Background The effect of the five positive randomized controlled trials on thrombectomy practices and procedural volume has yet to be defined. Further, few studies have attempted to define modern thrombectomy practices in terms of selection criteria and devices used.
Methods A 21 question survey of Society of Neurointerventional Surgery (SNIS) physicians was administered using the SurveyMonkey website, addressing current practices as well as changes from before January 1, 2015 to the months after this date.
Results A total of 78 responses were obtained (approximately 10% of SNIS membership). Prior to January 2015, two-thirds of respondents reported performing 1–5 thrombectomies per month (67%), with 31% performing more than 5 per month. Following January 2015, 62% of respondents reported performing more than 5 thrombectomies per month; 45% of respondents reported a higher number of thrombectomies after trial publication. 73% and 80% of respondents indicated that inpatient consultations and hospital to hospital transfers for thrombectomy have increased, respectively. A plurality of respondents reported using A Direct Aspiration First Pass Technique (40%) as the first strategy for revascularization. Most commonly, neurointerventionalists reported using conscious sedation (56%) for anesthesia. 74% of respondents indicated being successful with their primary technique in at least 70% of cases.
Conclusions This survey of predominantly academic SNIS physicians indicates that inpatient consultations, hospital to hospital transfers, and thrombectomy procedural volumes have increased modestly since the publication of the five major stroke trials this year. In addition, many respondents indicated an increase in aggressiveness in pursuing thrombectomy based on selection criteria.
Statistics from Altmetric.com
In 2013, three major randomized trials comparing intra-arterial therapies with medical management for emergent large vessel occlusion (ELVO) demonstrated no benefit for intra-arterial therapies.1–3 However, significant limitations to study design were noted, such as the lack of preprocedural confirmation of ELVO in some of the studies and use of antiquated revascularization technologies.4 ,5 These limitations led experts to question the relevance of these trial results to modern clinical practice, as the methodology used was not concordant with practices at many high volume stroke centers. For instance, many experienced centers performed angiographic imaging on all patients, used standard of care doses of intravenous tissue plasminogen activator for appropriate patients, and used retrievable stent or aspiration technologies.
In the last year, five prospective, multicenter, randomized clinical trials have confirmed the superiority of endovascular thrombectomy over medical management for the treatment of ELVO.6–10 A summary of these data clearly demonstrates thrombectomy (plus intravenous tissue plasminogen activator when indicated) as the standard of care treatment for patients presenting with ELVO.11 ,12 This fact is reflected in the new American Heart Association/American Stroke Association guidelines, which state that all candidates meeting certain criteria for thrombectomy should undergo this procedure as a class I, level of evidence A recommendation.13 These trials demonstrated significant benefits of thrombectomy, largely because these trials addressed a number of limitations learned from prior trials, and more closely emulated modern clinical practices.14
The effect of these five positive trials on thrombectomy practices and procedural volume has yet to be defined. Further, few studies have attempted to define modern thrombectomy practices in terms of selection criteria and devices used. We sought to answer these questions by performing a national survey of neurointerventional physicians.
Institutional review board approval was obtained prior to study initiation. The Society of Neurointerventional Surgery (SNIS) is the major North American physician organization to which the majority of neurointerventional fellowship trained physicians practicing in the USA belong as members. SNIS has approximately 800 members. The SNIS member website (http://www.snisonline.org) has discussion forums with regular email alerts for new forum discussion topics. We designed a 21 question survey, administered using the SurveyMonkey website (http://www.surveymonkey.com), addressing current practices as well as changes from before January 1, 2015 to the months after this date. We posted an announcement on the SNIS discussion forum on November 10, 2015 containing information about the survey, such as reasons for performing the study, time needed to complete, and the risks and benefits of participation. The post contained the link to the survey. A full list of questions and answer choices is displayed in table 1. The survey link remained accessible for 2 weeks. No compensation was offered to participants. All responses were anonymous.
A total of 78 responses were obtained (approximately 10% of SNIS membership). The full list of questions and respondent answers is shown in table 1.
Responses by region
Table 2 shows responses to questions based on region of the USA. The one respondent who selected ‘other’ was excluded from the regional analysis.
