Article Text


Original research
Relationship between stroke recurrence, infarct pattern, and vascular distribution in patients with symptomatic intracranial stenosis
  1. Karthikram Raghuram1,
  2. Aditya Durgam1,
  3. Jennifer Kohlnhofer1,
  4. Ayush Singh2
  1. 1 Department of Radiology, University of Texas Medical Branch, Galveston, Texas, USA
  2. 2 Department of Internal Medicine, University of Texas Medical Branch, Galveston, Texas, USA
  1. Correspondence to Dr Aditya Durgam, Department of Radiology, University of Texas Medical Branch, Galveston, TX 77555, USA; adurgam{at}


Objective In view of recent literature suggesting that stroke recurrence and risks related to intervention may be related to plaque physiology/instability, our study sought to discern the pattern of stroke and rates of stoke recurrence as they relate to the anatomy and presentation of the underlying stenosis.

Methods Retrospective chart as well as CT and MR angiographic imaging review of patients in the institutional stroke database was performed, including identification of patient risk factors, medical therapeutic optimization, compliance, serum cholesterol (low density lipoprotein) levels, blood pressure, physical therapy referrals, follow-up clinical status (using the modified Rankin Scales), and rate of recurrent stroke. 39 patients met the inclusion criteria. We evaluated infarct pattern (embolic, adjacent perforator, or watershed) and vascular distribution.

Results Basilar artery stenosis was most likely to present as a perforator stroke and least likely to recur. Patients discharged with suboptimal medical therapy were twice as likely to have a recurrent stroke. Among patients with optimized medical therapy, no recurrent strokes were seen in patients with an embolic infarct pattern, while a 57% recurrence rate was seen in patients with a watershed infarct pattern.

Conclusions Our results suggest that hemodynamic intracranial vascular stenoses may be less responsive to medical therapy, while stenotic lesions caused by plaque destabilization or in perforator territories may benefit from aggressive medical management with delayed or staged endovascular therapy. Recurrence of stroke may be affected both by vascular territory and by aggressive risk factor control, although the latter remains difficult to evaluate.

  • atherosclerosis
  • blood flow
  • embolic
  • stenosis
  • stroke

Statistics from


Intracranial atherosclerosis is a common cause of stroke worldwide and remains a challenge to treat effectively. The SAMMPRIS (Stenting vs Aggressive Medical Management for Preventing Recurrent Stroke in Intracranial Stenosis) study nearly halted all percutaneous transluminal angioplasty and stenting due to the 14.7% risk of periprocedural outcomes of either recurrent stroke or fatality.1 Medical management has become the mainstay of treatment due to the high risk of endovascular intervention, at least in the acute setting.2 3 However, further studies are needed to assess the validity of the SAMMPRIS study before extrapolating the conclusions to all patients presenting with a transient ischemic attack or confirmed stroke.

The purpose of this study is to elucidate stroke patterns in the setting of intracranial stenosis. Expected patterns are watershed, adjacent perforator, or embolic. Based on the pattern and underlying etiology, we try to assess which type is most vulnerable to recurrent infarcts. The objective is to assess which patterns are amenable to medical management for risk reduction. For patients not amenable to medical management, they should undergo further risk benefit analysis to determine if percutaneous transluminal angioplasty and stenting is the appropriate intervention to prevent further disability or death due to a probable future stroke and what the accurate procedural risk is for that type.

Materials and methods

The institutional stroke registry was reviewed from the time period of January 1, 2014 to January 31, 2016, returning a total of 1033 patients. Information on patient age, gender, suspected diagnosis at time of imaging, types of imaging performed (particularly confirmatory MR brain imaging and CT or MR angiography), clinical findings, discharge date, and follow-up were collected retrospectively through chart review. Patients excluded from the study included those that were deceased after initial presentation and patients who did not have concurrent CT or MR angiography of the brain (in addition to MR brain imaging confirmation of an acute infarct), clinical information regarding National Institutes of Health Stroke Scale measure/category and medical therapy regimen, and at least one follow-up visit (direct or telephonic), resulting in a total of 39 qualifying patients.

