More than 100,000 people in the UK have an ischaemic stroke every year. This means a blood clot blocks an artery supplying blood to their brain. Patients with ischaemic stroke often have sudden weakness affecting half of their body or face, and problems with speech, although other stroke symptoms are possible.

Treatment with a ‘clot-busting’ drug (thrombolysis) can reduce the amount of disability and death caused by ischaemic stroke by restoring the blood supply to the brain. Thrombolysis usually needs to be given within four and a half hours of stroke occurring. Unfortunately, nearly three quarters of patients with ischaemic stroke arrive in hospital later than this, or it is not clear when their stroke started. For these patients, it may still be possible to treat them with thrombolysis if the hospital can provide an additional advanced type of brain scan. However, many hospitals in the UK and worldwide cannot offer this advanced scan to patients with stroke, particularly hospitals that are not in major cities or developed nations. Patients arriving at these hospitals therefore do not currently have the same access to effective treatment for stroke.

We have developed a simple method for identifying which patients can be given thrombolysis even if they arrive at hospital later than four and a half hours, or where there is uncertainty about when their stroke started. Our method does not require any additional or advanced imaging, only the standard computed tomography (CT or CAT) scan that all patients with stroke get when they arrive at hospital. Our method is called the CT Clock. We ask doctors to look for stroke changes indicating ischaemic stroke in the brain on CT. If they find these stroke changes, they measure them compared to normal brain. If the stroke changes on CT are minor (less than 20% darker than normal brain), or the scan appears normal despite quite severe stroke symptoms, we would consider these patients suitable for treatment with thrombolysis.

In this study, we invited doctors working in the front-line of stroke care to try our method during their normal working day when looking after patients with acute stroke. We wanted to know how often they were able to use our CT Clock method in real life and we asked them whether they found it easy or difficult to use. We also checked how long it took them to use the method.

Thirty five patients with stroke agreed to join our study. On average, three doctors looking after each of these patients completed a CT Clock assessment. In total, 95 doctors took part. This included a mixture of radiologists (73) and stroke clinicians (22), some of these doctors were still in training (44). These doctors provided 130 CT Clock assessments, and all attempts to use our method were successful. On a scale of 1 (easy) to 5 (difficult), doctors mostly found it easy to find (average 2) and to measure (average 1) stroke changes on CT. Most CT Clock assessments (92%) took less than 5 minutes to complete.

We conclude that it is feasible for front-line doctors providing stroke care to use our CT Clock method during their normal working day. These results allow us to proceed with planning for a clinical trial of the CT Clock.

Introduction

When ischaemic stroke onset time is unknown, eligibility for thrombolysis and thrombectomy requires CT perfusion or MRI, but often only non-enhanced CT is available. We have developed a method for estimating ischaemic stroke onset time and brain tissue viability from non-enhanced CT alone, the CT Clock. We aim to test the feasibility of front-line clinicians using the CT Clock during acute stroke care.    

Participants and methods

Prospective single centre feasibility study. We recruited patients with ischaemic stroke, excluding lacunar syndrome and minor stroke (National Institutes of Health Stroke Scale, NIHSS ≤4). We recruited stroke clinicians and radiologists to apply the CT Clock (measure ischaemic lesion attenuation) in the acute setting. Primary endpoint: proportion of clinicians successfully using the CT Clock. We also assessed ease of use on Likert scales (1 Easy to 5 Hard) and time taken.

Results

We recruited 95 clinicians: 73 radiologists and 22 stroke clinicians; 51 consultants and 44 in-training. We recruited 35 patients in 12 months: 23 male (66%), median age 69 years (interquartile range, IQR 58-72); median NIHSS 8 (IQR 6-15) scanned at median 113 minutes after stroke. Clinicians provided 130 CT Clock assessments, median 3 per patient. All attempts were successful. Clinicians found it easy to find (median 2, IQR 1-3) and measure (median 1, IQR 1-2) ischaemic lesions (n=72). Most assessments (120/130, 92%) took <5 minutes.

Discussion and conclusion

It is feasible for front-line clinicians to use the CT Clock during acute stroke care. We are now planning safety and efficacy testing.