7 Role of Leptomeningeal Collaterals
7.1 Case Description
7.1.1 Clinical Presentation
A 72-year-old female presented to the emergency department with a 90-minute history of sudden-onset right hemiparesis, global aphasia, and neglect. Initial National Institutes of Health Stroke Scale (NIHSS) score was 25. Past medical history was significant for hypertension, dyslipidemia, and atrial fibrillation. Medications at the time of presentation included aspirin 81 mg daily and multiple antihypertensive agents. She had previously been anticoagulated with warfarin; however, this had been discontinued 2 years previously due to episodes of significant epistaxis.
7.1.2 Imaging Workup and Investigations
Noncontrast computed tomography (NCCT) and CT angiography (CTA) were completed within 2 hours from the onset of symptoms showing left M1 occlusion (Fig. 7.1).
Fig. 7.1 Non–contrast-enhanced CT brain (a–c) demonstrated a hyperdense vessel sign in the region of carotid termination and proximal M1 segment on the left (7.1a, arrow). There was loss of gray-white matter differentiation in the region of the caudate nucleus and insular ribbon on left (ASPECTS score of 8). No other early or established ischemic change was seen. There was no evidence of hemorrhage. Single-phase CT angiogram performed from the level of the aortic arch demonstrated occlusion of the left carotid terminus and proximal M1 segment of left MCA. There was, however, good leptomeningeal collateralization of the occluded left MCA territory, with filling of left MCA branches back to the M1 segment and distal face of the thrombus (d–f). The thrombus could be seen as a relatively short length filling defect in the vessel (d, arrow). Note was made of relative paucity of vessels in the region of the left lenticulostriate territory compared to the contralateral side (e, arrow) with absent opacification of the left internal cerebral vein. The left ACA was patent, filling through a patent anterior communicating artery.
7.1.3 Diagnosis
Left carotid terminus occlusion with good leptomeningeal collateralization to left middle cerebral artery (MCA) territory.
7.1.4 Treatment
Full-dose intravenous tissue plasminogen activator (IV-tPA) was administered which was initially used as a bridge to mechanical thrombectomy.
Material Used
An 8-Fr short angiographic sheath; 8-Fr MERCI balloon guide catheter; 5-Fr VTK slip catheter; 0.035 Terumo Advantage guidewire wire; Synchro 14 microguidewire; Trevo 18 microcatheter; Trevo Pro 4 × 20 mm stent retriever; 8-Fr Angio-Seal closure device.
Technique
Intervention was performed with conscious sedation, and local anesthetic. A single-wall right common femoral artery puncture was performed, and the 8-Fr short vascular access sheath inserted. Following puncture, 2,000 international units of heparin was administered intravenously.
The 8-Fr balloon guide catheter was advanced to the left internal carotid artery (ICA) over a 5-Fr VTK slip catheter with the aid of an advantage guidewire. Left internal carotid angiogram demonstrated persistent occlusion of the left carotid terminus (Fig. 7.2), with abrupt cut-off of contrast in the supraclinoid ICA at the level of the anterior choroidal artery origin.
A Trevo 18 microcatheter was navigated through the occluded left carotid terminus and M1 segment and into the left proximal inferior division of left MCA. Control injection through the microcatheter (not shown) confirmed position distal to thrombus in a patent good caliber M2 branch. A 4 × 20 mm Trevo Pro stent retriever was then deployed from the proximal M2 to the level of the supraclinoid ICA. The tip of the stent retriever is depicted by arrow in Fig. 7.2a.
After leaving the stent in situ for 5 minutes to allow incorporation of thrombus, retrieval was performed with flow arrest and continuous aspiration through the balloon guide catheter. Thrombus fragments were retrieved from the stent and aspiration tubing and further thrombus was obtained from aspiration of the guide catheter. The guide catheter balloon was deflated.
Control angiogram showed complete recanalization of the index lesion, and complete reperfusion of the distal MCA territory with no evidence of thromboembolic complication. The left ACA territory was patent and was now seen filling from the left ICA injection. The procedure was completed within 4 hours of symptom onset.
All devices were removed. An 8-Fr Angio-Seal closure device was placed for hemostasis.
Fig. 7.2 Left internal carotid angiogram showing persistent occlusion of the left carotid terminus (a), with abrupt cut off of contrast in the supraclinoid ICA at the level of the anterior choroidal artery origin and tip of the stent retriever (arrow). Control angiogram showed complete recanalization (b) after stent retriever and adjunctive suction, and distal reperfusion (c,d).
