ORIGINAL ARTICLE

Influence of Leptomeningeal Collateral Pattern on the Prognostic Value of Mismatch in Acute Anterior Circulation Stroke Alexander Spiessberger, MD, Christian Federau, MD, Dipl Phys ETH, Roman Guggenberger, MD, and Spyros Kollias, MD Objective: This study aimed to investigate whether and to what degree leptomeningeal collateral flow as detected on angiography influences the prognostic value of computed tomography perfusion-estimated mismatch in interventional treatment of acute anterior circulation stroke. Methods: Thirty-eight consecutive patients with acute anterior circulation stroke who received interventional neuroradiologic treatment were assigned one of 2 groups depending on the patient's degree of collateral flow (18 patients with poor collaterals, 20 patients with high degree collaterals) according to the American Society of Interventional and Therapeutic Neuroradiology/Society of Interventional Radiology grading system. In a multiregression model, we investigated a possible interaction between 2 independent variables mismatch ratio and degree of collateral flow using a “centered” variable approach. Results: The mismatch ratio per se showed a significant correlation with final clinical outcome (β coefficient, −0.79; P = 0.02); whereas, there was no interaction shown between mismatch degree of collateral flow (β coefficient, 0.54; P = 0.1). Conclusions: This study suggests that the predictive value of computed tomography perfusion-estimated mismatch is not influenced by the degree of leptomeningeal collateral flow. Key Words: stroke, mismatch, perfusion, CTP (J Comput Assist Tomogr 2015;39: 213–216)

n acute stroke, the clinical outcome1 is determined by many factors, such as time between onset of symptoms and treatment, type of treatment, localization and nature of occlusion. Another important determinant, receiving increasing attention recently, is the degree of collateralization of the vascular occlusion. Because of the possible serious adverse effects of both systemic and local thrombolysis, particularly cerebral hemorrhage, coupled with the fact that not every stroke patient benefits from revascularization, therapy indications and adequate selection of patients are of primary importance. Historically, the central paradigm in acute arterial occlusive stroke management used to be a rigid temporal treatment window, which in the last 2 decades, has been extended by newer concepts, such as core/penumbra mismatch, estimated with computed tomography perfusion (CTP) or diffusion weighted imaging/ perfusion weighted imaging. The concept of penumbra has been introduced in the late 1990s,2,3 and proposes that within a volume of ischemic brain tissue, there might be 2 compartments, namely, an irreversibly damaged infarcted core, and a surrounding halo of still salvageable tissue, named penumbra. This concept has been brought into clinical practice when the possibility of estimating

I

From the Institute of Neuroradiology, University Hospital Zurich, Zurich, Switzerland. Received for publication October 10, 2014; accepted November 21, 2014. Reprints: Alexander Spiessberger, MD, Klinik für Neurochirurgie, Kantonsspital Aarau, 5000 Aarau, Switzerland (e‐mail: [email protected]). The authors declare no conflict of interest. Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

the volumes of these 2 compartments through perfusion weighted imaging/diffusion weighted imaging and CTP studies4,5 emerged. In both methods, a number of parameters4 obtained through perfusion sequences are used to estimate the extent of penumbra and core. Although it is still under investigation, which parameters (and thresholds) most accurately reflect core and penumbra, especially in CPT imaging, clinical trials have shown the following: small core and large penumbra groups are more likely to benefit from revascularization6,7; the selection of patients for thrombolysis with a large penumbra/core ratio improves the benefit-risk ratio and is probably superior to patient selection based on time criteria alone.8 The potential of leptomeningeal vessels to perfuse the cerebral tissue distal to occlusions in a retrograde fashion has received increasing attention in the recent years. In the setting of interventional stroke therapy, it has been shown that a strong influence of the compensatory collateral flow on the final clinical outcome,9 on the rate of cerebral hemorrhage,10 on the rate of revascularization, and on the capacity of revascularization to improve clinical outcome.9,11 Moreover, Jung et al12 showed that the loss of penumbral tissue in cases of sustained vascular occlusion is faster in patients with poor collateral flow. In the present study, we focus our attention to patients with anterior territory infarcts. We aim to investigate whether the leptomeningeal collateral flow in these patients influences the CTPestimated mismatch and thus its prognostic value regarding clinical outcome after thrombectomy.