Selection criteria for thrombectomy
The vast majority of respondents used CT angiography with or without perfusion imaging for patient selection (92%), with most using a combination of imaging characteristics to determine candidacy (55%). Most commonly, respondents indicated performing thrombectomy on patients only rarely when the National Institutes of Health Stroke Scale (NIHSS) score was <8 (60%); therefore, approximately 70% of respondents predominantly perform thrombectomy on patients with an NIHSS score of ≥8. Most commonly, respondents reported performing thrombectomy based more on having favorable imaging (41%) than based on time since onset.
Approximately half of respondents indicated selecting patients with less severe strokes more frequently (46%), those with advanced age more frequently (56%), and those with longer times from onset more frequently (40%) since the beginning of 2015. However, the vast majority reported that selection based on CT or perfusion imaging had not changed (83%). Overall, nearly half of respondents (49%) reported being slightly more aggressive when selecting patients for thrombectomy since the beginning of the year. This trend was relatively consistent across all regions of the USA (table 2).
Prior to January 2015, two-thirds of respondents reported performing 1–5 thrombectomies per month (67%), with 31% performing more than 5 per month. Following January 2015, 62% of respondents reported performing more than 5 thrombectomies per month. Forty-five per cent of respondents reported a higher number of thrombectomies, based on the categories provided, from before to after January 2015. The number of thrombectomies performed and number of other stroke centers in the region are shown in table 3. Seventy-three per cent and 80% of respondents indicated that inpatient consultations and hospital to hospital transfers for thrombectomy have increased, respectively.
A plurality of respondents reported using A Direct Aspiration first Pass Technique (ADAPT; 40%) as the first strategy for revascularization. This was also the most common technique in four of the five regions of the USA (table 2). Most commonly, neurointerventionalists reported using conscious sedation (56%) for anesthesia. Seventy-four per cent of respondents indicated being successful with their primary technique in at least 70% of cases. Eighty per cent of respondents reported obtaining a Thrombolysis in Cerebral Infarction score of 2B or better in more than 75% of their cases.
This is the first national survey to assess the effect of the five major randomized trials on thrombectomy consultations and procedural volume, as well as to investigate revascularization strategies used and patient selection preferences. This survey produced several interesting findings. Most notably, the majority of respondents indicated that they are seeing more inpatient consultations for thrombectomy (73%) and more hospital to hospital transfers for thrombectomy (79%), and nearly half (45%) of respondents indicated that they are performing more thrombectomy procedures based on an increase to a higher category. These trends were relatively consistent across regions of the country and regardless of the number of nearby stroke centers. While the number of respondents in this survey was fairly low (10% of SNIS members), it does suggest a trend of increasing consultations and thrombectomy procedures since the publication of MR CLEAN (Multicentre Randomized Clinical trial of Endovascular treatment for Acute ischemic stroke in the Netherlands) on January 1, 2015.
This survey also demonstrates that neurointerventionalists have become more aggressive in pursuing thrombectomy for appropriate patients. Nearly two-thirds of respondents indicated an increase in aggressiveness in pursuing thrombectomy compared with the previous year, with about half of respondents indicating they are now selecting patients of older age, with less severe strokes, and increasing time from symptom onset. Interestingly, interpretation of perfusion imaging appears, for the most part, unchanged. Therefore, it is likely that the increase in mechanical thrombectomy procedures is related to both increases in consultation and hospital to hospital transfers for procedures as well as less strict selection criteria by neurointerventionalists.
In response to the publication of the five major trials, the American Heart Association/American Stroke Association published updated thrombectomy guidelines in June 2015.13 The guidelines provide a class I, level of evidence A recommendation that “patients should receive endovascular therapy with a stent retriever if they meet all the following criteria”, including causative occlusion of the internal carotid artery or proximal middle cerebral artery, age of at least 18 years, an NIHSS score of at least 6, Alberta Stroke Program Early CT Score (ASPECTS) score of 6 or higher, and treatment initiated within 6 h of symptom onset, among others. Importantly, these guidelines are based on the inclusion criteria from the five trials, which predominantly enrolled patients with internal carotid artery or proximal middle cerebral artery occlusions within 6 h from onset, although two of the trials used longer time criteria.6 ,10 Interestingly, a high percentage of survey respondents indicated using more liberal selection criteria, including patients for whom intervention may have ‘unclear benefit’ per the guidelines.13 For instance, only 9% of respondents used 6 h as a definitive time threshold; in fact, over half reported using imaging criteria as the primary determinant and/or frequently treat patients outside of the 6 h window. Treating patients at times beyond 6 h is a class IIB, level of evidence C recommendation. Further, only 23% and 12% of respondents used ASPECTS scoring alone and perfusion imaging alone, respectively, for determining candidacy; most used a combination of CT and/or perfusion characteristics (55%). This survey therefore suggests neurointerventionalists working in the field are using other criteria than those suggested by the published guidelines, predominantly imaging based, to select candidates for thrombectomy. Perhaps this is unsurprising given the tremendous benefit thrombectomy offers patients who are suffering the most devastating form of stroke, and the potential that this benefit may extend beyond the population of patients in which it is proven.
