An increasing number of reports highlight polymer coating embolism as an iatrogenic complication of intravascular medical devices [1-3]. Autopsies, histologic evaluations of thrombectomy specimens, samples of captured debris, resected or biopsied tissues, are available methods used to study polymer emboli post investigative catherizations or interventional procedures. Reported data highlight the prevalence of this phenomenon and/or its clinicopathologic impacts, however, fall short of identifying higher-risk polymer emboli interventional devices. Consequently, an optimal approach for future investigations related to polymer coating embolism is required.

Mehta et. al investigate the histologic frequency of polymer emboli among patients who underwent endovascular thrombectomy for treatment of acute ischemic stroke due to large vessel occlusion by retrospectively evaluating thrombectomy specimens [2]. In this study, the reported frequency of polymer emboli includes the use of various devices and techniques among selected cases. However, literature highlights polymer coating embolism is device specific and dependent on coating integrity measured by particulates released [4]. Thus, the use of alternate devices with higher or lower particulate release for a given procedure may result in a large variation in incidence rates from reported results. Also, as mentioned by the authors, subsequent statistical correlations unless appropriately powered provide limited information on higher-risk or culprit devices. Consequently, in addition to providing limited value, use of histologic evaluations of thrombi (or autopsies) to exclusively evaluate incidence rates may be time-consuming and have a high resource burden. Notably, this may not be the intent of the authors of this study who elude to an article series.

Few studies related to polymer coating embolism have included controls over procedures and specific devices used. For example, histologic evaluations of captured debris within cerebral protection devices during a mitral valve repair procedure highlighted polymer coating type material in 12 of 14 cases [5]. The source of polymer emboli was speculated to be from the mitral valve repair catheter and/or adjunct devices. In another study, intracranial polymer emboli incidence rates were determined for a branded hydrophilic coated guide sheath used for carotid and iliac stenting procedures respectively in Yucatan miniswine [6]. Since the experimental stents and stent delivery systems lacked any coating, impacts of polymer emboli were isolated to the coated guide sheaths, guide catheters and guidewires used during the procedures. These studies with controls over procedures and devices may be leveraged to understand an incidence rate from a specific set of interventional devices repetitively used during a given procedure. Notably however, even this approach is unable to provide device specific information or identify higher-risk interventional devices as the origin of emboli are difficult to determine.

Device type, coating composition-device substrate material combinations, coating application processes, coating thickness and degree of coating coverage are variables that impact polymer coating integrity on a device [4,7]. Particulate generation testing – the determination of a count, shape and size of particulates released from a device when used in an in-vitro vessel model – is the industry standard for evaluating polymer coating integrity from an intravascular device [8]. An intuitive correlation exists between particulates released in-vitro and the polymer emboli incidence rates [4]. Thus, particulate generation testing may be used to compare particulates released and understand relative incidence rates among intravascular devices. In-vitro particulate testing is an effective and efficient approach to determining higher-risk particulate release devices and may assist in identifying culprit interventional devices.

Based on the aforementioned study methodologies and outcomes, the following approach may be outlined for future investigations related to polymer coating embolism: a) Autopsy based studies that typically lack device information or controls over prior procedures should be limited to determining clinicopathologic impacts of polymer emboli. For these studies, estimating a particulate burden to correlate impacts with the quantity of polymer emboli for each area of assessment is relevant; b) Histologic evaluations of thrombi, captured debris, resected or biopsied tissues are also effective in determining localized disease processes. These methods may be used to determine incidence rates for a procedure if devices are consistently repeated for selected cases. For these studies (and postmortem investigations) combining incidence rates with particulate test data from devices will assist in categorizing embolic risk and determine higher risk devices; c) When clinical presentations preclude the consistent use of devices for a procedure, animal studies may be used to determine incidence rates. Extrapolating clinical impacts from healthy animals maybe acceptable, however, investigations that include relevant disease conditions are preferred; d) In-vitro particulate generation testing may be the optimal method to rank embolic risk and determine higher risk polymer emboli devices.

Inclusion and exclusion criteria for polymer emboli related investigations are essential for meaningful results. Patient history should be carefully evaluated as prior procedures may impact incidence rates. For this reason, patients with a history of multiple interventional procedures should be excluded from studies that attempt to determine polymer emboli frequencies. This patient subset may be included for studies determining potential impacts from polymer emboli for a worst-case scenario assessment. All investigations should include device specific information such as type, dimensions (e.g. length, diameter), brand, coating types (e.g. hydrophilic and/or hydrophobic) and coated dimensions. Other important parameters include patient baseline characteristics, procedural length, device in-dwell time (if available) and procedural outcomes.