MATERIALS AND METHODS Patient Selection and Study Design This study was conducted according to the local ethics committee regulations; written consent was waived by the institutional review board. We retrospectively reviewed the results of medical imaging of 38 patients consecutively admitted to the emergency ward of the University Hospital of Zurich, Switzerland, between January 2012 and August 2013, with an acute cerebral arterial infarction of the middle cerebral artery, involving at least the first segment of the vessel (M1), as seen on admission CT. The standardized CT protocol consisted of conventional CT, CT angiography of the neck and intracranial arteries, CTP, and postcontrast imaging. All patients received intravenous recombinant tissue plasminogen activator, followed by interventional neuroradiologic treatment according to a standardized treatment protocol. Imaging, patient selection, and treatment were performed following the guidelines for the management of ischemic stroke and transient ischemic attack in the version of 2008.13 The following parameters were collected for each patient: anatomic site of occlusion, core lesion and penumbra size applying the thresholds proposed by Wintermark et al,14 modified Rankin scale (mRs) score, National Institute of Health Stroke Scale (NIHSS) score,

J Comput Assist Tomogr • Volume 39, Number 2, March/April 2015

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

www.jcat.org

213

J Comput Assist Tomogr • Volume 39, Number 2, March/April 2015

Spiessberger et al

plasma glucose values, vascular risk factors (arterial hypertension, diabetes, smoking, dyslipidemia), and radiation dose of the admission CT scan. On the revascularization angiography images, grades of collateral flow according to the American Society of Interventional and Therapeutic Neuroradiology/Society of Interventional Radiology (ASITN/SIR) grading system,15 time to recanalization, and “thrombolysis in myocardial infarction” (TIMI) grade were recorded. Within 24 hours after the onset of symptoms, a further native CT scan was performed to rule out intracranial hemorrhage (ICH). Follow-up imaging was performed in cases of worsening neurological symptoms. The cause of death was recorded in cases of patients who died within 90 days after the ischemic event. Patients surviving the stroke completed the physical rehabilitation programs and were reexamined clinically at day 90, when mRs score was reevaluated. On the basis of the ASITN/SIR grading system proposed by Higashida et al15 for leptomeningeal collateral flow, the patients were assigned one of 2 possible groups: group 1 consisted of patients with poor collateral grade and incomplete retrograde filling of the occluded territory (ASITN/SIR grades 0, 1, 2), whereas group 2 included patients with good collateral flow (ASITN/SIR grades 3 and 4) and therefore complete retrograde filling. Figures 1 and 2 show angiographic examples of either group.

CT Protocol All CT examinations performed upon admission were performed on a dual-source Siemens SOMATOM definition scanner, type CT2010Awith 2  128 detector rows (Siemens AG, Erlangen, Germany). All cranial scans were performed at 420 mA s and 120 kV. The initial precontrast scan covered the entire brain parenchyma and the bony skull from the first cervical vertebra to the vertex. Images were reconstructed in axial, coronal, and sagittal planes using a H70 kernel for bone and a H30 kernel for brain parenchyma. The section thickness was 4 mm for parenchymal and 2 mm for bone reconstructions. Perfusion scans were then obtained within 45 seconds, beginning 5 seconds after the start of an intravenous injection of 60-mL

FIGURE 2. Anterior-posterior projection of diagnostic angiography on a patient with left-sided M1 occlusion. The patient has highgrade collateral flow (group 2) with marked retrograde flow through branches of the ipsilateral anterior cerebral artery.

