New machine learning models predict adverse postop abdominal hernia surgery outcomes

Editor's Note

Machine learning (ML) models designed by surgeons at the University of Texas MD Anderson Cancer Center in Houston show a high level of accuracy in predicting which types of patients are most likely to have a hernia recurrence or other complications after surgery in this study.

Of 725 patients who had open surgery for ventral hernia repair, the models achieved the following accuracy rates:

• 85% for predicting hernia recurrence
• 72% for predicting surgical site occurrence
• 84% for predicting 30-day hospital readmission.

Factors contributing to an increase in hernia recurrence were an existing breach of the rectus muscle of the front abdominal wall, obesity, and the bridged repair technique, which used mesh to span the hernia defect.

The authors note that research has shown a 1% reduction in the rate of hernia recurrence would save the healthcare system $30 million.