After waiting nearly a year since the first COVID-19 vaccines...Leer más
COVID-19 infection is associated with a unique vascular inflammation ‘signature’ that, when present in at-risk individuals, increases the risk of in-hospital mortality around three-fold, shows an innovative data analysis of thousands of UK angiography patients.
The results, presented at the British Cardiovascular Society 2021 Annual Conference on June 10, may also explain why the cheap and widely-used steroid dexamethasone is able to reduce mortality rates in COVID-19 patients.
Dr Charalambos Antoniades, Radcliffe Department of Medicine, University of Oxford, and colleagues were already conducting a large-scale, artificial intelligence (AI)-led analysis of patients undergoing computed tomography (CT) angiograms of their heart when the pandemic struck.
They quickly repurposed their study to examine whether feature extraction of the CT scans could be combined with RNA sequencing to identify a profile of vascular inflammation induced by COVID-19 infection.
One Extra Step
The study, which was co-funded by the British Heart Foundation (BFH), showed that not only was an inflammation signature identifiable but also that it could be used to predict in-hospital mortality, particularly in those not given dexamethasone therapy.
“By simply adding in one extra step to the routine care of people admitted to hospital with COVID-19 who already have a CT scan, we can now detect patients at high risk of life-threatening complications and could potentially tailor their treatment to aid long-term recovery,” Dr Antoniades said in a press release.
“There are still a lot of unknowns relating to how the virus can impact our health in the long-term, but this AI tool could ultimately help to save lives.”
Dr Antoniades said that the benefits of the tool “don’t stop there”, however.
“We know that this exaggerated immune response to the virus can also cause abnormal blood clotting, and so we are developing this AI platform to specifically identify COVID-19 patients who are most at risk of having a future heart attack or stroke.
“We can also pivot our platform with ease to develop a new scanning ‘signature’ to better understand future viruses and diseases that take hold of our population.”
Cardiovascular Risk Prediction Consortium
The UK C19 CT Cardiovascular Risk Prediction Consortium (CRT) study was originally designed to be the world’s largest registry of cardiovascular CT scans, linked to life-long outcome data and biological factors, with the aim of developing and validating novel biomarkers.
Sixteen sites in the UK and 14 international centres are enrolled in the study, and it incorporates the Oxford Risk Factors And Non-invasive imaging study (ORFAN), for which 10,000 patients undergoing coronary computed tomography angiography (CCTA) in the UK have been prospectively recruited.
In addition, data on 100,000 patients who underwent CCTA in the UK from 2010 have been gathered, with outcomes detected via NHS Digital, the National Institute for Cardiovascular Outcomes Research, and the Sentinel Stroke National Audit Programme.
To provide insights, the data are transferred via a hub into a translational science and imaging data integration platform, an AI system that combines RNA sequencing and radiographic imagery with clinical information.
Since the study began, it has already borne fruit. For example, it showed that perivascular tissue can be used as a biosensor of vascular inflammation, with different types of inflammation giving different ‘fingerprints’ in the texture of the adipose tissue.
In the light of the COVID-19 pandemic, it was decided to expand the original study to examine whether patients with high vascular inflammation are at risk of severe COVID-19, and whether COVID-19 induced inflammation is what drives in-hospital complications.
They pulled together data on patients from Oxford, Bath, and Leicester who had undergone CCTA as part of the original analysis and gone on to contract COVID-19, and then determine whether or not they had been hospitalised.
The patients had chest CT, CT pulmonary angiogram, and blood biochemistry analyses performed, and they were followed up to see if they improved or deteriorated. In the later case, their longer-term outcomes were collated.
The team also aligned the data with those from the landmark RECOVERY trial to identify 4000 patients from their own database who had also taken part in that study.
This UK trial showed that dexamethasone reduced death rates by around a third among the most severely ill COVID-19 patients admitted to hospital.
Running the findings though the AI system, the researchers found that COVID-19 induced vascular inflammation, causing cytokine upregulation that led to changes in perivascular adipose tissue.
Combining RNA sequencing with analysis of radiographic images, they found that this inflammation was associated with a novel radiotranscriptomic signature (C19-RS) that was significantly more prevalent in CCTA patients with COVID-19 than in those without (p<0.001).
Interestingly, they also found that the UK variant of COVID-19 led to a greater degree of vascular inflammation with this signature than the variant that was circulating at the beginning of the pandemic (p=0.025).
Turning to patient outcomes, the results from the Oxford cohort indicated that patients with high C19-RS vascular inflammation in response to COVID-19 infection had worse survival than those without, increasing the risk of in-hospital mortality at a hazard ratio of 3.31 (p=0.003).
This significant association was confirmed in the Bath and Leicester cohorts, where it was shown that high C19-RS vascular inflammation increased the risk of in-hospital mortality at a hazard ratio of 2.58 (p=0.028).
“An interesting aspect of this came from an analysis of the patients who did not receive dexamethasone” as part of the RECOVERY trial, Dr Antoniades said during his presentation.
Patients with high vascular inflammation who did not receive dexamethasone had a substantially increased risk of in-hospital mortality compared with those with low vascular inflammation who remained similarly untreated, at a hazard ratio of 8.24 (p=0.002).
However, this significant difference between high and low vascular inflammation was “eliminated” when patients were treated with dexamethasone (p=0.18).
“This suggests that, most likely, dexamethasone improves outcomes by targeting vascular inflammation, which is a major finding that needs to be further investigated in a trial setting,” Dr Antoniades said.
To those ends, they are working with their current trial population to see if “this observation is a true finding”.
Créditos: Comité científico Covid