Comparing particulate data among devices or reporting the magnitude of embolic burden may require assumptions to characterize total particulate volume. For the aforementioned studies, use of an efficient light obscuration methodology for particulate assessments that provides an equivalent circular diameter for particle areas may be used [9]. Combined with a uniform 1-micron height representing the typical lamellar nature of polymer particulates [3], a cylindrical volume calculation for each particle may be optimal. Summation of particle volumes may provide a total particulate burden per device or affected area of inspection.

The FDA continues to work with stakeholders to create tools which permit the standardization of particulate test methods and enable comparisons among devices [10]. Till these standards become available, investigations should include assumptions used to generate particulate data. In the future, an accumulation of procedural and device particulate data with associated incidence rates and clinicopathologic impacts may provide actionable input for regulators for setting device particulate limits. Given the sparse literature on this subject, more studies with controls over procedural parameters and devices used are required. Procedures with devices exposed to larger frictional forces (e.g. chronic total occlusions, aortic repair, or atherectomy), or aqueous environments for longer durations (e.g. percutaneous mechanical circulatory support) should be prioritized for future studies.

Acknowledgements, Funding Sources, Disclosures & Author Contributions
Acknowledgements: None. No persons other than the listed authors have made contributions to this manuscript.
Funding Sources: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Disclosures: None. Authors declare no current relationship with industry and no conflicts of interest.
Author Contributions: All authors have contributed to this manuscript.

References:
1. Chopra AM, Mehta M, Bismuth J, et al. Polymer coating embolism from intravascular medical devices - a clinical literature review. Cardiovasc Pathol 2017;30:45-54.
2. Mehta RI, Rai AT, Vos JA, Solis OE, Mehta RI. Intrathrombus polymer coating deposition: a pilot study of 91 patients undergoing endovascular therapy for acute large vessel stroke. Part I: Histologic frequency. J Neurointerv Surg. Epub ahead of print 18 May 2019. doi: 10.1136/neurintsurg-2018-014684.
3. Hickey TB, Honig A, Ostry AJ, et al. Iatrogenic embolization following cardiac intervention: postmortem analysis of 110 cases. Cardiovasc Pathol 2019;40:12-18.
4. Babcock DE, Hergenrother RW, Craig DA, Kolodgie FD, Virmani R. In Vivo Distribution of Particulate Matter from Coated Angioplasty Balloon Catheters. Biomaterials 2013;34(13):3196-205.
5. Frerker C, Schlüter M, Sanchez OD, et al. Cerebral Protection During MitraClip Implantation: Initial Experience at 2 Centers. JACC Cardiovasc Interv. 2016;9(2):171-9.
6. Stanley JR, Tzafriri AR, Regan K, et al. Particulates from Hydrophilic-Coated Guiding Sheaths Embolize to the Brain. EuroIntervention 2016;11(12):1435-41.
7. Work JW. Technical white paper: considerations for hydrophilic surface coatings on medical devices [internet]. Biocoat, Inc. Horsham, PA, USA. 2016. Available from www.biocoat.com. Accessed 22 Apr 2016.
8. Center for Devices and Radiological Health Recognized Consensus Standards. Recognition number 3-99: AAMI TIR42:2010 Evaluation of Particulates Associated with Vascular Medical Devices (cardiovascular) [Internet]. 2010. Available from: https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfStandards/detail.cf.... Accessed Jan. 13, 2019.
9. AAMI TIR42:2010. Evaluation of Particulates Associated with Vascular Medical Devices. Arlington, VA: Association for the Advancement of Medical Instrumentation, 2010.
10. ASTM WK60510. New Test Method for Simulated Use Testing of Neurointerventional Device in Tortuous Vasculature. West Conshohocken, PA: American Society for Testing and Materials International, 2018.

