Saving brain: Improving treatment for acute stroke

Posted: 22 March 2012 by Steffen Bayer and Evin Uzun Jacobson

Stroke occurs when the blood supply to part of the brain is disrupted off either by a blood clot or a haemorrhage. Stroke is a major cause of death and disability all around the world. Every year, 150,000 strokes occur in the UK and 780,000 strokes occur in the US consuming a big part of the health care budget. In 2009, stroke care costs were estimated to be $68.9bn in the US and £8.9bn in the UK, which was spent by many healthcare agencies and organizations needed for stroke care delivery including hospitals, ambulances and rehabilitation services.

Acute stroke care delivery is challenging and time critical because an urgent imaging of the brain is required to identify the patients eligible for clot busting drug treatment (thrombolysis). Thrombolysis can dramatically improve the outcomes for some patients who have suffered a stroke, but can only be given within a narrow time window (within four and a half hours of the first onset of symptoms) and only to a subset of stroke patients. Hence, when a patient suffering an acute stroke is brought to the hospital, an urgent brain scan (normally a CT scan) is needed to determine whether there is intracranial bleeding, which would rule out his or her candidacy for thrombolysis.

In this project led by HaCIRIC researchers at Imperial College London, one question that we address is how to shorten time to thrombolysis in order to improve clinical outcomes. This is a challenging question as the treatment such as thrombolysis is given only in specific centres and requires that services such as imaging and specialist care are aligned. Building on previous research carried out with colleagues at Imperial College Business School for the Department of Health to model the potential impact of innovations within the stroke care system , our current work explores ways to improve the hyper-acute phase of stroke care.

Long travel times to appropriate hospitals can be one factor making thrombolysis unavailable for some patients, especially in rural areas. Lack of out-hours services in smaller hospitals might compound the problem. For stroke patients in Scotland we investigated hospital travel distances to the nearest hospitals with or without thrombolysis provision. We observed that the average and median distance to a hospital that provides 24/7 thrombolysis service is almost twice the average distance to the nearest hospital (see figure below). Hence many more patients could reach a hospital in time and would be able to receive thrombolysis if all hospitals provided 24/7 service. Literature suggests that access to stroke centres that provide thrombolysis is also challenging in the rural US.

Delays between onset of stroke and the diagnosis based on an interpretation of a brain scan can result in missing the vital time window for thrombolysis. In our research we have considered three ways to shorten these delays and improve timely access to imaging: portable CT scanning in the A&E department; field-based ultrasound scanning in ambulances; teleradiology for the remote interpretation of CT scans.

Portable CT scanning within an A&E department can be used to identify candidates for thrombolysis treatment and avoid delays in hospital due to the time taken to transport the patient to the scanning facility. Led by a visiting Master’s student (Hendrik Ermers from the University of Nijmegen), a study was conducted to determine the cost-effectiveness of using portable CT scanning compared to usual care with fixed, non-portable CT scanning. A decision-analytic model was designed to assess the cost-effectiveness associated with portable CT scanning. The study concluded, based on effect and cost data from the literature, that using portable CT scanning to identify patients for thrombolysis is both cost saving and effective compared with the usual fixed CT scanning. However, the initial investment might create a barrier for hospitals to implement this approach.

Field-based imaging using ultrasound is proposed to improve stroke care by providing imaging in remote areas through ultrasound, hence enabling the more widespread access to thrombolysis. In a collaborative project including our colleagues, Henry Feldman from Harvard Medical School and Stan Finkelstein from MIT, we investigate point of care testing in stroke care.

CT scanning in remote areas is challenging as CT scanners are expensive and the output of a CT scan is normally interpreted by a radiologist typically at the site of the CT scan. Moreover, in rural areas the ability to bring a patient to the CT scanner in time is limited. Preliminary literature research has pointed to the feasibility of the use of portable ultrasound devices. We intend to conduct future research examining the possibility of the use of such devices within ambulances, thereby reducing the ‘time to scan’ especially in remote areas far from the nearest hospital providing CT scanning. The aim of this work is to establish whether ultrasound can be used to exclude haemorrhagic strokes so that the decision to thrombolyse the patient in the ambulance can be made.

Teleradiology is a way to provide a timely interpretation of a brain scan by a remote expert. This is of particular relevance to providing out-of-hours services in hospitals where 24/7 scanning is available, but no consultant to diagnose is present at night. In Scotland the experience of the use of teleradiology has been positive. A pilot telemedicine network in the south of Scotland was conducted in 2011, and the outcome of which has been promising. There has been planning in progress for west and north networks.

The analysis of the use of teleradiology to support out of hours imaging is a part of our collaborative research with the Scottish Stroke Clinical Audit Research Group. The aim of the work in Scotland is however broader than just the evaluation of teleradiology services. So far, we have also conducted an empirical data analysis of the data on all stroke patients in Scotland looking in particular at hospital delays. We also analyse networks of hospitals interacting with each other and compare alternative scenarios to the current system to include innovations such as reduction of in-hospital delays, field-based testing using portable ultrasound devices and teleradiology to provide out of hours services.

The HaCIRIC work on stroke care is ongoing; future work to look at some of the options available to improve acute stroke care in more depth is planned both in the form of simulation models and clinical trials.

Dr Steffen Bayer is a Research Coordinator, Imperial College London. Dr Evin Uzun Jacobson is a Research Associate, Imperial College London. Both researchers are funded by HaCIRIC.