Near-infrared spectroscopy (NIRS) is a non-invasive optical technique that estimates brain oxygenation by measuring oxygenated and deoxygenated haemoglobin. It is also been used to assess cerebral autoregulation (CA)—the brain’s ability to maintain stable blood flow despite changes in blood pressure. This thesis of Nick Eleveld evaluated what NIRS measurements represent and assessed a novel NIRS-only method for estimating CA without requiring arterial blood pressure measurements.
A systematic review confirmed that NIRS signals are substantially influenced by superficial tissues outside the brain, while a clinical study in patients undergoing treatment for acute ischaemic stroke showed that a commercial NIRS device failed to detect expected changes in brain oxygenation. Together, these findings highlight limitations in interpreting cerebral NIRS measurements.
To improve CA assessment, a semi-automatic algorithm (Haemosync) was developed to detect and correct timing shift between physiological signals.
The NIRS-only CA method was evaluated in healthy volunteers during experimentally induced systemic inflammation and in intensive care patients. The technique detected physiological changes and was feasible in critically ill patients after extensive preprocessing. However, inter-individual variability remained high and associations with clinical outcome were limited.
Finally, the NIRS-only approach was validated against an established CA method based on arterial blood pressure and transcranial Doppler ultrasound in patients undergoing carotid surgery. The two methods showed poor agreement, indicating that the NIRS-only technique does not yet provide a reliable substitute for reference methods. Overall, the thesis identifies both the potential and the current limitations of NIRS-based cerebral monitoring, emphasising the need for further validation before clinical implementation.
