Article Text


Stroke outcome measures: a tale of two trials
  1. Peter A Rasmussen
  1. Correspondence to Dr Peter A Rasmussen, Neurosurgery Department, Cerebrovascular Center, Cleveland Clinic, 9500 Euclid Ave. S-80, Cleveland, OH 44195, USA; rasmusp{at}

Statistics from

We are living in a medical world where our practice is dictated by proven medical outcomes as determined by multicenter randomized clinical trials. These types of trials lead to ‘evidence-based medicine’ and ‘best practices’ in how to manage our patients. Recently, two multicenter government-funded randomized clinical trials reached endpoints which would appear to provide us with guidance in the management of subacute and chronic cerebral ischemia. These two trials report that ‘aggressive medical management (AMM)’ is superior to AMM coupled with a surgical or endovascular procedure. Careful examination of these trials, in my opinion, casts significant doubt on the validity of their conclusions and, more importantly, the current assessment metrics used to track outcome for cerebrovascular diseases.

The two trials examined the role of angioplasty and stenting in the setting of severe intracranial atherostenosis (ICAD) (SAMMPRIS) and superficial temporal to middle cerebral artery bypass (STA-MCA bypass) in the setting of symptomatic carotid occlusion (COSS). Both trials were stopped prematurely, one (COSS) ostensibly secondary to futility and the other (SAMMPRIS) secondary to an increase in serious adverse events in the surgical (endovascular) treatment arm. Both trials used stroke as the primary endpoint and both trials ignored the disabling and unnerving effects of frequent transient ischemic attacks (TIAs). In addition, within this broad category of stroke, the only further clarification and scale of severity was the highly subjective metric disabling versus non-disabling stroke.

A subgroup analysis of the International Subarachnoid Aneurysm Trial (ISAT) known as N-ISAT1 examined the neurocognitive outcome of patients who had a ‘good outcome’ (modified Rankin Score (mRS) 0–1) in the ISAT trial. Detailed neuropsychological testing revealed that up to 40% of patients with a good outcome according to the mRS had significant impairments in at least one of six domains of cognitive function. Previous work has documented similar findings with good outcome as defined by the Glasgow Outcome Scale.2 These facts certainly led directly to the concept that the mRS is too coarse an outcome instrument to accurately assess subtle improvements in neurologic function or subtle declines that may occur as well in response to a surgical or endovascular procedure. Empirical observation has shown that it is more common than not that, following STA-MCA bypass, patients report that they feel better, think more clearly, find words easier, feel less depressed and generally enjoy life more following surgery than before. For example, a recent patient (mRS 1 pre-procedure) with a small stroke secondary to chronic carotid occlusion who was experiencing daily TIAs for over a year (yet no subsequent stroke occurred) reported the day after surgery that his sense of humor had returned. Since the day of his original stroke he had lost the capacity to tell a joke or comprehend humor. Postoperatively his mRS remained 1; quite obviously, the mRS lacks the capacity to account for such a major quality of life improvement. Imagine spending the rest of your life without humor. This very high level cognitive function is given back (it is doubtful if it would have returned with continued AMM), yet we do not have the ability to objectify such a crucial quality of life improvement with the mRS. In other words, if the patient had been enrolled in COSS he would not have been documented as having any benefit over AMM. No points would have been deducted from AMM for persistent TIAs yet, because of surgery, he is relieved of daily TIAs and is given back his sense of humor—no points awarded for this improvement.

Another patient with a small stroke (mRS 1) with severe left MCA stenosis has repeated daily hemiplegic TIAs and aphasia despite ASA/clopidogrel. He is an owner of two small businesses and swims 0.25 miles every day, has a normal lipid profile and his blood pressure is well controlled. SAMMPRIS tells us this man is best treated with AMM; but, he is already doing this. Because of the daily TIAs, he is considering selling his businesses (he is not safe to drive between offices) and he is going to give up swimming (recently, while swimming, he developed a TIA where he almost drowned—unable to swim and unable to call out for help). Which of the SAMMPRIS endpoints is going to report on his lost income, reduction in exercise and decreased overall quality of life?

Is the mRS the proper outcome metric for cerebrovascular disease? Does it have the ability truly to assess how a person gets on in the world after a stroke event? Clearly, this tool is woefully inadequate to assess a human's ability to function within his/her world. No finer example of the deficiency of this metric exists than in applying it to the person of Stephen Hawking. Even an inexperienced observer would classify him as mRS 4, a score hardly reflective of the contributions of one of the finest physicists of the 20th century.

The field of ophthalmology was confronted with similar difficulties with cataract surgery.3 The ophthalmologic correlate of stroke as an outcome measure is visual acuity. As we are beginning to understand, they discovered a lack of sensitivity to a range of functional visually oriented behaviors when measured visual acuity was 20/40 or better. To overcome this lack of fidelity in this particular outcome metric, a 14 point Visual Function Index (VF-14) was developed and has been shown to be highly correlated with both patient outcome satisfaction and visual difficulties. It has been found to have the resolving power needed to capture subtleties of the role of vision in day-to-day life.

Health outcomes, according to the Agency for Healthcare Research and Quality website, seek to understand the end results of particular healthcare practices and interventions. End results include effects that people experience and care about, such as change in the ability to function. In particular, for individuals with chronic conditions where cure is not always possible, end results include quality of life as well as mortality. By linking the care people get to the outcomes they experience, outcomes research has become the key to developing better ways to monitor and improve the quality of care.

It is within these last two paragraphs that we will find the secret to the success of future stroke surgery and endovascular trials. We must redefine what our outcome metrics are. We must develop the neurologic equivalent of the VF-14. What do we want to measure? What is important to our patients? What, based on our collective experience, are the problems that patients have and how have we helped them? We must be careful not to exclude TIAs and their disabling effect on productive individuals from our outcome measures. We should seriously consider volume of infarcted brain as a better outcome measure for acute stroke intervention trials. This is probably a better surrogate marker of cognitive/emotional/social performance than mRS. Perhaps more detailed instruments that measure such important human activities as short- and long-term memory, building and retaining friendships, degree of social interaction outside the home, ability to enjoy life and scales of a broad range of emotional health need to be developed and validated. Let this serve as our call to resolve the vast differences that exist within the mRS1 category; within this group of individuals, improved outcomes from surgery and endovascular therapy will be found.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.