Clin Neuroradiol DOI 10.1007/s00062-014-0347-x
O r i g i n a l A rt i c l e
The Impact of Histological Clot Composition in Embolic Stroke T. Boeckh-Behrens · M. I. Schubert · A. Förschler · S. Prothmann · K. Kreiser · C. Zimmer · J. Riegger · J. Bauer · F. Neff · V. Kehl · J. Pelisek · L. Schirmer · M. Mehr · H. Poppert
Received: 14 May 2014 / Accepted: 8 September 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose Thrombus composition has been suggested to have a decisive impact on the outcome of patients treated by mechanical thrombectomy because of embolic stroke. The recent development of stent retrievers allows collection and, hence, histopathological analysis of fresh thrombus material. Against this background, the aim of this prospective study was to assess the impact of thrombus composition on mechanical recanalization, clinical outcome and stroke etiology. Methods Thirty-four patients suffering from acute ischemic stroke due to occlusion of the distal internal carotid artery/ carotid-T, anterior cerebral artery, or middle cerebral arteries were mechanically recanalized, and thrombus material was obtained. Histological thrombus composition was compared with imaging, clinical, and neurointerventional data.
Results The main findings were that a higher percentage of white blood cells (WBCs) in the thrombus was associated with (i) cardioembolic etiology, (ii) extended mechanical recanalization time, and (iii) less favorable recanalization (Thrombolysis in Cerebral Infarction score) and clinical outcome (National Institute of Health Stroke Scale). Conclusion Our results suggest that thrombi with a high WBC fraction are related to more organized thrombi of cardioembolic origin associated with less favorable recanalization and clinical outcome in acute ischemic anterior circulation stroke. WBC-mediated immunological and coagulatory processes may play a key role in thrombus formation and pathogenesis of stroke warranting further investigation. Keywords Histopathology · Mechanical recanalization · Stroke · Thrombus
Electronic supplementary material The online version of this article (doi: 10.1007/s00062-014-0347-x) contains supplementary material, which is available to authorized users. T. Boeckh-Behrens, MD () · M. I. Schubert, MD · A. Förschler, MD · S. Prothmann, MD · K. Kreiser, MD · C. Zimmer, MD · J. Bauer, MD Department of Neuroradiology, University Hospital Rechts der Isar, Technical University Munich, Ismaninger Str. 22, 81675 Munich, Germany e-mail: [email protected] J. Riegger, MD Department of Experimental Cardiology, German Heart Centre Munich, Munich, Germany F. Neff, MD Department of Pathology, German Research Centre for Environmental Health, Munich, Germany
V. Kehl, PhD Institute of Medical Statistics and Epidemiology, University Hospital Rechts der Isar, Technical University Munich, Munich, Germany J. Pelisek, PhD Clinic of Vascular and Endovascular Surgery, University Hospital Rechts der Isar, Technical University Munich, Munich, Germany L. Schirmer, MD · M. Mehr · H. Poppert, MD Department of Neurology, University Hospital Rechts der Isar, Technical University Munich, Munich, Germany
13
2
T. Boeckh-Behrens et al.
Introduction
Clinical and Imaging Variables
Irrespective of the underlying etiology in acute ischemic stroke, chemical or mechanical removal of the occluding thrombus has been explored for efficient treatment in the acute phase. The concept of a relationship between successful restoration of arterial blood flow and favorable clinical outcome of patients is supported by multiple publications [1]. In addition to the well-established therapeutic options of intravenous and/or intra-arterial thrombolysis with recombinant tissue plasminogen activator (rt-PA) for the treatment of acute ischemic stroke, mechanical recanalization seems to be an alternative or additional therapeutic strategy [1, 2]. The current end point of this development is the so-called stent retriever, with which recanalization rates of approximately 90 % appear to be achievable without an increase in complication rates [3, 4]. This approach allows fresh thrombus to be retrieved in vivo and histopathologically analyzed, overcoming the limitations of postmortem and animal studies. Thus far, it has been shown that the thrombus predominantly consists of three components: fibrin/platelet accumulations and red and white blood cells (WBCs) [5, 6]. Preclinical models of thromboembolic stroke showed erythrocyte-rich clots to be more susceptible to intravenous and intra-arterial thrombolysis [7, 8] with rt-PA than fibrin-dominant clots. In addition to thrombus composition, other factors such as location, age and origin of clot, and thrombus size have been reported to be decisive factors predicting the success of intravenous thrombolysis [9, 10]. Although there are some data demonstrating the influence of thrombus composition on the success rate of intravenous thrombolysis as described earlier in the text [7, 8, 11], there is a lack of knowledge about how thrombus composition may impact on the course of mechanical thrombectomy and clinical outcome of patients. Against this background, our retrospective study aimed primarily at assessing the impact of thrombus composition on the procedure and clinical outcome of mechanical recanalization in patients with thromboembolic stroke in the anterior circulation. Secondly, we aimed to explore whether differences in thrombus composition allowed determination of thrombus etiology, origin, and organization.