7.2 Postprocedure Care/Outcome
The patient demonstrated on table improvement to NIHSS score of 7, with mild right sided weakness and dysphasia remaining. Postprocedure the patient was transferred to high-dependency stroke unit for further care.
Repeat NCCT of the brain was performed at 24 hours postprocedure (Fig. 7.3). This demonstrated low attenuation change consistent with small volume infarction in the left caudate and lentiform nucleus, as well as insular and frontal opercular region. Gray-white matter differentiation was otherwise preserved in the left hemisphere with no evidence of more widespread infarction. Following CT, the patient was commenced on 81 mg aspirin daily. In view of the atrial fibrillation and prior history of significant epistaxis, the patient was assessed by the ENT service, and subsequently recommenced on anticoagulation at 10 days poststroke. Blood pressure control was optimized and lipid lowering medication was commenced. The patient continued to improve throughout her inpatient stay with almost complete resolution of symptoms, and she was discharged 5 days following presentation to rehabilitation.
7.3 Companion Case
7.3.1 Clinical Presentation
A 75-year-old male presented to the emergency department with a 90-minute history of sudden onset right facial droop, hemiparesis, global aphasia, and neglect. NIHSS score was 28. Past medical history was significant for hypertension, dyslipidemia, and atrial fibrillation. Medications at the time of presentation included warfarin, with INR of 2.7 on testing.
7.3.2 Imaging Workup and Investigations
NCCT of the brain and CT angiogram were completed within 2 hours of symptom onset. CT of the brain (Fig. 7.4) demonstrated loss of gray-white matter differentiation in the left insula, and cortical zones M5 and M6. CT ASPECTS score 7. Hyperdense vessel sign was noted in the region of the left carotid terminus and proximal M1 segment.
Single-phase arch-to-vertex CT angiography was performed. This demonstrated occlusion of the left supraclinoid ICA to carotid terminus as well as nonfilling of left A1 segment of ACA and left MCA. There was filling of the more distal left ACA territory presumably through the A Comm from the contralateral side. Axial reconstructed MIP images demonstrated overall poor collateralization to the majority of the left MCA territory. There was some filling of leptomeningeal collaterals in the left posterior temporal and inferior parietal region but relative paucity of collaterals elsewhere in the left MCA territory, particularly in the left frontal and anterior parietal region.
Fig. 7.4 NCCT within 2 hours of symptom onset showing (a–c) loss of gray-white matter differentiation in the left insula, and cortical zones M5 and M6, (ASPECTS score of 7) and hyperdense vessel (d). Single-phase CTA, showing occlusion of the left supraclinoid ICA to carotid terminus as well as non-filling of left A1 and M1 segments (e). Axial reconstructed MIP images demonstrated overall poor collateralization to the majority of the left MCA territory (f–h).
7.3.3 Diagnosis
Left carotid terminus occlusion with poor leptomeningeal collateralization to left MCA territory.
7.3.4 Treatment
As the patient was on warfarin with INR of 2.7, IV-tPA was contraindicated and so not administered. The patient was therefore brought directly to the interventional suite for endovascular treatment. Vitamin K was administered prior to intervention.
Material Used
8-Fr short angiographic sheath.
8-Fr MERCI balloon guide catheter.
5-Fr H1 slip catheter.
0.035 angled Terumo guidewire wire.
Synchro 14 microguidewire.
Rebar 18 microcatheter.
Solitaire AB 6 × 30 mm.
Trevo Pro 4 × 20 mm stent retriever.
Technique
Intervention was performed with conscious sedation and local anesthetic. A single-wall right common femoral artery puncture was performed, and the 8-Fr short vascular access sheath was inserted. The 8-Fr balloon guide catheter was advanced to the left ICA over a 5-Fr H1 slip catheter with the aid of a Terumo guidewire.
Left internal carotid angiogram demonstrated persistent occlusion of the left carotid terminus, with abrupt cutoff of contrast in the supraclinoid ICA. There was no filling of left ACA or MCA territory from the carotid injection.
A Rebar 18 microcatheter was navigated through the occluded left carotid terminus and M1 segment with the aid of a Synchro 14 guidewire. A 6 × 30 mm Solitaire AB stent was deployed from the M1 segment of MCA to the left supraclinoid ICA (Fig. 7.5a). There was no evidence of antegrade flow through the deployed stent on left ICA control injection.
