Cerebral Revascularization After the Carotid Occlusion Surgery Study

What Candidates Remain, and Can We Do Better?

Timothy G. White, MD; Hussam Abou-Al-Shaar, MD; Jung Park, MD, PhD; Jeffrey Katz, MD; David J. Langer, MD; Amir R. Dehdashti, MD


Neurosurg Focus. 2019;46(2) 

In This Article


Patient Demographics

In total, 36 patients underwent cerebral revascularization for nonmoyamoya cerebrovascular occlusive disease. One of the 36 patients was not included in the study because the cerebrovascular occlusion and subsequent hypoperfusion was secondary to carotid sacrifice during head and neck surgery. The average patient age was 55 years (range 22–74 years). All patients except 1 were on at least single-antiplatelet therapy prior to presentation. Clinical presentation included recurrent strokes, recurrent TIA, perfusion-dependent examination, hemorrhagic transformation of stroke, and headache with severe hypoperfusion. The most common presentation was recurrent stroke in 74% (26/35) of patients, followed by recurrent TIA in 11% (4/35), and perfusion-dependent examination in 9% (3/35). Both hemorrhagic transformation of stroke as well as headache and severe hypoperfusion occurred in 1 patient each. Of note, most patients presented with metabolic syndrome including the combination of hypertension, hyperlipidemia, and insulin resistance. There were 3 unique cases in which patients did not present with classic symptoms of atherosclerosis: one patient with fibromuscular dysplasia, one with sickle cell anemia, and one with antiphospholipid syndrome.

Imaging Characteristics

All patients were evaluated using multimodal imaging techniques. During the workup, patients underwent MRI followed by vessel imaging. In 57% (20/35) of patients MRI results demonstrated multiple watershed-area infarcts, which was the most frequent MRI finding, whereas other patients presented with multiple MCA but no watershed area strokes. All patients underwent some form of hemodynamic imaging. Most commonly the imaging workup included QMRA NOVA as well as SPECT with and without Diamox. QMRA NOVA was done in 60% (21/35) of patients, and in 100% (35/35) of those patients flow-limiting stenosis was demonstrated when compared to historical age-based normal values. SPECT was done in 91% of patients (32/35), and of those patients 81% (26/32) showed decreased CVR when challenged with Diamox. All patients without diminished CVR who received Diamox did not improve with the administration of the drug, and at baseline had severe hypoperfusion on SPECT along with recurrent stroke on optimal medical therapy. Patients were categorized as having internal carotid artery (ICA) steno-occlusion, bilateral ICA steno-occlusion, or MCA steno-occlusion. ICA steno-occlusion was the most common, occurring in 49% of patients (17/35), followed by MCA steno-occlusion in 31% (Table 1).

Perioperative Outcomes

Thirty-day perioperative outcomes were assessed in all patients to accurately determine perioperative morbidity in this theoretically high-risk population. Immediate postoperative graft occlusion occurred in 2 patients; therefore immediate graft patency was 94% (33/35). Postoperative complications occurred in 11% of patients (4/35). There were 3 cases of postoperative ischemic events (9% of patients), which were permanent in only 1 case. Further description of these 3 events can be seen in Table 2. All events occurred within 3 days postoperatively. There was also 1 case of postoperative seizure treated successfully with antiepileptic drugs. One patient with postoperative graft occlusion underwent a subsequent encephaloduroarteriosynangiosis (EDAS) procedure when postoperative angiography demonstrated the occlusion. In univariate and multivariate analysis no factors (age, sex, preoperative hemodynamic imaging findings, site of occlusion) were identified as being associated with the risk of a perioperative ischemic event or the likelihood of bypass patency (p > 0.05).

Long-term Outcomes

Long-term clinical outcomes were assessed in 80% (28/35) of patients, whereas long-term graft patency outcomes were assessed in 60% (21/35) of patients. Bypass grafts were found to be patent in 95% of patients (20/21) (Table 2). In one patient who experienced immediate graft failure on long-term follow-up the graft was found to be open, and in the other patient with immediate occlusion who underwent the revision EDAS procedure long-term patency was not assessed. Clinically, no patient experienced subsequent stroke; however, 1 patient did develop a persistent seizure disorder postoperatively. Duration of follow-up was on average 25.5 months (range 3–84 months).