Another important finding from this survey is related to the primary devices used for revascularization. The recent guidelines recommend the use of retrievable stents, as these were the predominant devices used for revascularization in the five major trials. Fiorella et al have recently argued against device specific guidelines due to the rapid pace at which technology is evolving in acute stroke care.15 This point is further emphasized by the findings from this survey, which indicate that 40% of respondents (the most common response) use ADAPT, a technique that is gaining support as a safe, fast, and cost effective means of obtaining revascularization.16 ,17 Furthermore, ADAPT was the most commonly used technique in four of the five regions. An additional 28% of respondents used retrievable stents with aspiration at the clot (‘Solumbra’), another technique which has been shown to be effective and combines aspiration and stent-triever technologies that is not formally recognized by the guidelines. Therefore, less than one-third of neurointerventionalists use retrievable stents as the primary revascularization strategy, as recommended by the guidelines that were published only 5 months prior to the survey being completed.
An additional interesting finding is the respondent's anesthetic preferences. In a survey of neurointerventionalists in 2010, most respondents preferentially placed patients under general anesthesia for thrombectomy.18 In this survey, approximately 80% of proceduralists now use conscious or minimal sedation as their anesthetic preference, a fairly dramatic change from the survey performed 5 years previously. This is likely due to the increasing literature suggesting a negative effect of general anesthesia on outcomes. For instance, a recent meta-analysis of nine studies comparing outcomes after conscious sedation versus general anesthesia confirmed the detrimental effect of general endotracheal anesthesia on thrombectomy outcomes.19
There are several important limitations to this study. First, the sample size was small, amounting to only about 10% of SNIS members. Second, there may be proceduralists performing thrombectomies that are not members of SNIS, and therefore these individuals could not be identified for participation. The survey was not designed to capture actual thrombectomy procedural volume, but to identify observed trends.
This survey of predominantly academic SNIS physicians indicates that inpatient consultations, hospital to hospital transfers, and thrombectomy procedural volumes have increased modestly since the publication of the five major New England Journal of Medicine stroke trials this year. In addition, many respondents indicated an increase in aggressiveness in pursuing thrombectomy based on selection criteria, with many physicians choosing to perform thrombectomy on older patients, less severe strokes, or longer times from symptom onset than they did last year. Further, most neurointerventionalists are using aspiration technologies for revascularization and are now using minimal or conscious sedation for procedures instead of general anesthesia. Although the survey sample size was small, these trends appear consistent across regions of the USA. Multicenter studies will be necessary to further define modern clinical practice.
Contributors Each author listed above should receive authorship credit based on the material contribution to this article, their revision of this article, and their final approval of this article for submission to this journal.
Competing interests None declared.
Ethics approval The study was approved by the institutional review board of the Medical University of South Carolina.
Provenance and peer review Not commissioned; externally peer reviewed.
Conflicts of interest Fargen: None. Arthur: consultant for Medtronic, Microvention, Penumbra, Sequent, Stryker, Codman and receives research support from Sequent and Siemens. Spiotta: consultant for Microvention, Penumbra, Pulsar Vascular, and Stryker. Lena: consultant for Penumbra. Chaudry: consultant for Covidien, Medina Medical, Microvention, Pulsar Vascular, Stryker, and Three Rivers Medical. Turner: consultant for Blockade Medical, Codman, Covidien, Microvention, Penumbra, and Pulsar Vascular. Turk: consultant for Codman, Covidien, Lazarus Effect, Medina Medical, Microvention, Penumbra, Pulsar Vascular, Siemens, and Three Rivers Medical.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.