Ultravist 300 contrast agent (Bayer Healthcare AG, Leverkusen, Germany) at the rate of 5 mL/s. The scan was performed at a fixed table position, covering the area of the basal ganglia. After a 10-minute break for the contrast to washout from the vasculature, CT angiography was performed using the dualenergy mode. A triggered scan was performed (trigger point placed in the ascending aorta) during the injection of another 80 mL of Ultravist 300 at 4 mL/s covering the area from the aortic arch to the superior sagittal sinus. Using a H10 kernel axial, coronal and sagittal reconstructions were generated with a section width of 2 mm. Furthermore, 3-dimensional maximum-intensity projections and volume-rendered images were generated after applying the automated bone-removal mode. Finally, a postcontrast scan of the brain parenchyma was obtained 20 seconds after ending the acquisition of the angiography scan.

CTP—Postprocessing

FIGURE 1. Anterior-posterior projection of diagnostic angiography on a patient with right-sided M1 occlusion. The patient has poor grade collateral flow (group 1) with only poor retrograde flow through branches of the ipsilateral anterior cerebral artery.

214

www.jcat.org

Postprocessing was performed on a desktop computer using the software Olea Sphere V2.0 by Olea Medical (La Ciotat, France). The neuroradiologist performed the following tasks: visual inspection of the automatically detected arterial and venous reference vessel (anterior cerebral artery contralateral to the ischemic hemisphere and transverse sinus), definition of the midline, and definition of the affected hemisphere. The program uses an oscillating deconvolution approach to calculate parametric maps of the following parameters: mean transit time (MTT), time to peak (TTP), cerebral blood flow (CBF), cerebral blood volume (CBV), and maximal time (Tmax). There are several guidelines published4 on how to best define core and penumbra in CTP. The ones proposed by Wintermark et al14 (core, 145% MTT)2,3 are among the most accurate, and therefore applied in the present study. After entering the thresholds, the program calculates and visually displays both volumes on separate maps (Fig. 3). © 2015 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

J Comput Assist Tomogr • Volume 39, Number 2, March/April 2015

Leptomeningeal Collateral Pattern

FIGURE 3. CTP-estimated maps of MTT and CBV (upper row right and left, respectively) in a patient with distal M1 occlusion on the right side. After applying the thresholds proposed by Wintermark et al, a map outlining core and penumbra area (bottom line; core lesion in red and penumbra lesion in blue) is generated.

Neuroradiologic Intervention The patient was transferred to the interventional suit and a 4vessel diagnostic angiography was performed via a transfemoral route. The site of occlusion was verified and an assessment of leptomeningeal collateral flow was performed according to ASITN/ SIR grading system15 as described previously. Intra-arterial thrombolysis and thrombectomy were performed by an experienced interventional neuroradiologist.

approach, both variables were centered to create a centered product of both terms. Finally, a multiregression analysis was performed using the following parameters: mRs(90) as dependent variable; centered mismatch, centered collateral grade, and centered mismatch  collateral grade as independent variables. The result of this analysis is, that there was not shown to be a significant interaction between mismatch ratio and grade of collateral flow was shown (β coefficient, 0.54; P = 0.1). It was confirmed,

Statistical Analysis Statistical analysis was performed using PSPP version 0.8.3 by GNU project (www.hnu.org). Analysis of variance and multiregression analysis were established with mRs(90) being the dependent and mismatch ratio and collateral grade the independent variables. To check for possible interaction between the 2 independent variables, the approach described by Aiken et al16 and McClelland and Judd17 using centered variables was used. The following basic demographic and clinical data of our cohort are given in Table 1: sex, age, number of vascular comorbidities (arterial hypertension, nicotine, diabetes, dyslipidemia, coronary artery disease), glucose level upon admission (mmol/L), anatomic side of vascular occlusion (left/right), NIHSS upon admission, TIMI grade, time to recanalization (in minutes), number of deaths, number of intracerebral hemorrhage, mRs 90 days after the event, CTP-estimated core lesions size (in milliliter) and penumbra size (in milliliter), and radiation dose of admission CT scan.

