Improvements in Delivering Care to Patients Following an Acute Stroke

This review was submitted from Beaumont Hospital, Geriatric Medicine Journal Club.  It was compiled by Dr Barry Moynihan, Consultant Geriatrician.

An editorial comment is provided by Dr Tim Cassidy, Consultant Physician and Geriatrician in St Vincent’s University Hospital.

We are interested in any additional comments you have about this paper and initiatives and resources that your local services are accessing or using to improve the care delivered to people who have had a stroke.   We look forward to you sharing any tips, hints or informed comments you may have and that you can post on the topic of “Stroke”.


Improvements in Delivering Care to Patients Following an Acute Stroke

Dr B Moynihan

The DAWN trial

Thrombectomy 6 to 24 Hours after Stroke with a Mismatch between Deficit and Infarct

Nogueira et al N Engl J Med. 2017 Nov 11


Mechanical thrombectomy (MT) for large vessel occlusion (LVO) in acute ischaemic stroke is now established as standard care up to 6 hours from stroke onset. DAWN used advanced CT or MRI imaging to test MT versus medical care in patients who were treated between 6 and 24 hours from when they were last seen well.

206 patients were randomised. Patients had clinical stroke with at least moderate stroke severity (NIHSS of 10 or more) and proven occlusion of the middle cerebral artery (MCA) or internal carotid artery (ICA). They also needed to have a volume of ischaemic brain tissue at risk but without large infarction using CT or MR perfusion. 25% of the patients were aged 80 or over. There were two co-primary endpoints based on the modified Rankin score (mRS) and a weighted mRS. Independence was defined as mRS 0-2.

DAWN stopped early due to the results of a pre-specified interim analysis. 53% of patients were wake-up strokes. 80% were MCA occlusions. Median NIHSS was 17. Median infarct volume was small (around 8mls). The main trial result was a clear difference in independence at 3 months - 49% of the MT group versus 13% in the control group. Intracerebral haemorrhage was non-significantly more common in the MT group (6 v 3%). There was no difference in mortality.

DAWN showed that using clinical and radiological measures, MT improves outcomes in patients from 6 to 24 hours from stroke onset.


Stroke is a leading cause of adult disability and ischaemic stroke accounts for 85% of stroke. Up to 40% of acute ischaemic strokes have a LVO. 8 trials have now reported improved outcomes with MT versus best medical care in LVO. The HERMES meta-analysis confirmed benefit up to 6 hours based on 5 initial studies. International guidelines state MT is standard of care in patients with appropriate M1/ICA occlusions up to 6 hours.

What does DAWN add? Many patients are admitted beyond 6 hours. The DAWN trial looked at this patient population, although over 50% of patients were wake-up stroke. The trial is notable for using advanced CT or MR perfusion to select patients with viable tissue and no large core infarct. The difference in outcomes was stark in favour of MT with poor outcomes without recanalisation. In centres that can perform CTP or MRP, there is an evolving evidence base for treating these patients with MT. How many people will this apply to in routine clinical practice? The DAWN authors felt one-third of patients would meet their criteria but did not report their own data.

The DAWN trial has moved LVO treatment from a time-based to a tissue-based paradigm. DAWN has identified a group with small cores that benefit from MT. However, patients whose imaging is not so favourable may still benefit from MT. Plain CT with ASPECTS, and multiphase-CTA can give relevant information on tissue status although further trials are needed. Given the natural history of untreated LVO, identification of suitable patients for MT is essential.

One barrier to widespread adoption of CTA/CTP is the availability of Radiology expertise on call. The EXTEND-IA and DAWN studies have shown that automated CTP analysis can help. Automated CTP software (eg MISTAR programme used in DAWN) can give near-instant results without specialist training. The proportion of patients eligible for MT has expanded significantly and all hospitals need to rapidly develop the imaging capability to identify and appropriately treat such patients.

One note of warning; this trial does not allow more time to decide on treating patients with LVO. Time is brain. Rapid access to CT and CTA with early contact with interventional centres will increase the proportion of patients benefiting from MT, reducing the disability caused by stroke in Ireland. It’s up to all of us to advocate for our patients in this exciting era of evidence-based stroke care.

Take-home messages

  1. Large vessel occlusion is common in acute ischaemic stroke. Untreated LVO is associated with poor outcomes.
  2. Reperfusion with mechanical thrombectomy up to 6 hours is now standard of care for LVO.
  3. The DAWN trial extends the time window for treating LVO with MT to 24 hours by selecting patients using advanced CT or MR imaging.
  4. Automated CT perfusion software is readily available.
  5. All acute hospitals need to develop 24/7 rapid access to plain CT, CT angiography and CT perfusion to allow patients to benefit from reperfusion. Time is brain.



Editorial Comment – Dr Tim Cassidy


The Changing Stroke Treatment Paradigm

For two and a half thousand years doctors have made pessimistic and nihilistic statements about treatment of the acute stroke patient. The NINDS study in 1995 and subsequent ECASS studies demonstrated benefit with intravenous thrombolysis within a time window, initially of 3 hours and subsequently 4.5 hours in the acute ischaemic stroke patient. This led to the theory of “time is brain” and by clearing the occlusion and restoring blood supply the patient received benefit. However, in patients with a Large Vessel Occlusion (LVO), IV thrombolysis often did not clear that occlusion and the patient was left with significant disability and handicap.

In 2015, 5 major endovascular thrombectomy trials (EVT) were published all suggesting benefit in the acute stroke patient with a LVO. The meta-analysis of these trials (Hermes collaboration, Lancet 2016) clearly showed benefit up to 5 hours. But in patients randomized after 5 hours and patients with tandem lesions there was no significant benefit on functional independence after thrombectomy.

The publication of the DAWN tria (NEJM 2017), summarized here suggests benefit for thrombectomy in patients 6 to 24 hours after stroke. Before widely implementing these new guidelines, thought has to be given about applying research evidence into clinical practice.

The Dawn trial used advanced neuro-imaging to identify a subgroup of patients with a small ischaemic core and a large area of potentially salveagble tissue. This suggests that we move away from the traditional time window to a tissue defined (biological) window. The question remains of how many patients would meet the entry criteria for the Dawn trial. While the investigations for the Dawn Trial would suggest a third of patients would meet the entry criteria, no screening logs were kept. Screening logs from 1 trial suggested only 1% of stroke patients were eligible (Campbell et al. NEJM 2015; 372: 1009 - 18).

In the clinical world, we extrapolate the research to our clinical scenario. Patients were included in the Dawn Trial who were all previously independent (mRS of 0 – 1). This leads to the uncomfortable question of do we extrapolate the research to the stroke patient who is seen in the ED with a mRS of 2 or greater. In addition, subgroup definitions are constrained by original trial criteria. Small numbers of patients with distal anterior circulation vessel occlusion were included and no patients with posterior circulation occlusions were included. The benefit of EVT in these patients therefore remains unproven.

What is certain is that the stroke physician needs to change from a time window model to a tissue based algorithm. This requires access to advanced neuro-imaging to allow these decisions to occur. In order to achieve this hospitals must evolve to develop these systems or we must develop Regional Stroke Centres to meet these increasing and important demands.