The stent was left in situ for 5 minutes and then retrieved with flow arrest and continuous aspiration. A large volume of thrombus was retrieved in the stent. Control angiography demonstrated recanalization of the carotid terminus, left A1, and proximal left MCA to the level of the distal M1 segment with some distal filling of temporal and anterior division of MCA branches (Fig. 7.5b, c).
Two further passes were performed in a similar fashion, from the M2 to M1 segment of MCA, first with the Solitaire 6 × 30 mm and secondly with a 4 × 20 mm Trevo Pro stent retriever. This resulted in complete recanalization of the left M1 segment of MCA and reperfusion of the distal MCA territory (Fig. 7.5d–f).
There was spasm in both the M1 segment and ICA. A total of 5 mg of intra-arterial (IA) verapamil was administered through the guide catheter with good effect. The patient was recanalized within 5 hours of symptom onset.
Fig. 7.5 Solitaire stent deployed from the M1 segment to the left supraclinoid ICA (a) with no antegrade flow. Control DSA (b,c) after continuous aspiration and stent retrieval shows recanalization of the carotid terminus, left A1, and proximal left MCA to the level of the distal M1 segment with some distal filling of temporal and anterior division of MCA branches. DSA (d–f) after two further passes from the M2 to M1 segment with the Solitaire 6 × 30 mm and secondly with a 4 × 20 mm Trevo Pro stent retriever showing complete recanalization of the left M1 segment of MCA and reperfusion of the distal MCA territory.
There was great difficulty in attempting to place closure device at the puncture site in view of extreme tortuosity of the iliac vessels and due to the patient’s body habitus, as there was a large amount of soft tissue between the skin surface and the sheath entry site into the femoral artery. A groin hematoma was also present. Hemostasis was achieved by manual compression.
7.4 Postprocedure Care/Outcome
The patient’s clinical condition was not significantly changed postprocedure. The following morning, NIHSS score was 26, which was a slight improvement from 28 on admission. CT at 24 hours postthrombectomy (Fig. 7.6a–d), however, demonstrated large left MCA infarct, with relative sparing of the left posterior temporal lobe (a site where there were collaterals present on initial CTA). There was hemorrhagic transformation of the infarct in the region of the left basal ganglia with associated mass effect, midline shift, effacement of the left lateral ventricle, and also intraventricular extension of blood. Neurosurgery was consulted; however, it was felt that prognosis even with decompressive craniectomy would be poor; therefore, further intervention was not performed. The poor prognosis was discussed with the patients’ family and palliative care service was consulted. The patient died on the eighth day of hospital admission.
7.4.1 Discussion
The greatest differentiating factor between these two similar cases with very different outcomes was most likely the degree of collateralization to the occluded left MCA territory. In the first case, there was excellent leptomeningeal collateralization with retrograde filling of MCA branches back to the M1 segment on CTA at presentation. In the second patient, who went on to infarct almost the entire territory, CTA performed at the same time from symptom onset showed poor collateralization particularly in the left frontal and parietal region, and there was already some loss of frontal gray-white matter differentiation at 2 hours from symptom onset.
Leptomeningeal collaterals, also referred to as leptomeningeal anastomoses or pial collaterals, are direct arteriole–arteriole connections, approximately 50 to 400 μm in caliber, joining terminal cortical branches of major cerebral arteries (i.e., anterior, middle, and posterior cerebrals) along the brain surface. Dormant under normal conditions (i.e., when blood flow from the major cerebral arteries is not impeded), they provide a route for retrograde filling of the territory distal to an occluded artery.
Carotid terminus occlusion has been shown to be an independent predictor for poor outcome in stroke, most likely because the occlusion of the carotid terminus cuts off the most important channel for collateral blood supply, the circle of Willis. Provided the patient has a patent anterior communicating artery, the ACA ipsilateral to the occlusion will fill antegradely from the contralateral side. There is, however, no possibility of antegrade supply to the MCA territory, and collateral supply, if it exists, must therefore be retrograde due to filling of leptomeningeal collaterals from the ipsilateral ACA or PCA territory.
Numerous studies, using several imaging modalities and grading methods, suggest that good leptomeningeal collaterals confer a benefit in stroke. Relationship has been shown between collateral flow and the “known” predictors of good outcome in stroke. For example, good leptomeningeal collateral supply has been shown to be associated with lower NIHSS score at presentation, higher baseline ASPECTS score, lower admission diffusion-weighted imaging (DWI) lesion volume, less infarct growth, smaller final infarct size, and higher follow-up CT ASPECTS score. Poor collateralization has been associated with higher incidence and larger size of hemorrhage following IA thrombolytic therapy.