The available pre-interventional computed tomography (CT) scans were evaluated regarding the presence of a hyperdense artery sign (HAS). If quality and source format was adequate, an additional region-of-interest (ROI) analysis in the center of the HAS was performed. Clinically, two full neurological examinations of all patients, including the National Institute of Health Stroke Scale (NIHSS) score assessment by the neurologist on duty, i.e., directly prior to the neuroradiological intervention and on the day of discharge, were performed. Additionally, the clinical outcome was measured by the modified Rankin Scale (mRS) within 90 days after discharge. The most likely cause of the acute ischemic stroke was defined according to the international Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification [12], based on all diagnostic and clinical information available during the hospital stay, including cerebral CT in all cases, CT angiography and magnetic resonance imaging (MRI) in some cases, and transcranial and extracranial duplex sonography and long-term ECG recording in all cases.
Patients and Methods The methods were approved by the local ethics committee. All patients (n = 54) who were treated in our university hospital by mechanical recanalization with stent retrievers between October 2010 and July 2011 by reason of distal internal carotid artery (ICA)/carotid-T, middle cerebral artery (MCA), or anterior cerebral artery (ACA) occlusion were included in the study. Evaluable thrombus material could be collected during the mechanical recanalization procedure in 34 patients.
13
Procedure of Mechanical Thrombectomy Allowing for individual patients’ needs, prerequisites, and circumstances, mechanical recanalization was performed according to the institutional standard procedure with distal access catheter without proximal balloon occlusion under general anesthesia. Depending on the individual situation, one of the following stent retrievers was used: Solitaire 4-20, Solitaire 6-30 (ev3 Endovascular Inc., Plymouth, MN, USA), Trevo, Trevo pro 4 (Concentric Medical Inc., Mountain View, CA, USA), or Penumbra Pulse (Penumbra Inc., Alameda, CA, USA). Thrombus material retrieved was collected for further histopathological analysis. The following parameters regarding the mechanical recanalization were assessed: time between initial symptoms and start of the angiographic procedure (time to treat), time between the first angiographic series and final angiographic assessment after mechanical recanalization (time to recanalization), time between initial symptoms and final assessment of recanalization result (total time), selected stent retriever, and the recanalization result based on the adjusted Thrombolysis in Cerebral Infarction (TICI) score [13]. Processing and Histological Analysis of Thrombus Material On retrieval, thrombus material was immediately fixed in phosphate-buffered 3.5 % formalin, transferred to 70 % ethanol within 48 h, and then embedded in paraffin. The formalin-fixed and paraffin-embedded thrombus material
The Impact of Histological Clot Composition in Embolic Stroke
was cut into 2-μm slices using a Microm HM 335 E microtome (Microm International GmbH, Walldorf, Germany), followed by hematoxylin–eosin and Elastica van Gieson staining of slices. The slides with the stained specimens were scanned at high resolution (× 400) with a Hamamatsu Nano-Zoomer 2.0-RS scanner (Hamamatsu Photonics K.K., Hamamatsu City, Japan) and digitally stored. Histological analysis of thrombi was performed by a senior neuropathologist blind to clinical and interventional data. Besides organization and degree of the main clot components (fibrin/platelets, red blood cells (RBCs), and WBCs), histological assessment included the presence of calcifications, lipid/cholesterol accumulations, endothelialized surfaces, or abnormal cell structures. Additionally, the relative quantitative fraction of the different thrombus components was evaluated using semi-automated color-based segmentation (Adobe Photoshop CS4 Extended, Version 11.0, Adobe Systems, San Jose, CA, USA) of the scanned slides of the complete retrieved thrombus material (Supplementary Figs. 1, 2A–C). In the presence of multiple fragments, all fragments were included in the relative quantitative fraction analysis. Statistical Analysis Two-sided Spearman’s rank correlation coefficient was used to assess significant correlations between the thrombusspecific parameters and clinical and interventional parameters. In addition, between-group differences in thrombus and clinical and interventional parameters were analyzed for statistical significance using the nonparametric Kruskal–Wallis-test. Statistical analyses were performed using SPSS software (IBM SPSS Statistics, Version 20). P-values 60 % of fibrin) 12 RBC (> 60 % of RBC) Mixed 38
