Clinical characteristics and imaging patterns of cerebral infarction with outcomes of carotid artery stenting in symptomatic carotid stenosis: An eight-year journey

INTRODUCTION

Stroke constitutes the fourth leading cause of mortality and the fifth leading cause of death in India.^[1] Extracranial carotid atherosclerosis contributes to 18–25% of ischemic strokes worldwide.^[2] Symptomatic carotid stenosis (SCS) predisposes to high recurrence rates of ischemic stroke and merits immediate management.^[3,4] The treatment for SCS has gained clarity in recent times with optimal medical management for all patients and additional carotid artery stenting (CAS) or endarterectomy for those with more than 50% stenosis.^[3,4] Less invasiveness, shorter post-operative recovery time, and the growing availability of interventionists surely make carotid stenting a more attractive option.^[3,4]

Topographical patterns of border-zone infarcts (BZI) and territorial infarcts (TIs) stand described with BZIs having more seizures, transcortical aphasia, TIA, and anomalous circle of Willis.^[5,6] Carotid endarterectomy (CEA) was previously reported to be safer with fewer complications in patients with TI than with BZI.^[7]

Patterns of cerebral infarction in symptomatic carotid disease and the outcomes of carotid stenting are inadequately addressed in Indian studies. Keeping this in mind, we compiled our 8-year data on patients with SCS who underwent carotid stenting and explored their clinical and imaging characteristics, peri-procedural complications, and predictors of outcome after stenting.

MATERIALS AND METHODS

Selection and defining the groups

Patients who underwent carotid stenting for symptomatic carotid stenting with at least six months of follow-up were included in the study. Patients with incomplete medical records, pregnant women, and patients with very severe strokes (National Institutes of Health Stroke Scale [NIHSS] ≥20) were not considered for stenting and were excluded from the study.

Symptomatic carotid artery stenosis refers to sudden onset focal neurological symptoms attributable to ipsilateral carotid atherosclerotic disease, including one or more ischemic strokes or transient ischemic attack within the previous six months.^[4] Both the European Society of Cardiology and American Heart Association/American Stroke Association guidelines currently recommend optimal medical management with antiplatelet and statins for all patients with SCS. Interventional management with CEA or CAS is recommended for >50% stenosis when the estimated periprocedural complication rate (death or stroke) is <6% for the latter and preferably performed within 14 days of the ictus.^[3,4] CAS emerged as an alternative to the gold standard CEA in 1989. In a recently conducted meta-analysis, both CAS and CEA had comparable rates of periprocedural stroke, myocardial infarction, and death.^[9]

Patients were divided into three groups based on the topographical pattern of infarcts on imaging [Figure 1]. BZIs occur at the junction of two neighboring arterial territories. Two main patterns of infarction are described under BZI: cortical border-zone infarct (CBZ) and internal BZI.^[10] The CBZ are further subdivided into anterior and posterior infarcts, with anterior infarcts at the junction of anterior and middle cerebral arteries and posterior BZI between the middle and posterior cerebral arteries (PCAs). The internal BZI are sub-cortical at junctions of anterior choroidal, recurrent artery of Heubner, and lenticulostriate arteries territories with the anterior, middle, and PCA territories.^[11] These zones are highly vulnerable to infarction due to their inherent lower perfusion pressure, which is compounded by a severe proximal arterial stenosis resulting in a hemodynamic infarct.

Close

Figure 1:

(a) Territorial infarct in left middle cerebral artery (MCA) region, (b) Right anterior cerebral artery (ACA)-MCA cortical and internal border zone, (c) Right MCA-posterior cerebral artery cortical border zone infarct (BZI), this patient had a mixed infarct pattern, (d) Mixed infarct: Internal and posterior cortical BZI, (e) Right MCA sub-cortical and ACA-MCA cortical border zone, (f) Computed tomography angiogram showing severe carotid stenosis of the left internal carotid (white arrow).

Export to PPT

An anterior circulation infarct not involving the topography of the described BZI was considered as TI and infarcts with involvement of both these regions (TI and BZI) were categorized as mixed infarcts (MIs) in this study.

