ROUTINE blood tests are now showing promise as tools for predicting the severity and survival chances after spinal cord injury, thanks to recent advances in artificial intelligence. These findings could transform emergency care and resource planning for injury patients in hospitals around the world.
Why Early Spinal Cord Injury Prediction Matters
Each year, over 930,000 people worldwide experience a spinal cord injury, a condition marked by unpredictable symptoms and recovery prospects. Early and reliable assessment is critical, yet traditional neurological exams depend heavily on a patient’s alertness, limiting their usefulness in intensive care. This has left clinicians searching for fast, objective alternatives to guide intervention and care from the outset.
Machine Learning Unlocks Blood Test Power
A University of Waterloo-led study analysed data from more than 2,600 patients in the US, using routine blood measurements such as electrolytes and immune cell counts taken in the first three weeks after injury. By employing machine learning to find hidden patterns, researchers could forecast survival and injury severity as soon as one to three days after hospital admission. The accuracy of these predictions improved with more test results over time, outperforming standard severity scores like SAPS II. These patterns proved robust even without neurological exams, and were validated in a separate patient group. The results highlight the untapped power of ordinary blood samples to flag risky cases of spinal cord injury, and guide emergency management before advanced scans or specialist markers become available.
Clinical Impact and Future Directions
This research paves the way for faster, more equitable spinal cord injury care. Hospitals can now use commonplace, low-cost blood tests to get actionable predictions for spinal cord injury severity and mortality risk, even in environments without advanced imaging. Such insights could help clinicians make early treatment decisions, prioritise critical resources, and tailor care for high-risk patients.
Reference
Mussavi Rizi M et al. Modeling trajectories of routine blood tests as dynamic biomarkers for outcome in spinal cord injury. NPJ Digital Medicine. 2025;8(1):470.