percentage of leukocytes and the mRS scores up to 90 days (r = 0.358, P = 0.057), which was statistically not significant. There was a significant positive correlation between the percentage of leukocytes in the thrombi and the recanalization time (r = 0.370, P = 0.037), and a positive correlation with the number of recanalization maneuvers (r = 0.309, P = 0.086), although statistically not significant. Table 3 summarizes the results of the correlation analyses.
Group Comparisons After mechanical recanalization, 71 % of our patients (n = 24) achieved a TICI score of 3, and the remaining 10 patients (29 %), a score of 2b. No TICI score less than 2b emerged from our 34 patients. Patients achieving a TICI score of 3 showed a significantly lower proportion of WBCs in their thrombi than patients ending up with a TICI score 2b (Fig. 3a). In a search for group differences in clot composition in relation to stroke etiology, there was a significantly higher proportion of WBCs in the thrombi of group TOAST 2 (cardioembolic) than in group TOAST 1 (arterioembolic/ atherosclerotic; P = 0.034; Kruskal–Wallis test, two-sided) or TOAST 5 (other undetermined cause; P = 0.038). A similar trend was found in the group TOAST 2 compared with the group TOAST 4 (other determined cause; P = 0.054; Fig. 3b). There was no significant difference in clot composition between the groups “HAS,” “no HAS,” and “no CT avail-
13
6
T. Boeckh-Behrens et al.
Table 3 Results from correlation analysis of the three thrombus components with NIHSS score and recanalization time as well as number of recanalization maneuvers RBC (%) Fibrin (%) WBC (%) r-Value P-value r-Value P-value r-Value P-value 0.000** 0.047* RBC (%; n = 34) − 0.940 − 0.343 0.000** 0.139 0.432 Fibrin (%; n = 34) − 0.940 0.047* 0.139 0.432 WBC (%; n = 34) − 0.343 0.492 0.103 0.567 0.130 0.470 NIHSS pre (n = 32) − 0.124 0.130 0.237 0.185 0.394 0.023* NIHSS post (n = 31) − 0.269 0.125 0.494 0.358 0.320 NIHSS pre–post (n = 31) − 0.168 − 0.182 0.270 0.180 0.349 0.358 0.057 mRS up to 90 days (n = 29) − 0.212 0.420 0.083 0.653 0.370 0.037* Recanalization time (n =32) − 0.148 0.934 0.749 0.309 0.086 Number of recanalization maneuvers (n = 32) − 0.015 − 0.059 0.350 0.164 0.378 0.196 0.292 Total time (n = 31) − 0.174 0.255 0.253 0.727 0.890 Thrombus density in HU (n = 22) − 0.79 − 0.31 NIHSS National Institute of Health Stroke Scale, NIHSS pre NIHSS before mechanical thrombectomy, NIHSS post NIHSS after mechanical thrombectomy (taken at discharge), mRS modified Rankin Scale, RBC red blood cells, Total time time between initial symptoms and final assessment of recanalization result, WBC white blood cells, HU Hounsfield units *P
O r i g i n a l A rt i c l e
The Impact of Histological Clot Composition in Embolic Stroke T. Boeckh-Behrens · M. I. Schubert · A. Förschler · S. Prothmann · K. Kreiser · C. Zimmer · J. Riegger · J. Bauer · F. Neff · V. Kehl · J. Pelisek · L. Schirmer · M. Mehr · H. Poppert
Received: 14 May 2014 / Accepted: 8 September 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose Thrombus composition has been suggested to have a decisive impact on the outcome of patients treated by mechanical thrombectomy because of embolic stroke. The recent development of stent retrievers allows collection and, hence, histopathological analysis of fresh thrombus material. Against this background, the aim of this prospective study was to assess the impact of thrombus composition on mechanical recanalization, clinical outcome and stroke etiology. Methods Thirty-four patients suffering from acute ischemic stroke due to occlusion of the distal internal carotid artery/ carotid-T, anterior cerebral artery, or middle cerebral arteries were mechanically recanalized, and thrombus material was obtained. Histological thrombus composition was compared with imaging, clinical, and neurointerventional data.