RESULTS

Ninety-six cases of CAS done for SCS with a minimum of six months of follow-up were included between 2015 and 2022. There were 46 (47.91%), 12 (12.50%), and 38 (39.58%) patients in the territorial, border-zone, and MI groups, respectively. No significant differences existed among the groups in mean age, gender, vascular risk factors such as hypertension, diabetes, dyslipidemia, current smoking or tobacco chewing, and ischemic heart disease [Table 1]. Among the 12 patients in BZI, 2 (16.66%) had isolated internal BZI, and 10 (83.33%) had a combination of both internal and cortical BZIs. BZI patients had significantly more minor strokes with NIHSS <5 (P = 0.01) and none had NIHSS >15 [Table 1]. Transcortical aphasia (sensory, motor, or mixed) was observed in 25% (3/12) BZI while it was 6.52% (3/46) and 10.52% (4/38) in territorial and MIs groups, respectively. TIA was significant in BZI (P = 0.009). Circle of Willis anomalies, percentage of stenosis, treated vessel, and bilateral disease were comparable [MI vs. BZI vs. TI, Table 2]. Transient bradycardia (52/96, 54.16%) was commonly encountered during the stenting procedure, and vessel spasm (19/96, 19.79%) and transient hypotension necessitating inotrope use (11/96, 11.45%) were also noted [Table 3]. Technical success of revascularization (TICI 3 or residual stenosis <20%) was achieved in 97.87% (92/94) cases [Figures 2a-c]. Peri-procedural (<30 days) death rate was 4.1% [4/96, Table 3]. Two (2.08%, 2/96) patients with multiple comorbidities (diabetes, hypertension, and ischemic heart disease) died of cardiac arrest shortly after stenting. Two (2.12%, 2/94) other deaths occurred, one attributable to cerebral hyperperfusion syndrome (CHS) (basal ganglia hemorrhage with raised intracranial pressure) and the other to myocardial infarction [Table 3] in the 1^st week after stenting. Subsequently, between one to six months after stenting three (3.26%, 3/92) deaths occurred. One is secondary to cardiac arrhythmia and two attributable to multiorgan dysfunction in patients with multiple vascular risk factors [Table 3]. We also compared outcomes (mRS ≤2 vs. mRS >2) at one and six months after stenting using the previously mentioned variables and dichotomizing NIHSS at 10. NIHSS ≤10, early intervention (≤2 weeks after stroke), and absence of hypertension, diabetes, and ischemic heart disease were associated with favorable outcomes at the end of one month on univariate analysis [Table 4]. After six months, low NIHSS, early intervention, and absence of hypertension and diabetes were associated with a good outcome [Table 4]. Multivariate binary logistic regression analysis was applied to these significant variables. NIHSS ≤10 and absence of hypertension (P = 0.03, 0.04) at 1 month and absence of diabetes with low NIHSS (P = 0.04, 0.01) were predictors of favorable outcomes at one and six months, respectively [Table 5].

Close

Figure 2:

(a) Mixed infarct-Left middle cerebral artery (MCA) TI with MCA- posterior cerebral artery border zone infarct, (b) Pre-stenting digital subtraction angiography (DSA) showing severe >70% stenosis (white arrow), (c) Post-stent deployment DSA showing no residual stenosis, (d) territorial infarct in right MCA, (e) DSA showing tight stenosis of right internal carotid artery (white arrow), (f) Cerebral hyperperfusion syndrome – Basal ganglia hemorrhage with mass effect and vasogenic edema in same patient post-stenting.

Export to PPT

DISCUSSION

Concurrent to previous studies, BZI patients had significantly more minor strokes (NIHSS <5, P = 0.01) in comparison to TI and MI.^[3,15] In addition, transcortical aphasia (25% vs. 6.52 and 10.52 in TI and MI, respectively) was more common in BZI (though it did not reach statistical significance) akin to existing literature.^[5] TIAs were significant in the BZI group (P = 0.009), similar to existing studies.^[5] In contrast to Joinlambert et al., seizures had no predilection and were comparable among the groups (TI vs. BZI vs. MI, P = 0.84 and 0.87, respectively).^[5] The pattern of BZI, isolated internal border-zone (2/12, 16.66%), and combined (internal and cortical border-zone) infarcts (10/12, 83.33%) was in harmony with the current literature, wherein, carotid stenosis associated hemodynamic infarcts were associated with IBZ and combined (IBZ with CBZ) pattern while cardioembolic had more of cortical BZIs.^[6,15] No significant angiographic (circle of Willis or vertebral artery hypoplasia) differences were detected in BZI versus MI or TI. This agreed with ElGammal et al. and Szabo et al. and was in contrast to that of Kumral et al. who noted a higher incidence of a hypoplastic PCA in patients with BZI.^[6,16,17]

The European trials of Endarterectomy versus Angioplasty in Patients with Symptomatic Severe Carotid Stenosis (EVA-3S) showed more clinical events (death and disabling stroke) with CAS than CEA and the International Carotid Stenting Study randomized controlled trial showed more minor strokes with CAS.^[18,19] Later, these trials were criticized for the non-utilization of EPDs and the limited experience of the interventionalists. Subsequently, short- and long-term outcomes of the Carotid Revascularization versus Endarterectomy Trial showed no difference in rates of myocardial infarction, stroke, or death between CEA and CAS.^[20,21] In this study, 4.34% (4/92) suffered a minor stroke or TIA in the peri-procedural period (<30 days) which was comparable to the 5.7%, 4.4%, and 6.3% in SPACE, SAPPHIRE and EVA-3S studies, respectively.^[18,22,23] A single case of acute visual loss due to central retinal artery occlusion occurred within 24 h of stenting in an elderly lady with bilateral severe carotid bifurcation disease and multiple vascular risk factors (hypertension, diabetes, and ischemic heart disease). Intraprocedural embolization of the ophthalmic artery, atherosclerosis of the culprit vessel, and anastomosis of the ophthalmic artery with external carotid artery branches were the possible mechanisms.^[24]