For each of the qualifying patients, MR imaging and MR or CT angiography were carefully reviewed by one faculty neuroradiologist and determinations were made regarding the location of the infarct by pattern (watershed infarct, embolic infarct, or perforator infarct) (figure 1) and territory, as well as determination of the vessel demonstrating the greatest narrowing on concurrent MR or CT angiography (middle cerebral artery (MCA), internal carotid artery (ICA), or basilar artery) (figure 2). Additional clinical information procured retrospectively from the electronic medical record by the faculty neuroradiologist, radiology resident, or medical student included time to follow-up, optimal versus suboptimal medical management, rates of compliance with medical therapy at the time of follow-up, and recurrence of stroke symptoms after initial encounter/imaging. Optimal medical management was defined by both dual antiplatelet therapy and statin administration for at least 90 days after the presenting event. Low density lipoprotein cholesterol (LDL) values were collected at follow-up to assess for adequacy of statin therapy, with 70 mg/dL used as a cut-off value to determine the effectiveness of therapy. A physical therapy consult was placed for patients at the time of initial diagnosis but there was no formal method to track compliance with physical therapy other than verbal discussion with the patient at the time of follow-up, and hence this metric was not tracked. Effectiveness of blood pressure management was determined by target blood pressure (systolic blood pressure ≤140 mm Hg) achieved on follow-up visits. However, there were limitations in data collection because of variability in the follow-up provider and a non-standard recording template as this study was retrospective in nature. Additionally, some of the patients were lost to follow-up.

Figure 1

Diffusion restriction weighted axial MR imaging showing representative examples of embolic infarct pattern from right anterior cerebral artery atherosclerotic disease (left), perforator infarct pattern affecting the left thalamus (middle), and watershed infarct pattern resulting in multifocal infarcts through the left cerebral hemisphere (right).

Figure 2

Left: Axial time of flight MR angiography demonstrating high grade narrowing of the basilar artery (arrowhead). Middle: Axial CT angiography demonstrating high grade narrowing of the proximal M1 segment of the right middle cerebral artery (arrow). Right: Axial CT angiography demonstrating high grade narrowing of the supraclinoid segment of the left internal carotid artery (curved arrow).


Of the 39 patients that met our inclusion criteria, mean age at the time of initial infarct was 62 years (range 41–87 years). The male (26) to female (13) ratio was 2:1.

At the first follow-up, compliance with outpatient medical therapy was reported (by provider documentation) in 95% of cases. In 59% of cases, optimal medical therapy was administered, while suboptimal medical therapy was given in 41% of cases. Also, only 23% of our patients were able to achieve the target LDL value, suggesting the need for further optimization of care delivery and monitoring therapy in our system. Systolic blood pressure target was achieved in 43% of patients on follow-up. A total of 15% of patients were lost to follow-up; 33% of patient were discharged to a skilled nursing facility or long term care facility and were followed-up by telephone discussion with the caregiver.

Of the three infarct patterns evaluated, approximately 41% were embolic and another 41% were perforator infarcts; the remaining 18% were watershed infarcts. Evaluation of the vessels with greatest narrowing in each type of infarct pattern (table 1) showed that embolic infarcts were more likely when the high grade stenosis involved the larger/proximal vessels (50% being basilar artery narrowing and 44% being ICA) rather than the MCA (6%). As expected, perforator infarcts were most often associated with basilar artery stenosis (75%) and watershed infarcts were seen with both ICA (57%) and MCA (43%) narrowing but not with basilar narrowing.

Table 1

Infarct patterns distributed by vessel of greatest narrowing

When rates of stroke recurrence were matched against medical optimization of therapy, patients who were medically optimized had a recurrence rate of 13% while patients who were not medically optimized had a recurrence rate of 29% (table 2).

Table 2

Effect of medical optimization on stroke recurrence rate

None of the patients demonstrating embolic infarct patterns was shown to develop recurrence of stroke at follow-up, while 19% of patients with perforator pattern infarcts and 57% of patients with watershed pattern infarcts developed recurrence (table 3) and (table 4).

Table 3

Recurrence of stroke according to infarct pattern


Despite recent advances in neurointerventional techniques, treatment of intracranial atherosclerosis remains a formidable frontier. Following the SAMMPRIS trial, there has been an increased focus on identifying subsets of intracranial atherosclerotic stenosis in an attempt to streamline management.4 5 Our study looks at the occurrence of recurrent stroke in patients with intracranial atherosclerosis disease treated by medical management.

