LAUSR

To the Editor: The recent article by Feng et al, “Stroke Mechanisms in Symptomatic Intracranial Atherosclerotic Disease: Classification and Clinical Implications,” is an interesting analysis of stroke mechanisms in symptomatic intracranial atherosclerosis (ICAS), which is crucial for identifying the most effective secondary prevention therapies. The authors report good-to-excellent intrarater and interrater agreement classifying the probable mechanism of the index stroke event. Patients were followed for 1 year to determine the rate of recurrent stroke or transient ischemic attack in the index stroke parent artery’s vascular distribution. The authors found a higher risk of recurrence in patients with a mix of arteryto-artery embolism and hypoperfusion. This study has several major limitations. First, only 17 patients had recurrent stroke in 1 year. The low number of outcome events prevented several crucial analyses, the most important of which is analyzing which of the infarct patterns is associated with the high risk of recurrence. In addition, due to the small number of outcomes, the authors could not adjust for important potential confounders, such as stenosis severity of the qualifying symptomatic ICAS, collateral status, or differences in medical management. Second, hypoperfusion was present in at least 91 patients (59.4%; 46 with isolated hypoperfusion and 45 with artery-toartery embolism and hypoperfusion). On the contrary, 50% (61 of 122) of patients had <70% stenosis of the affected artery. Therefore, some patients with luminal stenosis <70% were coded as border-zone infarcts, which contradicts prior studies showing that impaired cerebral blood flow is unlikely when the luminal stenosis is <75%.2 Third, this study does not differentiate between cortical and internal border-zone infarcts, which may not portend a similar risk of recurrence. In fact, one study suggests that while internal border-zone infarcts are likely related to impaired distal perfusion, cortical border-zone infarcts are more likely related to reduced clearance of emboli in the cortical border-zone territory and less likely related to severe stenosis, occlusion, or impaired distal perfusion. Fourth, the lack of perfusion imaging is a major challenge. It has been suggested that the presence of a mismatch profile on perfusion imaging is associated with a significantly increased risk of recurrent stroke as compared with those without a target mismatch profile. Perfusion imaging offers important information to adjudicate stroke mechanism that is lacking with infarct pattern alone, which does not determine whether there is viable tissue at risk beyond the core infarct. Therefore, it would be clinically meaningful to determine whether patients with any hypoperfusion-related infarcts have an increased risk of stroke as compared with those without hypoperfusion-related infarcts. This is important as treatment strategies for secondary stroke prevention should ideally target the underlying mechanism. While endovascular treatment of ICAS has substantial risk, those with impaired distal perfusion may constitute a subgroup that could benefit from stenting, angioplasty, or novel treatment modalities. In fact, a recent analysis of the SAMMPRIS trial (Stenting Versus Aggressive Medical Therapy for Intracranial Arterial Stenosis) of patients with symptomatic proximal anterior circulation stenosis showed that those with border-zone infarcts had fewer adverse events with stenting as compared with medical treatment. Furthermore, it is crucial to know whether there is a difference in event rates between internal and cortical border-zone infarcts. In conclusion, although this article highlights the importance of understanding the mechanism implicated in symptomatic ICAS, it has several major limitations discussed above, and, therefore, these findings need to be replicated in future studies that adjust for several confounders and include perfusion imaging data, which is the most direct way of assessing distal blood flow and may be a more robust predictor of recurrent stroke in patients with ICAS.