Recently, we have read with great interest the article by Benson et al. “Clot permeability and histopathology: is a clot’s perviousness on CT imaging correlated with its histologic composition?” [1].
It is pleasing, that research in the field of thrombus characterization by perviousness and its association to thrombus composition is emerging. Benson et al. report a higher clot perviousness for RBC rich clots in comparison to fibrin dominant thrombi [1]. These results stand in contrast to our previously published study [2], that shows an association between perviousness and fibrin rich clots. We furthermore validated those findings in a large collective by showing a relationship between perviousness and cardioembolic origin. Further research to this special topic is scarce. However, there is another experimental and therefore well controllable study on artificial clots, that showed a strong association of fibrin content and contrast agent uptake [3], similar as it has been shown for in vivo thrombi in our study [2].
Consequently, these contradictory results demand further explanations. In our opinion, the differing results might be caused by methodological differences, which we want to discuss.
First, thrombus localizations should be taken into account. Benson et al. used a collective of 57 thrombi with different thrombus locations (38 MCA, 6 ICA, 5 ICA/MCA, 3 basilar artery, 2 posterior cerebral artery, 2 ICA/MCA/ACA, 1 ICC/MCA). It is at least questionable, if histological composition can be compared between different circulations as it is known that there are different pathophysiological mechanisms (e.g. more underlying stenosis in the posterior circulation) as well as different flow conditions that would influence thrombus evolution and consequently composition [4].
Second, and more important, using an in-homogenous collective for perviousness assessment would stringently lead to technical difficulties in data acquisition: to make measurements comparable, direct contact from fresh flooding contrast agent in CTA is required, preventing the problem of a stationary blood column, which can be observed in occlusions of the ICA, for example. From own experience, identifying exact thrombus location is challenging for cases of ICA occlusions, that complicates the perviousness assessment. To circumvent these risks for falsified measurements, we used in our study a homogenous collective of 32 MCA occlusions. It would be interesting if Benson et al. would reproduce their results in the subgroup of 38 MCA occlusions.
Third, technical issues about perviousness assessment should be discussed as they differ between the groups. We used a co-registration process for native CT and CTA images to exactly identify the thrombus, although the thrombus is not visible in the native CT scan. Benson et al. did not apply a co-registration process between native CT and CTA images, and they excluded patients, “if the inciting clot was too small to be visualized”. Especially fibrin-rich clots are not obviously visible on native CT, and they can be included in the analysis by using a co-registration process to avoid a systematical selection bias.
Fourth, we excluded patients because of non-occluding thrombi when contrast agent passed the thrombus. We think, that perviousness cannot be assessed adequately for these cases, because perviousness would be measured artificially high. Consequently, these cases also present with tendentially better clinical outcome. It would be interesting, if such cases are included in the study of Benson et al. and how they present.
From the majority of histological studies, an association between fibrin-rich clots and cardioembolic origin is known [5, 6]. This fact makes the perviousness assessment interesting as it can predict stroke cause early on (for detailed discussion see [2]). Benson et al. report a cardioembolic origin as most common (59.6% of thrombi). They also report that 66.7% appeared pervious, however with higher RBC density. It would be interesting if their data would show a possible correlation between histological thrombus composition and etiology, and, secondly, a correlation between perviousness and etiology.
Benson et al. based their hypotheses on a positive correlation between pervious clots and good clinical outcome, that seems plausible as the dependent tissue behind an occlusion is better supplied with blood and nutrients when the clot is pervious [7]. RBC-rich clots show better clinical outcome, that is, among others, based on an easier removal in mechanical thrombectomy [8]. Due to the predominant subgroup of cardioembolic, presumably fibrin-rich clots in the anterior circulation, these clots dominate the analyses. Possibly, the assumed tendency between perviousness and good clinical outcome is based on this predominant subgroup. Consequently, it would not contradict the possible association between higher perviousness and fibrin-rich clots.
In summary, association between perviousness, clot composition, etiology and clinical outcome is not conclusively clarified yet and warrants further research, especially in a larger patient group in a multi-centric setting.