Results The main findings were that a higher percentage of white blood cells (WBCs) in the thrombus was associated with (i) cardioembolic etiology, (ii) extended mechanical recanalization time, and (iii) less favorable recanalization (Thrombolysis in Cerebral Infarction score) and clinical outcome (National Institute of Health Stroke Scale). Conclusion Our results suggest that thrombi with a high WBC fraction are related to more organized thrombi of cardioembolic origin associated with less favorable recanalization and clinical outcome in acute ischemic anterior circulation stroke. WBC-mediated immunological and coagulatory processes may play a key role in thrombus formation and pathogenesis of stroke warranting further investigation. Keywords Histopathology · Mechanical recanalization · Stroke · Thrombus
Electronic supplementary material The online version of this article (doi: 10.1007/s00062-014-0347-x) contains supplementary material, which is available to authorized users. T. Boeckh-Behrens, MD () · M. I. Schubert, MD · A. Förschler, MD · S. Prothmann, MD · K. Kreiser, MD · C. Zimmer, MD · J. Bauer, MD Department of Neuroradiology, University Hospital Rechts der Isar, Technical University Munich, Ismaninger Str. 22, 81675 Munich, Germany e-mail: [email protected] J. Riegger, MD Department of Experimental Cardiology, German Heart Centre Munich, Munich, Germany F. Neff, MD Department of Pathology, German Research Centre for Environmental Health, Munich, Germany
V. Kehl, PhD Institute of Medical Statistics and Epidemiology, University Hospital Rechts der Isar, Technical University Munich, Munich, Germany J. Pelisek, PhD Clinic of Vascular and Endovascular Surgery, University Hospital Rechts der Isar, Technical University Munich, Munich, Germany L. Schirmer, MD · M. Mehr · H. Poppert, MD Department of Neurology, University Hospital Rechts der Isar, Technical University Munich, Munich, Germany
13
2
T. Boeckh-Behrens et al.
Introduction
Clinical and Imaging Variables
Irrespective of the underlying etiology in acute ischemic stroke, chemical or mechanical removal of the occluding thrombus has been explored for efficient treatment in the acute phase. The concept of a relationship between successful restoration of arterial blood flow and favorable clinical outcome of patients is supported by multiple publications [1]. In addition to the well-established therapeutic options of intravenous and/or intra-arterial thrombolysis with recombinant tissue plasminogen activator (rt-PA) for the treatment of acute ischemic stroke, mechanical recanalization seems to be an alternative or additional therapeutic strategy [1, 2]. The current end point of this development is the so-called stent retriever, with which recanalization rates of approximately 90 % appear to be achievable without an increase in complication rates [3, 4]. This approach allows fresh thrombus to be retrieved in vivo and histopathologically analyzed, overcoming the limitations of postmortem and animal studies. Thus far, it has been shown that the thrombus predominantly consists of three components: fibrin/platelet accumulations and red and white blood cells (WBCs) [5, 6]. Preclinical models of thromboembolic stroke showed erythrocyte-rich clots to be more susceptible to intravenous and intra-arterial thrombolysis [7, 8] with rt-PA than fibrin-dominant clots. In addition to thrombus composition, other factors such as location, age and origin of clot, and thrombus size have been reported to be decisive factors predicting the success of intravenous thrombolysis [9, 10]. Although there are some data demonstrating the influence of thrombus composition on the success rate of intravenous thrombolysis as described earlier in the text [7, 8, 11], there is a lack of knowledge about how thrombus composition may impact on the course of mechanical thrombectomy and clinical outcome of patients. Against this background, our retrospective study aimed primarily at assessing the impact of thrombus composition on the procedure and clinical outcome of mechanical recanalization in patients with thromboembolic stroke in the anterior circulation. Secondly, we aimed to explore whether differences in thrombus composition allowed determination of thrombus etiology, origin, and organization.
