Monthly Archives: January 2015

It is known from NASCET that CEA’s prevent stroke in symptomatic carotid stenosis. Prior literature has pointed to a restenosis rate however of 5-15%. Many have argued that lack of using a patch during the CEA increases the rate of restenosis. The idea is that closure of an arteriotomy with a patch minimizes the effect of neointimal hyperplasia and scarring, maintaining arterial lumen diameter after the procedure. A handful of retrospective studies, a meta-analysis and one small randomized trial have contributed to answering this question. While some data points to improvement of stenosis rate with application of the patch and improvement in recurrent stroke prevention as well; the data is still a mixed bag with both positive and negative results. Malas et al evaluate whether patching in carotid revascularization reduces restenosis and provides the benefit of preventing further strokes.


This study’s strength lies in the number of patients it has to evaluate. Using the CREST results, 1082 patients were included. As a reminder, CREST randomly assigned patients with symptomatic or asymptomatic carotid stenosis to undergo carotid-artery stenting or carotid endarterectomy. For this study, only the CEA patients were evaluated. Duplex ultrasounds were performed at baseline, 1,6,12,24 and 48 months. Restenosis rates at two years were assessed, defined as 70% or greater diameter reducing stenosis based on elevated peak systolic velocity of 3.0 m/s or higher.

With regards to results: 753 (65%) underwent CEA with patch and 329 (29%) underwent primary closure. 44 patients were excluded due to undergoing eversion CEA and 25 patients had missing data. By specialty, 89% of vascular surgeons, 6% of neurosurgeons and 76% of thoracic surgeons patched. There was significant reduction in the 2-year risk of restenosis, even when adjusted by surgeon specialty (HR = 0.35, p=.006). There was however no significant difference in the rates of periprocedural stroke and death (HR 1.58, P=.57) or 4-year risk of ipsilateral stroke (HR=1.23, P=0.71).

The results that Malas et al generate in terms of restenosis reduction at 2 years are quite convincing. Prior studies may have shown mixed results due to the small sample size which may not have adequately powered for a difference. In this sense, one could argue that patching should be routine for CEA patients. The study however, does not evaluate for complications of patching such as infection which is a function of the CREST data set. Curiously, even though there is improvement of the 2 year restenosis rates, there is no clear effect on the 4 year risk of ipsilateral stroke. While prior smaller studies have shown this benefit (a recent CREST analysis showed that patients with restenosis had significantly higher risk of stroke), one could argue that at 4 years, evaluating a stroke as a function of that restenosis may be not long enough of a time span to see that benefit. It is unclear what the average duration of restenosis is in these patients, but if restenosis develops at 2 years post CEA, we know that the benefits of CEA aren’t potentially seen until 5 years in moderate stenosis as per NASCET2. Thus evaluation of a stroke at 4 years (which is only 2 years after the cut off for evaluation of restenosis) may be too short of a time window to actually capture strokes which are a function of that particular stenosis.

By in large, there is likely something beneficial to patching in CEA. Multiple studies have found similar findings, although at times, the data are not completely in harmony. Although this study only hit home on one of its’ two endpoints, it still lends support to “PRO Patch,” and, in my opinion, there is enough data out there otherwise to suggest a benefit.

According to our 2013 AHA/ASA guidelines for the early management of patients with acute ischemic stroke, patients with “minor and isolated or rapidly improving neurologic signs” are defined as such:

“Minor and isolated symptoms are those that are not presently potentially disabling. Although most patients with potentially disabling symptoms will have NIHSS scores ≥4, certain patients, such as those with gait disturbance, isolated aphasia, or isolated hemianopia, may have potentially disabling symptoms although their NIHSS score is just 2.”


These guidelines are reasonable for everyday practice, as many acute strokes are mild and “minor and rapidly improving signs” is the most frequently cited reason to withhold tPA in patients with acute ischemic stroke who are otherwise eligible. However, there seems to be a fair amount of subjectivity in determining “disabling” deficits and there is nothing concrete about a NIHSS cutoff of 4. To continue what was said about minor symptoms in the guidelines:

“Several studies have now reported that approximately one third of patients who are not treated with intravenous rtPA because of mild or rapidly improving stroke symptoms
on hospital arrival have a poor final stroke outcome. A persistent large-artery occlusion on imaging, despite minor symptoms or clinical improvement, may identify patients at increased risk of subsequent deterioration. In light of these observations, the practice of withholding intravenous fibrinolytic therapy because of mild or rapidly improving symptoms has been questioned, which justifies further study.”