Literature
1. Benson JC, Fitzgerald ST, Kadirvel R, Johnson C, Dai D, Karen D et al. Clot permeability and histopathology: is a clot's perviousness on CT imaging correlated with its histologic composition? J Neurointerv Surg. 2019. doi:10.1136/neurintsurg-2019-014979.
2. Berndt M, Friedrich B, Maegerlein C, Moench S, Hedderich D, Lehm M et al. Thrombus Permeability in Admission Computed Tomographic Imaging Indicates Stroke Pathogenesis Based on Thrombus Histology. Stroke. 2018;49(11):2674-82. doi:10.1161/STROKEAHA.118.021873.
3. Borggrefe J, Kottlors J, Mirza M, Neuhaus VF, Abdullayev N, Maus V et al. Differentiation of Clot Composition Using Conventional and Dual-Energy Computed Tomography. Clin Neuroradiol. 2017. doi:10.1007/s00062-017-0599-3.
4. Boeckh-Behrens T, Pree D, Lummel N, Friedrich B, Maegerlein C, Kreiser K et al. Vertebral Artery Patency and Thrombectomy in Basilar Artery Occlusions. Stroke. 2019;50(2):389-95. doi:10.1161/STROKEAHA.118.022466.
5. Sporns PB, Hanning U, Schwindt W, Velasco A, Minnerup J, Zoubi T et al. Ischemic Stroke: What Does the Histological Composition Tell Us About the Origin of the Thrombus? Stroke. 2017;48(8):2206-10. doi:10.1161/STROKEAHA.117.016590.
6. Boeckh-Behrens T, Kleine JF, Zimmer C, Neff F, Scheipl F, Pelisek J et al. Thrombus Histology Suggests Cardioembolic Cause in Cryptogenic Stroke. Stroke. 2016;47(7):1864-71. doi:10.1161/STROKEAHA.116.013105.
7. Santos EM, Marquering HA, den Blanken MD, Berkhemer OA, Boers AM, Yoo AJ et al. Thrombus Permeability Is Associated With Improved Functional Outcome and Recanalization in Patients With Ischemic Stroke. Stroke. 2016;47(3):732-41. doi:10.1161/STROKEAHA.115.011187.
8. Hashimoto T, Hayakawa M, Funatsu N, Yamagami H, Satow T, Takahashi JC et al. Histopathologic Analysis of Retrieved Thrombi Associated With Successful Reperfusion After Acute Stroke Thrombectomy. Stroke. 2016;47(12):3035-7. doi:10.1161/STROKEAHA.116.015228.

We read with interest the article by Soize et al. “Can early neurological improvement after mechanical thrombectomy be used as a surrogate for final stroke outcome?”[1] Based on their results, the authors concluded that early neurological improvement (ENI) 24 hours after thrombectomy is a straightforward surrogate of long-term outcome. However, all patients in this study were treated with conscious sedation (CS), and not general anesthesia (GA). The residual effects of GA may mask ENI and limit its utility as a surrogate for long-term outcome.[2]

We performed a similar analysis of patients enrolled in a prospective single-center registry. The ability of ENI to predict 3-month functional independence was assessed by the area under the receiver operating characteristic curve (AUC) and compared using the independent-samples Hanley test. Multivariable linear regression assessing the relationship between anesthetic technique and ENI was also performed. The analysis received ethics approval.

291 patients were treated with thrombectomy, with 261 (89.7%) procedures performed with GA, and 30 (10.3%) with CS. All patients were de-sedated and extubated more than 12 hours before 24-hour National Institutes of Health Stroke Scale assessment. 174 (59.8%) patients achieved 3-month functional independence. Baseline and procedural characteristics did not differ between GA and CS patients (all P>0.05). ENI demonstrated better prognostic ability in CS (AUC 0.91, 95% confidence interval [CI], 0.80-1.00) than it did with GA treated patients (AUC 0.73, 95% CI, 0.67-0.80; P=0.008). Multivariable regression showed that GA was independently associated with attenuated ENI (P=0.03).

Our findings are in agreement with Soize et al, in that ENI does seem to predict long-term outcome in thrombectomy patients treated with CS, with comparable AUCs (0.91 and 0.93 respectively).[1] However, our results also suggest that ENI is worse at predicting long-term outcome following thrombectomy performed with GA. Furthermore, GA was independently associated with attenuated ENI, suggesting that GA might mask early neurologic recovery. These trends would appear to be in agreement with the SIESTA trial, which reported a greater likelihood of achieving 3-month functional independence in GA than CS patients, in the absence of significant differences in ENI.[3] Post-hoc analysis of SIESTA showed that propofol dose during thrombectomy was independently associated with reduced ENI.[2] The possible mechanisms for GA attenuating ENI may include the residual pharmacological effects of benzodiazepines, opioids, neuromuscular blockers, and intravenous or volatile anesthetic agents, or transient complications of endotracheal intubation such as ventilator-associated complications.[4]

[1] Soize S, Fabre G, Gawlitza M, et al. Can early neurological improvement after mechanical thrombectomy be used as a surrogate for final stroke outcome? J. Neurointerv. Surg. 2019;11(5):450-454.