Jean-Baptiste et al. and Kazandjian et al. previously reported procedural complications to be higher in BZI than in TI when these patients underwent CEA for SCS.^[7,25] In light of this, we compared procedural complications of bradycardia, vessel spasm, and inotrope use among the groups (TI vs. BZI vs. MI) and found no significant differences. Cumulative death and stroke rates were 13% in BZI versus 0% in TI in Jean-Baptiste’s study.^[25] In contrast, no deaths and a single TIA occurred in BZI against four deaths and 3 minor strokes (1 ipsilateral and 2 contralateral) in the TI and MI group in this study. No major strokes occurred in either group; however, a single case of CRAO occurred in the TI subset.

CHS comprises a clinical spectrum ranging from headache (most common) to seizures, altered mentation, new-onset neurological deficits, and life-threatening intra-cerebral hemorrhage (ICH) with raised intracranial pressure.^[26] Single-photon emission CT and CT perfusion have been used for demonstrating CHS which is characterized by a >100% increase in cerebral blood flow of the revascularized region.^[26] Hypertension and carotid stenosis of ≥90% along with poor collaterals are known to be risk factors for CHS.^[26] We maintained post-procedural blood pressure at ≤140/90 mmHg to reduce the incidence of CHS.^[26,27] Minor CHS, characterized by ipsilateral headache and seizure (9.57% and 2.12%, respectively) occurred, this agreed with the existing studies [Table 3]^[27] Ipsilateral headache peaked within 6–12 h of the procedure and improved with cerebral decongestants and stringent control of blood pressure. One patient became drowsy 12 h after the procedure but showed immediate recovery with anti-edema measures. However, two patients (2.17%) developed severe CHS, one resulting in death, and the other was discharged with mRS 4. The first patient had severe carotid disease (≥90% stenosis), postoperatively he developed severe headache, altered mentation, and worsening deficit, and imaging showed ipsilateral basal ganglia hemorrhage with midline shift necessitating an emergency decompressive hemicraniectomy. The second patient with a near total carotid occlusion developed altered mentation and imaging showed worsened vasogenic edema. Cerebral decongestants were started. Headache and mentation improved over the next 72 h, however, he had a poor outcome at discharge and after one and six months. Development of severe CHS in these patients agreed with the existing literature, wherein, most cases occurred secondary to severe carotid disease or contralateral carotid occlusion.^[26,27] ICH occurring in the immediate post-procedure period could be secondary to the rupture of perforating arteries supplying the basal ganglia, consequent to the sudden normalization of the perfusion pressure [Figures 2d-f].^[26,27]

Similar to Mihindu et al., we dichotomized NIHSS at 10 and analyzed the outcomes.^[14] 64.13% (59/92) and 75.28% (67/89) of patients had favorable mRS 1 and 6 months after stenting. Of the 24.72% of patients with unfavorable mRS at the end of 6 months, the majority had mRS 3 and only those with severe strokes (3/89, NIHSS 16–20) had mRS 4. The absence of hypertension and NIHSS ≤10 after 1 month and the absence of diabetes, NIHSS ≤10 after 6 months were predictors of a favorable outcome [Table 5]. Patients with severe stroke (NIHSS ≥16) had more deaths in the immediate and one-month period after stenting (3/4, 75%). These cases also had multiple risk factors (diabetes, hypertension, ischemic heart disease). The AHA guidelines recommended early revascularization within 2 weeks of the ictus in 2006 and we tried to adhere to this (69/92, 75%).^[28] Mihindu et al. reported an unfavorable outcome (mRS ≥3) at discharge when urgent (<48 h from stroke) carotid interventions (CEA or CAS) were performed.^[14] No urgent interventions were performed in this study. Early intervention (≤2 weeks) was significant in the favorable mRS population at 1 and 6 months (0.003 and 0.027, respectively), though it was not a predictor of outcome.

CONCLUSION

With the majority of patients achieving a favorable outcome, shorter post-procedure recovery time, and few complications in trained hands CAS for SCS appear to be the procedure of choice in the near future. Furthermore, with more neurologists getting trained as neurointerventionalists, CAS may soon replace CEA at most centers with equally good outcomes. A close watch on the development of CHS in those with severe carotid disease is warranted.