World-first test to reduce post-surgery risks to brain health
Any surgical procedure can be risky, especially for older people. To minimise the dangers, EU-funded researchers have developed the world's first personalised test to assess the risk of patients developing post-surgical sensory and cognitive disorders, allowing doctors to choose the safest treatment.
© Gorodenkoff #224040273 , source: adobe.stock.com 2019
Post-operative delirium (POD) causes progressive deterioration of sensory or cognitive function in 30-80 % of people after major surgery. Although usually short-lived, among elderly patients in particular it is frequently followed by longer-lasting post-operative cognitive dysfunction (POCD), which can resemble chronic dementia and may accelerate cognitive deterioration in people with Alzheimers disease. Treating these conditions is estimated to cost health-care providers EUR 30 billion per year in Germany alone.
To address this challenge, the EU-funded BIOCOG project carried out ground-breaking research that could underpin a substantial reduction in the incidence of post-operative cognitive impairments via the pre-surgical testing of patients for risk factors.
Our society is growing older and that means the socio-economic implications of POD/POCD are profound, says project coordinator Georg Winterer, CEO of Pharmaimage Biomarker Solutions and head of the Clinical Neuroscience Research Group at university hospital Charité Campus Berlin-Buch. POD/POCD generally means more frequent and longer hospital stays, loss of independence and quality of life, and increased mortality, as well as a higher burden on patients families, carers and health-care services.
The project team developed a predictive algorithm that is expected to be made available to doctors as a clinical software app in the next two years, enabling personalised testing for POD/POCD risks to take place on an unprecedented scale. This will enable doctors and patients to more accurately weigh the risks and benefits of surgical procedures and make better-informed choices about treatment options.
To develop the bioinformatics-based algorithm, the BIOCOG team carried out the largest study ever conducted worldwide to identify valid biomarkers for risk prediction of post-operative cognitive impairments, involving more than 1 000 patients across Europe.
The team used cutting-edge neuroimaging solutions to obtain data on brain structure and function, such as functional magnetic resonance imaging and diffusion tensor imaging, in combination with genetic data and analyses of blood plasma to study patients before and after surgery.
This landmark research enabled the identification of a number of clinical, cognitive, neuroimaging and molecular biomarkers indicating with high accuracy the risk parameters associated with a patient developing POD/POCD. An individual pre-operative elderly patient can therefore be positioned on a multi-parameter risk scale, referenced against the worlds largest population-based databank for POD/POCD.
Beta testing ahead
According to the project team, the databank and predictive algorithm will not only support decision-making by doctors before surgical interventions, but in the longer term will enable the development of novel therapies in cases where POD/POCD does still occur.
The multivariate prediction algorithm developed in BIOCOG is a very significant development for personalised risk prediction, says Winterer. The pre-operative identification of high-risk patients will enable clinicians to assess whether a surgical intervention is really necessary or whether alternatives should be considered, such as non-surgical treatment, a less invasive form of surgery, modified pre-operative medication or a different choice of anaesthetics etc.
The solution is due to undergo beta testing in hospitals, with project partner Pharmaimage Biomarker Solutions aiming to gain regulatory approval to launch the app commercially from 2021. The product was recently named best business project in the European Institute of Innovation and Technologys Health Wildcard competition.
We will start with a simple version of the software using a set of as few as five or six parameters, says Winterer. Since the software is self-learning, we will be able to incrementally improve the risk prediction as its use becomes more widespread. This will also enable us to add additional prediction parameters, for instance for rare conditions that may increase the risk of developing POD/POCD.