7.4.2 Workup and Diagnosis
Patient History
Targeted patient history should be obtained with a view to determining time interval from onset of symptoms, and contraindications to IV thrombolysis.
Examination and Investigations
Findings on physical examination in patients presenting with acute stroke will depend on the territory involved, and collateral supply to the occluded territory. Stroke severity should be assessed with the NIHSS. Patients with good leptomeningeal collateral supply have been shown to have lower NIHSS score at presentation. For example, Miteff et al 1 demonstrated significant difference in median acute NIHSS between good and reduced collateral groups in patients presenting with carotid T and M1 occlusion (NIHSS: 16 vs. 18; p = 0.012). Left and right hemisphere strokes were equally distributed between the groups. Menon et al 2 demonstrated that in multivariable analysis a poor collateral score was associated with higher baseline NIHSS score (odds ratio [OR]: 1.1 per 1 point increase in NIHSS; p = 0.04).
In determining eligibility for IV thrombolysis in a patient on warfarin, point-of-care INR testing in the emergency department will allow rapid determination of INR. IV-tPA administration can be considered in patients on warfarin with INR below 1.7.
NCCT will rule out hemorrhage as a cause for presentation and should be scrutinized for evidence of early ischemic changes such as loss of gray-white matter differentiation and early sulcal effacement. This is particularly important in patients with poor leptomeningeal collateral status as infarction may already be present even at short intervals from symptom onset as highlighted in section “Companion Case.” Lima et al 3 demonstrated that patients with “equal” or “greater” collaterals had significantly higher baseline CT ASPECTS score than those with “less” collaterals (p = 0.02).
CTA is an important tool in the workup of the acute stroke patient. It is widely available, is easily accessible, and serves as a quick and highly accurate method for detecting occlusive thrombus in proximal cerebral arteries, and if performed from the level of the aortic arch it allows assessment of the extracranial vasculature. CTA has the added benefit of giving additional information regarding collateralization to the occluded vascular territory, and allows assessment of leptomeningeal collaterals. There are numerous proposed grading systems for leptomeningeal collaterals in the literature, as discussed in greater detail later in the “Imaging” section. Throughout numerous studies in both medically managed and endovascular cohorts, good collateral status has been shown in multivariate analysis to be an independent predictor for good functional outcome. Patients with poor collateral status are more likely to have a poor outcome following large artery occlusive stroke, despite treatment. Collateral status therefore does have a role to play in patient selection.
CT perfusion can be used as an adjunctive technique in patients presenting with acute stroke. Territory distal to the occlusion will typically show reduced time to peak (TTP), prolonged mean transit time (MTT), reduced relative cerebral blood flow (rCBF), and preserved or reduced relative cerebral blood volume (rCBV). A region of rCBV is taken to represent already infarcted tissue (i.e., “core infarct”). Regions of tissue with reduced TTP, rCBF, and prolonged MTT that do not have reduced rCBV are taken to represent tissue that is hypoperfused, and therefore at risk, but not yet infarcted, referred to as ischemic “penumbra.” There are, however, downsides to CT perfusion. In addition to the extra radiation, CT perfusion requires postprocessing which takes time, there is a lack of standardization of postprocessing tools across vendors, and there is a lack of robust evidence validating its use in reliably identifying penumbra. There remains disagreement in the literature about its utility in routine clinical practice to guide early treatment decisions.
As a modality, CT is usually more accessible than MRI in the acute setting, it is also a quicker study to perform, and it does not require the extensive patient safety screening of MRI. MRI, however, remains the initial hyperacute stroke imaging investigation of choice in some centers. The basic MRI stroke protocol can be relatively fast and usually includes an axial DWI, axial T2*/susceptibility-weighted imaging, and axial fluid-attenuated inversion recovery, as well as intracranial time-of-flight or gadolinium-enhanced MRA to demonstrate site of occlusion. While MRA may provide information about the status of circle of Willis collaterals, it does not possess enough spatial resolution to evaluate the more distal leptomeningeal bed. CTA is superior to MRA for depicting and evaluating leptomeningeal collaterals. 4 Significant negative correlation has been demonstrated between CTA collateral score and baseline DWI lesion volume 5 ; patients with higher and therefore better collateral scores have been shown to have lower lesion volume on DWI at baseline (p < 0.001).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