[2] Schönenberger S, Uhlmann L, Ungerer M, et al. Association of Blood Pressure With Short- and Long-Term Functional Outcome After Stroke Thrombectomy: Post Hoc Analysis of the SIESTA Trial. Stroke. 2018;49:1451–1456.

[3] Schönenberger S, Uhlmann L, Hacke W, et al. Effect of Conscious Sedation vs General Anesthesia on Early Neurological Improvement Among Patients With Ischemic Stroke Undergoing Endovascular Thrombectomy: A Randomized Clinical Trial. JAMA. 2016;316:1986–1996.

[4] Sinclair RCF, Faleiro RJ. Delayed recovery of consciousness after anaesthesia. Continuing Education in Anaesthesia, Critical Care & Pain. 2006;6:114–118.

Congratulations to Annika Keuler et al¹ on their experience with the wireless microcatheter technique preventing vessel perforations in endovascular thrombectomy. Based on their results, the authors conclude that in most cases of mechanical recanalization, the clot can be passed more safely with a wireless microcatheter. In our daily work, we also find the wireless microcatheter technique seems to reduce subarachnoid hyperdensity resulting from vessel perforations. However it seems difficult to confirm this correlation; the details of which will be discussed as follows. After reading and analyzing the article carefully, we have some opinions about the study which we would like to communicate with the authors because the conclusions of the paper directly relate to our clinical experience.
In the article, two radiological manifestations are defined as vessel perforations——contrast extravasation during angiography and angiographically occult ipsilateral circumscribed subarachnoid contrast extravasation which is identified by post-interventional CT scans. As confirmed by previous studies2-3, we agree with the authors on using immediate post-interventional CT examination to identify the subarachnoid hyperdensity due to intraoperative contrast extravasation. Based on their results, post-thrombectomy subarachnoid hyperdensity was observed on CT scans in 22 patients, in 18 of whom, the clot was passed using a microwire, and in the other four, using a wireless microcatheter. The authors concluded that the complication rate for post-thrombectomy hyperdensity was significantly higher when a microwire was used to pass the clot. However, Omid Nikoubashman et al² report that post-interventional subarachnoid hyperdensities are associated with a long interval between clinical onset and recanalization, a long procedure time, and a high number of recanalization attempts. Perry P Ng et al ³ also mention that an increased number of stent retriever passes, distal device positioning, and presence of severe vasospasm were associated with post-thrombectomy subarachnoid hyperdensity. Additionaly, the interventional neuroradiologists used a microwire to pass the clot when first-pass microcatheter passage was not successful. There is confounding bias in this practice itself——It could be more difficult to pass the clot and need more stent retriever attempts in the microwire use group.
To sum up, hyperdensity on immediate post-thrombectomy CT scans, a manifestation of vessel perforation, is associated with many factors. The conclusion that the wireless technique can reduce post-thrombectomy hyperdensity would be more convincing if the authors ruled out the association between subarachnoid contrast extravasation and potential risk factors.

Yuan Ma, Pei-Cheng Li, Long Chen
Department of Interventional Radiology, The First Affiliated Hospital of Soochow University, Suzhou, China.

Correspondence to Dr. Long Chen, Department of Interventional Radiology, The First Affiliated Hospital of Soochow University, 188 Shizi Street,215006 Suzhou, China；lchen76@163.com

References:
1 Keulers A, Nikoubashman O, Mpotsaris A, et al. Preventing vessel perforations in endovascular thrombectomy: feasibility and safety of passing the clot with a microcatheter without microwire: the wireless microcatheter technique. J Neurointerv Surg 2018: 2018-14267.
2 Nikoubashman O, Reich A, Pjontek R, et al. Postinterventional subarachnoid haemorrhage after endovascular stroke treatment with stent retrievers. Neuroradiology 2014;56: 1087-1096.
3 Ng PP, Larson TC, Nichols CW, et al. Intraprocedural predictors of post-stent retriever thrombectomy subarachnoid hemorrhage in middle cerebral artery stroke. J Neurointerv Surg 2018;11: 127-132.