RESULTS To investigate a possible interaction between mismatch ratio and grade of collateral flow, a statistical approach described by Aiken et al16 and McClelland and Judd17 was used. According to this

TABLE 1. Clinical and Demographic Characteristics of Patients Given Either as Absolute Numbers or as Median and SD in Brackets n Sex (f) Age, y No. vascular comorbidities Glucose level on admission, mmol/L Side of occlusion (left) NIHSS on admission TIMI grade Time to recanalization, min No. death No. ICHs mRs on day 90 Core volume, mL Penumbra volume, mL Radiation dose, mGy cm

© 2015 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

38 16 65.2 (13.5) 2 (1) 7.3 (2.1) 20 14 (4) 2 (1) 340 (135) 9 9 3.5 (2) 46.2 (30.2) 50.3 (25.2) 4325 (680)

www.jcat.org

215

J Comput Assist Tomogr • Volume 39, Number 2, March/April 2015

Spiessberger et al

however, that mismatch does show a significant correlation with mRs(90): β coefficient, −0.79; P = 0.02.

DISCUSSION There is an increasing amount of evidence that the pattern of leptomeningeal collateral vessels has an important impact on stroke dynamics and tissue fate.18 Several previous studies have shown the strong impact of collateral flow on the potential of therapeutic revascularization to improve clinical outcome,9 on the occurrence of adverse effects (namely, ICH),10 and on the rate of revascularization.9 Inefficient collaterals are an important factor for the rapid transformation of penumbra into core lesion. Patients with adequate collateral flow, most probably have a slow loss of penumbra tissue due to sufficient retrograde flow, which indicates that the penumbra is not at severe enough risk to be deemed prognostically relevant. Including the assessment of collateral flow pattern seems essential for the therapeutic decision making, especially in conjunction with the core/penumbra mismatch concept. The transformation of penumbra into core lesion is affected by the degree of leptomeningeal collateral flow, whereas a poor collateral flow results in faster transformation rates.12 Many factors influence18 the extent of compensatory collateral flow (eg, blood pressure, congenital vascular organization, blood glucose level, cardiac function) with a high degree of individual variations as a result. A common concept in stroke management is subdividing an ischemic zone into core and penumbra lesion to determine the socalled mismatch. It is thought that a patient with high mismatch value would benefit more from therapeutic revascularization than a patient with low mismatch.2,6,7,19,20 Without recanalization, it is thought that the mismatch value will slowly decrease over hours as the penumbra is transformed into core, resulting in a decreasing effectiveness of the therapeutic recanalization. With all these factors shown to be influenced by an individual's degree of collateral flow, we aimed to assess if the prognostic information of mismatch differs among patients with different degrees of collateral flow, warranting different interpretation depending on a patient's collateral grade. To do so, we set up a multiregression analysis and checked for intervariable interaction between mismatch and collateral grade. Statistical analysis failed to show a significant interaction between these 2 variables (β coefficient, 0.54; P = 0.1). The interpretation of this is that a certain value of mismatch does have the same prognostic information in patients with high-grade collateral flow as well as in patients with low-grade flow. Our study has some limitations. First, the number of 38 patients included in our study is relatively small to allow more confident statements on the prognostic value of collaterals and their relation with the evolution of penumbra. Second, the assessment of collateral flow itself is a matter of some degree of subjectivity. Third, the division of patients in 2 groups of collaterals is a simplification of the existing grading system, which proposes 5 possible grades (0–4) and was chosen to allow comparison at the level of our case number (38 patients). A larger cohort allowing analysis of patients in 5 groups could alter the result.

CONCLUSIONS The prognostic value of mismatch was not influenced by the degree of leptomeningeal collateral flow, meaning that a certain mismatch ratio holds the same information in patients with different collateral grades.

