Using blood proteins to predict complications after heart surgery

Theme Translational data science

Workstream Omics for prediction and prognosis

Status: This project is ongoing

When people have heart surgery, some develop serious complications afterwards. Two of the most common are acute kidney injury (AKI) and atrial fibrillation (AF, an irregular heartbeat).

Each can affect around 1 in 3 patients. These problems are thought to be linked to how the body’s immune system responds to the stress of surgery, but we do not yet fully understand why some patients are more vulnerable than others.

One way to study this is through molecular profiling – looking at patterns of small molecules and proteins in the blood. These ‘signatures’ may reveal early warning signs of complications or even point to new treatment options.

So far, research has mainly focused on metabolites (small molecules). Proteins, however, may be even more useful, since they directly reflect inflammation and could be targeted with medicines.

The Outcome Monitoring After Cardiac Surgery (OMACS) study at the Bristol Heart Institute has already collected blood samples and health data from hundreds of patients, providing an excellent resource for this research.

Project aims

This project will:

  • Measure changes in blood proteins linked to inflammation before and after heart surgery
  • Test whether these proteins can help predict which patients develop AKI or AF
  • Explore whether combining protein data with existing metabolite data improves prediction

What we hope to achieve

By analysing these samples, we aim to discover biomarkers – measurable signals in the blood that show higher risk of AKI or AF. Using advanced statistical and machine learning methods, we will build models that combine clinical information with molecular data to predict complications more accurately.

This work could help identify patients at risk early, so that doctors can give more personalised care and reduce the chance of serious complications after surgery. It will also create a valuable dataset for future research and support new grant applications to help bring these findings into routine clinical practice.