News

Can advanced digital modelling help reduce patients’ stay in NHS Critical Care units, facilitate more timely treatment and better survival rates, and lead to lower associated costs to the health system?

Researchers at the Nottingham BRC are working on advanced computational modelling to mirror human respiratory and cardiovascular systems, with the ambition of applying the technology to critically ill patients.

According to data from 2019, 172,000 patients required critical care treatment each year in the UK, with a mortality rate of 20%. Almost half of these patients needed specialist support for two or more organ systems (including mechanical ventilation) before they were well enough to be discharged from Critical Care units.

Making better use of data

Figures from NHS England show a typical stay for an adult in the Critical Care Unit is around six days, at an annual cost of £2 billion. This excludes the cost of stays on burns units.

Assistant Professor Dr Marianna Laviola of the Nottingham BRC Musculoskeletal, Surgery, Inflammation and Recovery Theme, is working with collaborators in the UK and internationally on research with the aim of making better, real-time use of data from patients’ vital signs.
As well as reducing patients’ length of stay in Critical Care units and improving survival rates, the project aims to reduce human suffering following trauma.

The digital twin

Dr Laviola and her team - which includes Professor Jonathan Hardman - are working with Critical Care specialists including those at Nottingham University Hospitals; they aim to create a ‘digital twin’.

This in future will enable exactly mirroring of a patient and their vital organs, facilitating prompt and accurate treatment once a patient is admitted to a Critical Care unit, and reducing the ‘guesswork’ required in critical care diagnosis and organ support.

Over the last 25 years, a computer simulation model has been developed with 2,000 lines of code, 100 physiological variables and 250 parameters. Having already achieved robust simulation of a patient’s organ systems offline, the next step is to create a real-time simulation of patient pathophysiology, monitoring vital signs and providing a platform for testing planned interventions.

The next phase in the project includes development of “personalised and optimised resuscitation strategies” tailored to each individual patient, according to Dr Laviola. This will build on previous work encompassing computer modelling in the treatment of cardiac arrest.

Treatment protocols “could be improved”

Research by Dr Laviola and colleagues published in the journal Resuscitation last year, and funded by the National Centre for the 3Rs, evaluated the current guidelines for cardiopulmonary resuscitation, using a computational model of the cardiopulmonary system.

Investigators developed and validated their computational model against available human data. They used an algorithm to find CPR protocol parameters that optimised results associated with the return of spontaneous circulation in 10 virtual subjects.

Their results indicated that current protocols for CPR have the potential for improvement, resulting in improved organ oxygenation during cardiac arrest, with the potential for improved survival rates. The authors recommend that future clinical trials aimed at developing CPR protocols should specifically consider interactions between chest compressions and ventilation.

The journal describes computational simulation as “a novel and promising alternative to animal and clinical trials”, allowing “complete reproducibility of methods and results, and offers mechanistic insight into the effectiveness of treatment strategies”.