Liver resection surgery, crucial for removing liver cancer and other issues, has a high postoperative mortality rate of over 10%. A French multidisciplinary team developed a digital twin model of blood circulation, predicting liver failure risks with personalised pre-operative insights. Tested on 47 patients, the model accurately matched real blood pressure data, offering surgeons a vital tool to improve outcomes and avoid risky surgeries.
Even with advanced surgical techniques and technologies, some surgeries still have a high risk of failure. For instance, liver resection, used to remove liver cancer and other liver-related issues, has a postoperative mortality rate exceeding 10%. However, digital twins are now available to help reduce these risks by providing surgeons with personalised, life-saving pre-operative insights.
Liver resection is a complex surgery required to remove liver cancer, metastases, benign tumours, cysts, and more. The primary cause of postoperative deaths is increased pressure in the portal vein, a major blood vessel carrying 75% of the liver's blood flow from the intestine, gallbladder, and pancreas. This complication is severe and needs quick detection to prevent fatal outcomes.
In recent decades, pre-existing portal hypertension, or high pressure in the portal vein, changed from being a surgical contraindication to a risk factor, but of limited predictive power. In fact, the real issue is postoperative hypertension, for which no predictive current tool exists. Therefore, personalised methods are needed to prevent complications
A multidisciplinary team from France, including experts from AP-HP and Inria, created a mathematical model, or digital twin, of the entire blood circulation which can, in a few minutes, predict the risk of postoperative liver failure for a given patient.
The team conducted a feasibility study on humans, creating a model of the entire blood circulation for liver resection. They used data from 47 patients who underwent liver resection. The surgery was simulated in the model, and the predicted liver blood pressure was compared to the actual measurements taken from the patients at the end of the surgery. The results were very promising and matched well with the measured data.
In conclusion, the team developed and tested a digital twin of blood flow that can predict the risk of liver failure after extensive liver resection. This model could become an essential pre-operative tool for surgeons, allowing them to provide better treatments and avoid risky surgeries with highly personalised predictions.