216

www.jcat.org

REFERENCES 1. Mullen MT, Pisapia JM, Tilwa S, et al. Systematic review of outcome after ischemic stroke due to anterior circulation occlusion treated with intravenous, intra-arterial, or combined intravenous+intra-arterial thrombolysis. Stroke. 2012;43:2350–2355. 2. Kumar G, Goyal MK, Sahota PK, et al. Penumbra, the basis of neuroimaging in acute stroke treatment: current evidence. J Neurol Sci. 2010;288:13–24. 3. Konstas AA, Wintermark M, Lev MH. CT perfusion imaging in acute stroke. Neuroimaging Clin N Am. 2010;21:215–238, ix. 4. Dani KA, Thomas RG, Chappell FM, et al. Computed tomography and magnetic resonance perfusion imaging in ischemic stroke: definitions and thresholds. Ann Neurol. 2011;70:384–401. 5. Dani KA, Thomas RG, Chappell FM, et al. Systematic review of perfusion imaging with computed tomography and magnetic resonance in acute ischemic stroke: heterogeneity of acquisition and postprocessing parameters: a translational medicine research collaboration multicentre acute stroke imaging study. Stroke. 2012;43:563–566. 6. Albers GW, Thijs VN, Wechsler L, et al. Magnetic resonance imaging profiles predict clinical response to early reperfusion: the diffusion and perfusion imaging evaluation for understanding stroke evolution (DEFUSE) study. Ann Neurol. 2006;60:508–517. 7. Hacke W, Albers G, Al-Rawi Y, et al. The Desmoteplase in Acute Ischemic Stroke Trial (DIAS): a phase II MRI-based 9-hour window acute stroke thrombolysis trial with intravenous desmoteplase. Stroke. 2005;36:66–73. 8. Kohrmann M, Juttler E, Fiebach JB, et al. MRI versus CT-based thrombolysis treatment within and beyond the 3 h time window after stroke onset: a cohort study. Lancet Neurol. 2006;5:661–667. 9. Bang OY, Saver JL, Kim SJ, et al. Collateral flow predicts response to endovascular therapy for acute ischemic stroke. Stroke. 2011;42:693–699. 10. Bang OY, Saver JL, Kim SJ, et al. Collateral flow averts hemorrhagic transformation after endovascular therapy for acute ischemic stroke. Stroke. 2011;42:2235–2239. 11. Bang OY, Saver JL, Buck BH, et al. Impact of collateral flow on tissue fate in acute ischaemic stroke. J Neurol Neurosurg Psychiatry. 2008;79:625–629. 12. Jung S, Gilgen M, Slotboom J, et al. Factors that determine penumbral tissue loss in acute ischaemic stroke. Brain. 2013;136:3554–3560. 13. European Stroke Organisation (ESO) Executive Committee. Guidelines for management of ischaemic stroke and transient ischaemic attack 2008. Cerebrovasc Dis. 2008;25:457–507. 14. Wintermark M, Flanders AE, Velthuis B, et al. Perfusion-CT assessment of infarct core and penumbra: receiver operating characteristic curve analysis in 130 patients suspected of acute hemispheric stroke. Stroke. 2006;37:979–985. 15. Higashida R, Furlan A, Roberts H, et al. Trial design and reporting standards for intraarterial cerebral thrombolysis for acute ischemic stroke. J Vasc Interv Radiol. 2003;14:S493–S494. 16. Aiken LS, Stephen Gl. Multiple Regression: Testing and Interpreting Interactions. S. 75–87. Newbury Park, CA: Sage; 1991. 17. McClelland GH, Judd CM. Statistical difficulties of detecting interactions and moderator effects. Psychol Bull. 1993;114:376–390. 18. Shuaib A, Butcher K, Mohammad AA, et al. Collateral blood vessels in acute ischaemic stroke: a potential therapeutic target. Lancet Neurol. 2011;10:909–921. 19. Furlan AJ, Eyding D, Albers GW, et al. Dose Escalation of Desmoteplase for Acute Ischemic Stroke (DEDAS): evidence of safety and efficacy 3 to 9 hours after stroke onset. Stroke. 2006;37:1227–1231. 20. Nambiar V, Sohn SI, Almekhlafi MA, et al. CTA collateral status and response to recanalization in patients with acute ischemic stroke. AJNR Am J Neuroradiol. 2014;35:884–890.

© 2015 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

Influence of leptomeningeal collateral pattern on the prognostic value of mismatch in acute anterior circulation stroke.

This study aimed to investigate whether and to what degree leptomeningeal collateral flow as detected on angiography influences the prognostic value o...
2MB Sizes 0 Downloads 12 Views