Intracranial atherosclerotic lesions can be classified by the ensuing pattern of stroke. Hemodynamically significant lesions that reduce brain perfusion in conditions of reduced cerebral perfusion pressure cause watershed infarcts.6 Lesions that result from acute plaque destabilization and resulting platelet aggregation result in embolic infarcts.7 Perforator infarcts can result from either plaque remodeling or acute plaque destabilization.8

Given the underlying pathophysiology, one could postulate that lesions with acute plaque destabilization are most likely to be responsive to medical therapy and also most likely to cause periprocedural strokes in patients that undergo intracranial angioplasty or stenting.9 10 Conversely, hemodynamically significant, flow limiting lesions would be expected to be most responsive to endovascular interventions and least likely to respond to medical management. Perforator stroke related to intracranial atherosclerosis may be more complex, with plaque remodeling from intraplaque hemorrhage being a contributor to recurrent stroke among other effects, including flow limitation or local embolism. The ‘snowplowing’ effect of intracranial angioplasty and stenting could result in periprocedural perforator compromise, more likely if there are underlying risk factors for perforator disease (such as diabetes and hypertension).11

Reflecting these postulations, our study showed a tendency for hemodynamic stenosis to be less responsive to medical therapy while embolic strokes were the subgroup least likely to have recurrent strokes. In our dataset, basilar lesions had a high association with perforator and embolic strokes but overall a lower recurrence rate of stroke. Subgroup analyses of data from the SAMMPRIS trial did not show any difference in outcome between medical management and endovascular therapy12 in the hypoperfusion subgroup, defined as symptoms related to change in position, effort on exertion, or recent change in antihypertensive, but not defined using MRI criteria. Also, stenoses in the 90–99% range also did not show any significant differences between the medical and interventional arms.

Additionally, given the vulnerable nature of intracranial stenosis caused by plaque destabilization, initial stabilization of the underlying atherosclerotic lesion with maximum and aggressive medical therapy would be prudent to minimize additional neurologic injury using current revascularization techniques. Delayed or staged management by endovascular methods is likely the most appropriate approach in these patients. Stenotic lesions in perforator rich arterial segments are probably also best managed by the same method, although these lesions remain a difficult group to manage.

Regarding medical therapy, studies have shown the superiority of dual antiplatelet therapy in the management of intracranial stenosis over aspirin monotherapy. Additionally, high dose statin therapy has also been shown to be beneficial in reducing the risk of recurrent stroke in patients with intracranial stenosis. Aggressive risk factor control has been shown to be effective in reducing the risk of recurrent stroke13 14 but some reports have questioned whether14 15 the SAMMPRIS results of medical therapy of intracranial stenosis can be reproduced in a real world setting. In our dataset, given a real world setting, 41% of patients received suboptimal medical therapy and only 23% were able to meet the target value set for LDL. Only 43% of patients met the target systolic blood pressure value on follow-up. This has prompted our group to re-evaluate the process for care delivery to this subgroup of patients and suggests an area for improvement.

Limitations of our study include the retrospective nature of the study as well as the limited sample size. Additionally, we only included patients with MRI evidence of stroke in the territory served by the stenotic artery, thereby possibly excluding patients with MRI occult stroke or transient ischemic attack.

Further directions in understanding intracranial stenosis would include studies analyzing outcomes in stenosis subtypes and their suitability for endovascular therapy, as well as addressing the efficacy of emerging techniques such as drug eluting balloon angioplasty.16 Optimization of medical therapy and having a standardized approach to medical management with consistent follow-up by stroke specialists and careful monitoring to achieve target values would also help reduce variability in results across medical centers and physician groups.

Table 4

Recurrence of stroke according to vascular territory


View Abstract


  • Contributors Contibutorship statement: JK and AS reviewed the patient electronic medical records, assembled the data sheet, performed the data analysis, and reviewed the manuscript. AD reviewed the patient electronic medical records, performed the data analysis and statistical analysis, drafted and reviewed the manuscript, and submitted the manuscript (as corresponding author). KR oversaw the project, assembled and reviewed the data sheet, performed the data analysis, and drafted and reviewed the manuscript.

  • Funding This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent Not required.

  • Ethics approval The study was approved by the institutional review board.

  • Provenance and peer review Not commissioned; externally peer reviewed.

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.