The available pre-interventional computed tomography (CT) scans were evaluated regarding the presence of a hyperdense artery sign (HAS). If quality and source format was adequate, an additional region-of-interest (ROI) analysis in the center of the HAS was performed. Clinically, two full neurological examinations of all patients, including the National Institute of Health Stroke Scale (NIHSS) score assessment by the neurologist on duty, i.e., directly prior to the neuroradiological intervention and on the day of discharge, were performed. Additionally, the clinical outcome was measured by the modified Rankin Scale (mRS) within 90 days after discharge. The most likely cause of the acute ischemic stroke was defined according to the international Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification [12], based on all diagnostic and clinical information available during the hospital stay, including cerebral CT in all cases, CT angiography and magnetic resonance imaging (MRI) in some cases, and transcranial and extracranial duplex sonography and long-term ECG recording in all cases.
Patients and Methods The methods were approved by the local ethics committee. All patients (n = 54) who were treated in our university hospital by mechanical recanalization with stent retrievers between October 2010 and July 2011 by reason of distal internal carotid artery (ICA)/carotid-T, middle cerebral artery (MCA), or anterior cerebral artery (ACA) occlusion were included in the study. Evaluable thrombus material could be collected during the mechanical recanalization procedure in 34 patients.
13
Procedure of Mechanical Thrombectomy Allowing for individual patients’ needs, prerequisites, and circumstances, mechanical recanalization was performed according to the institutional standard procedure with distal access catheter without proximal balloon occlusion under general anesthesia. Depending on the individual situation, one of the following stent retrievers was used: Solitaire 4-20, Solitaire 6-30 (ev3 Endovascular Inc., Plymouth, MN, USA), Trevo, Trevo pro 4 (Concentric Medical Inc., Mountain View, CA, USA), or Penumbra Pulse (Penumbra Inc., Alameda, CA, USA). Thrombus material retrieved was collected for further histopathological analysis. The following parameters regarding the mechanical recanalization were assessed: time between initial symptoms and start of the angiographic procedure (time to treat), time between the first angiographic series and final angiographic assessment after mechanical recanalization (time to recanalization), time between initial symptoms and final assessment of recanalization result (total time), selected stent retriever, and the recanalization result based on the adjusted Thrombolysis in Cerebral Infarction (TICI) score [13]. Processing and Histological Analysis of Thrombus Material On retrieval, thrombus material was immediately fixed in phosphate-buffered 3.5 % formalin, transferred to 70 % ethanol within 48 h, and then embedded in paraffin. The formalin-fixed and paraffin-embedded thrombus material
The Impact of Histological Clot Composition in Embolic Stroke
was cut into 2-μm slices using a Microm HM 335 E microtome (Microm International GmbH, Walldorf, Germany), followed by hematoxylin–eosin and Elastica van Gieson staining of slices. The slides with the stained specimens were scanned at high resolution (× 400) with a Hamamatsu Nano-Zoomer 2.0-RS scanner (Hamamatsu Photonics K.K., Hamamatsu City, Japan) and digitally stored. Histological analysis of thrombi was performed by a senior neuropathologist blind to clinical and interventional data. Besides organization and degree of the main clot components (fibrin/platelets, red blood cells (RBCs), and WBCs), histological assessment included