In light of this ambiguity, some SUNY and Tufts investigators conducted a survey of active urban academic neurologists as to how they make tPA decisions in the setting of minor acute ischemic stroke. The standardized survey was composed 110 clinical cases with permutations in NIHSS, symptom type, pattern of changes, and age/occupation.

The survey composition has its flaws and the makeup of the respondents, including only 43% of those invited, calls the generalizability into question but, NIHSS is not the only determinant of tPA provision, although it was a significant factor. Symptom type (language/neglect over motor and visual/sensory/ataxia) and age/occupation (35/violinist, 52/taxi driver, and 65/lawyer over 80/retired) were also significant predictors of tPA provision when controlling for NIHSS. These are the first quantitative data on tPA decision-making for minor acute stroke.

The findings of this survey are interesting if not unsurprising. It would be shortsighted to make tPA decisions based on NIHSS alone, but what of the symptom clusters separated them? Or are we reading too much into it because of the symptom clusters are not individual features as we discern at the bedside? And what about age? Is the withholding of tPA to the 80 year old retiree an acknowledgment of series suggesting lower tPA efficacy and higher bleed risk or ageism? As the investigators say, future similarly designed surveys (hopefully with acknowledged limitations addressed) cast additional light to better understand practice patterns in this controversial realm. That is an important step before we can constructively scrutinize “typical practice” and make recommendations for minor stroke.

Samarasekera N, Fonville A, Lerpiniere C, Farrall AJ, Wardlaw JM, White PM, et al. Influence of Intracerebral Hemorrhage Location on Incidence, Characteristics, and Outcome: Population-Based Study. Stroke. 2015

Just as in real estate, location within the brain is of utmost importance to a stroke neurologist! Intracerebral hemorrhage outcomes have unfortunately not changed as much as we would have hoped over the past 20 years, with only 60% of patients surviving a month. Those who survive have comparable risks for ischemic v/s hemorrhagic stroke in the future. Given the grim nature of the entity, there has been a lot of focus trying to identify and quantify prognostic factors so that we can help patients and their families make informed decisions. The ICH score was developed in 2001 to predict 1 month mortality, and included age, Glasgow Coma Scale, ICH location, ICH volume, and intraventricular extension as components. Among the components, age and location (supratentorial or infratentorial) were the strongest predictors. ICH grading scale (ICH-GS) was proposed in in 2007 and improved discrimination slightly by using different cutoffs and points assignment. However, location was still classified as supra versus infratentorial. As four fifths of all ICH are supratentorial, there have been few studies focussing on differences in outcome contingent on whether ICH is lobar versus non-lobar, as opposed to supratentorial versus infratentorial. The influence of supratentorial ICH location on outcome is not well established.



In the current issue of Stroke, Samarasekera et al. study the effect of lobar and non-lobar location on outcome in a prospective population-based cohort in Scotland of 695,335 people aged >=16 years. 128 incident primary ICHs between June 2010 – June 2011 were included. An experienced neuroradiologist classified ICH location into non-lobar (brainstem, cerebellum, basal ganglia, internal or external capsule, thalamus) and lobar. Lobar location, constituting 53% of the ICHs was more often associated with prior dementia, lower GCS score, larger volume, as well as subarachnoid and subdural extension. However, lobar hemorrhages were less likely to be dead at 1 year (adjusted OR 0.21, CI 0.07 – 0.63) as compared to non-lobar hemorrhages, after adjustment for known predictors. Supratentorial hemorrhages were associated with lower odds of death at one year as compared to infratentorial hemorrhages. There were 4 recurrent ICHs, all among the lobar ICH group. When the authors reclassified location in the ICH score as lobar, deep or infratentorial, it did not improve the discrimination of ICH score for outcome at 1 year. This is an important study as it validates several observations such as better outcome in lobar ICHs but higher risk of recurrence. Also, it validates the ICH score in a population based cohort and reasserts its discriminative ability. The higher prevalence of dementia in lobar ICH may stem from cerebral amyloid angiopathy.