowel dilation, wall thickening, mesenteric vasoconstriction).
• The score demonstrated good diagnostic performance with an area under the receiver operating characteristic curve of 0.82 in both training and validation cohorts.
• Reliable probability predictions were achieved for probabilities ≤ 30% or ≥ 70%.

Conclusions:

• CTA findings alone are insufficient to rule out GIisch in post-cardiothoracic surgery patients.
• A scoring system integrating CTA findings with other clinical variables enhances GIisch diagnosis in this critical care setting.