the presence of calcifications, lipid/cholesterol accumulations, endothelialized surfaces, or abnormal cell structures. Additionally, the relative quantitative fraction of the different thrombus components was evaluated using semi-automated color-based segmentation (Adobe Photoshop CS4 Extended, Version 11.0, Adobe Systems, San Jose, CA, USA) of the scanned slides of the complete retrieved thrombus material (Supplementary Figs. 1, 2A–C). In the presence of multiple fragments, all fragments were included in the relative quantitative fraction analysis. Statistical Analysis Two-sided Spearman’s rank correlation coefficient was used to assess significant correlations between the thrombusspecific parameters and clinical and interventional parameters. In addition, between-group differences in thrombus and clinical and interventional parameters were analyzed for statistical significance using the nonparametric Kruskal–Wallis-test. Statistical analyses were performed using SPSS software (IBM SPSS Statistics, Version 20). P-values 60 % of fibrin) 12 RBC (> 60 % of RBC) Mixed 38
percentage of leukocytes and the mRS scores up to 90 days (r = 0.358, P = 0.057), which was statistically not significant. There was a significant positive correlation between the percentage of leukocytes in the thrombi and the recanalization time (r = 0.370, P = 0.037), and a positive correlation with the number of recanalization maneuvers (r = 0.309, P = 0.086), although statistically not significant. Table 3 summarizes the results of the correlation analyses.
Group Comparisons After mechanical recanalization, 71 % of our patients (n = 24) achieved a TICI score of 3, and the remaining 10 patients (29 %), a score of 2b. No TICI score less than 2b emerged from our 34 patients. Patients achieving a TICI score of 3 showed a significantly lower proportion of WBCs in their thrombi than patients ending up with a TICI score 2b (Fig. 3a). In a search for group differences in clot composition in relation to stroke etiology, there was a significantly higher proportion of WBCs in the thrombi of group TOAST 2 (cardioembolic) than in group TOAST 1 (arterioembolic/ atherosclerotic; P = 0.034; Kruskal–Wallis test, two-sided) or TOAST 5 (other undetermined cause; P = 0.038). A similar trend was found in the group TOAST 2 compared with the group TOAST 4 (other determined cause; P = 0.054; Fig. 3b). There was no significant difference in clot composition between the groups “HAS,” “no HAS,” and “no CT avail-
13
6
T. Boeckh-Behrens et al.
Table 3 Results from correlation analysis of the three thrombus components with NIHSS score and recanalization time as well as number of recanalization maneuvers RBC (%) Fibrin (%) WBC (%) r-Value P-value r-Value P-value r-Value P-value 0.000** 0.047* RBC (%; n = 34) − 0.940 − 0.343 0.000** 0.139 0.432 Fibrin (%; n = 34) − 0.940 0.047* 0.139 0.432 WBC (%; n = 34) − 0.343 0.492 0.103 0.567 0.130 0.470 NIHSS pre (n = 32) − 0.124 0.130 0.237 0.185 0.394 0.023* NIHSS post (n = 31) − 0.269 0.125 0.494 0.358 0.320 NIHSS pre–post (n = 31) − 0.168 − 0.182 0.270 0.180 0.349 0.358 0.057 mRS up to 90 days (n = 29) − 0.212 0.420 0.083 0.653 0.370 0.037* Recanalization time (n =32) − 0.148 0.934 0.749 0.309 0.086 Number of recanalization maneuvers (n = 32) − 0.015 − 0.059 0.350 0.164 0.378 0.196 0.292 Total time (n = 31) − 0.174 0.255 0.253 0.727 0.890 Thrombus density in HU (n = 22) − 0.79 − 0.31 NIHSS National Institute of Health Stroke Scale, NIHSS pre NIHSS before mechanical thrombectomy, NIHSS post NIHSS after mechanical thrombectomy (taken at discharge), mRS modified Rankin Scale, RBC red blood cells, Total time time between initial symptoms and final assessment of recanalization result, WBC white blood cells, HU Hounsfield units *P
