Response to Letter by Phan et al
We thank Dr Phan and his colleagues for their comments. They pointed out that the relationship between internal border-zone (IBZ) infarcts and severe carotid artery stenosis/occlusion has not been demonstrated in some studies. As we mentioned in our article, this inconsistency is likely to have resulted from (a) not using diffusion-weighted imaging (DWI) and (b) not considering the stenosis of the middle carotid artery.1 The articles that Dr Phan et al mentioned are no exception.
We agree that stroke topography does not equate with stroke mechanism, ie, hemodynamic versus nonhemodynamic. In addition, it may be difficult to separate these 2 mechanisms of stroke because a synergic association of these 2 mechanisms has been postulated.2 As Phan et al commented, the use of hemodynamic and nonhemodynamic mechanisms rather than the topographic terms for border-zone infarcts may be more ideal. However, the pathological and imaging characteristics of border-zone infarcts are well described and easily recognizable in clinical practice, while it is impossible to perform comprehensive workups in all patients with border-zone infarcts. More importantly, different types of border-zone infarcts may require different therapeutic approaches. In a previous analysis of DWI lesion patterns, an unstable hospital course (often with enlargement of the infarct) and poor long-term outcome were frequently observed in patients with IBZ infarcts, whereas recurrent strokes after the index stroke were more common in those who had small superficial infarcts, including cortical border-zone (CBZ) infarcts. These findings suggest that the DWI lesion pattern is helpful in recognizing the likely differences in the early prognostic end points after ischemic stroke, and the topographic approach may guide targeted interventions to prevent negative outcomes.3
Concerning the topographic approach, it is well known that the vascular territories can vary greatly across individuals, which might cause a territorial infarct to be misinterpreted as a border-zone infarct. However, almost every article on border-zone infarcts has used the topographic site as the definition. (How else could one define a border-zone infarction?) This is perhaps an inherent limitation of all studies of border-zone infarction.
The significant difference in the degree of stenosis (Figure 2 in the article1) was not the sole evidence that led us to propose different mechanisms in the 2 stroke groups. We also found that patients with IBZ infarcts had fewer small cortical infarcts outside the border-zone territory, and more often assumed a rosary pattern, both of which indicate that IBZ infarction is closely related to hemodynamic compromise. Recently, a comprehensive review of cerebral perfusion studies examined the mechanisms of border-zone infarcts in carotid disease.4 It showed that although severe hemodynamic compromise appears to underlie combined CBZ and IBZ infarction, artery-to-artery embolism might play an important role in isolated CBZ infarcts. Although Phan et al worry about there being insufficient evidence to rule out embolic sources, we believe that the differences we have put forward for the 2 groups strongly indicate that these 2 stroke syndromes have different pathogeneses. Increasing applications of diffusion- and perfusion-weighted imaging and microemboli detection using transcranial Doppler ultrasonography may allow a better understanding of the mechanisms of border-zone infarcts.
Yong SW, Bang OY, Lee PH, Li WY. Internal and cortical border-zone infarction: clinical and diffusion-weighted imaging features. Stroke. 2006; 37: 841–846.
Bang OY, Lee PH, Heo KG, Joo US, Yoon SR, Kim SY. Specific DWI lesion patterns predict prognosis after acute ischaemic stroke within the MCA territory. J Neurol Neurosurg Psychiatry. 2005; 76: 1222–1228.
Momjian-Mayor I, Baron JC. The pathophysiology of watershed infarction in internal carotid artery disease: review of cerebral perfusion studies. Stroke. 2005; 36: 567–577.