You are here

Article Text

Article menu

Download PDFPDF

Ischemic Stroke
Original research
Reasons for failed endovascular recanalization attempts in stroke patients
  1. Hannes Leischner1,
  2. Fabian Flottmann1,
  3. Uta Hanning1,
  4. Gabriel Broocks1,
  5. Tobias Djamsched Faizy1,
  6. Milani Deb-Chatterji2,
  7. Martina Bernhardt1,
  8. Caspar Brekenfeld1,
  9. Jan-Hendrik Buhk1,
  10. Susanne Gellissen1,
  11. Götz Thomalla2,
  12. Christian Gerloff2,
  13. Jens Fiehler1
  1. 1 Department of Diagnostic and Interventional Neuroradiology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
  2. 2 Department of Neurology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
  1. Correspondence to Dr PhD Hannes Leischner, Department of Diagnostic and Interventional Neuroradiology, University Medical Center Hamburg-Eppendorf, Hamburg 20246, Germany; h.leischner{at}uke.de

Abstract

Purpose Mechanical thrombectomy (MT) is a highly effective therapy in patients with acute ischemic stroke due to large vessel occlusion (LVO). However, complete recanalization of the occluded vessel cannot be achieved in all patients, leading to poor clinical outcome. We analyzed the reasons for failed recanalization to help direct future improvements in therapy.

Methods 648 consecutive stroke patients with LVO and an MT attempt were retrospectively analyzed for none or minimal recanalization, assessed according to the Thrombolysis in Cerebral Infarction (TICI) score (0/1). Procedural parameters were evaluated in a standardized approach. Among other variables, number of retrieval attempts, devices, duration of the intervention, and rescue methods were analyzed.

Results TICI 0/1 was observed in 72/648 patients (11%). In these patients, the thrombus could not be reached in 21% (n=15/72), was reached but not passed in 21% (n=15/72), and was reached and passed in 58% (n=42/72). Only a minor degree of initial recanalization was achieved in 19% (n=8/42) of patients with a reached occlusion during the course of the intervention. Furthermore, a higher number of passes with a single retriever device led to significant prolongation of the intervention. Therefore, major reasons for failed endovascular recanalization were difficult anatomical access and hard or resistant occlusions that might reflect hard thrombi or pre-existing atherosclerotic stenosis. Procedural complications such as dissection or perforation played a minor role.

Conclusion In stroke patients with failed MT attempts, approximately 60% of occlusions can be passed. In such cases, rescue therapy might be considered to improve recanalization and clinical outcome. Further development of access devices might help in the remaining cases where the microcatheter could not be manipulated to or through the occlusion.

  • stroke
  • thrombectomy
  • angiography

https://doi.org/10.1136/neurintsurg-2018-014060

Statistics from Altmetric.com

    Request Permissions

    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.

    Introduction

    Mechanical thrombectomy (MT) is highly effective in recanalyzing large vessel occlusion (LVO) in stroke patients.1 The close association between high recanalization rates and good clinical outcome is well established. Despite positive functional outcome in randomized studies, successful reperfusion was achieved in only 59–88% of patients2–6 whereas the rate of poor recanalization results (Thrombolysis in Cerebral Infarction (TICI) Scale scores of 0 and 1), if reported, ranged between 8% and 18% in these studies.1–5

    Little is known about the reasons for failed recanalization attempts. Identifying causes and possible alternative treatment options is crucial to further improve recanalization methods and patient outcome. We report our experiences in patients with failed recanalization (TICI 0 and 1) attempts and analyzed both patient related and technical causes for unsuccessful recanalization to improve understanding of these conditions.

    Such information may direct future developments of endovascular techniques and devices. We hypothesized that the absolute majority of occlusions can be reached and crossed with a microcatheter but fail to remain open at the end of the procedure.

    Material and methods

    This retrospective analysis was conducted using patient data from a stroke database from one university hospital with 24 h/7 day neurological and neuroradiological attendance (between January 2010 and July 2017). In the Department of Neuroradiology, MT is performed by 3–5 experienced neuroradiologists. Consecutive patients with acute ischemic stroke in the anterior circulation with MT attempts and a post interventional TICI score of 0 or 1 were included in this analysis.

    Primary treatment rationale was a significant neurological deficit (National Institutes of Health Stroke Scale (NIHSS) score >4) and aphasia caused by LVO in the anterior circulation (verified by CT angiography or MRI angiography) following exclusion of intracranial hemorrhage, mass effect, or advanced early ischemic changes. There was no absolute treatment time window and no age limit as exclusion criteria for MT. Intravenous thrombolytic therapy was administered and dosed following national and international guidelines.

    Data acquisition and processing

    Data were acquired using a standardized data entry form and included baseline demographics and previous medical history. Time of symptom onset and stroke severity, as measured by the NIHSS, were recorded. Time points of the procedure (eg, groin puncture and end of the procedure) were captured based on acquisition time created by the angiography system as well as by the treatment charts.

    Location of the occlusion was determined on the pre-therapeutic angiographic runs. Medication and devices used during the intervention procedures, thrombectomy maneuvers count, and intraprocedural complications were determined from the treatment protocols.

    We pre-specified three groups of patients: group 1, in which the occlusion could not be reached; group 2, in which the occlusion could be reached but not passed with a microcatheter; and group 3, in which the occlusion could be reached and passed with a microcatheter (figure 1).

    Figure 1
    Figure 1

    Total number of patients with specific interventional characteristics. Numbers are depicted as total numbers as well as percentages of the whole patient group or subgroup. TICI, Thrombolysis in Cerebral Infarction Scale.

    Complications and final perfusion result (TICI score) were assessed based on DSA images. Because this analysis was designed to focus on technical and angiographic parameters, clinical mid term or long term follow-up data were not recorded.

    Ethics approval and statistical analysis

    The stroke database was approved by the local ethics committee. Due to the retrospective nature of the data collection and analysis, the need for written informed consent was waived.

    For categorical data, absolute and relative frequencies are given. Univariate distribution of metric variables is described by median (IQR).

    Two tailed t tests were used for testing the time between angiographic key steps for patients treated with 0–3 or >3 retrieval attempts. A two tailed level of significance with a P value ≤0.05 was considered to indicate a significant difference.

    Data are shown as mean (SD) or median (IQR), as appropriate. Analyses were performed with SPSS V.21 software package (IBM Corporation, Armonk, New York, USA).

    Interventional parameters

    Our standard technique includes access via a 8 F femoral sheath, an 8 F guiding catheter/balloon catheter for the carotid artery, and a 6 F guide catheter for vertebral arteries. Access through brachial, radial, or carotid punctures is not performed at our institution. MT stentrieval devices included the Solitaire (Solitaire FR revascularization device; Covidien, Mansfield, Masachusetts, USA), Trevo ProVue Stentriever (Stryker, Kalamazoo, Michigan, USA), ERIC retrieval device (Microvention, Tustin, California, USA), APERIO (Acandis, Pforzheim, Germany), Embotrap (Neuravi, Galway, Ireland), and Revive (FAST Self-Expanding Basket Thrombectomy Device, Micrus Endovascular, San Jose, California, USA). Intermediate catheters (ie, 5 F and 6 F Sofia, Microvention; Navien A+058 Intracranial Support Catheter, Medtronic, Minneapolis, Minnesota, USA) were employed at the discretion of the interventional neuroradiologist for distal thrombus aspiration. The decision to end the procedure or to perform another recanalization attempt was taken by the interventionalist.

    Results

    Patient characteristics

    Between January 2010 and July 2017, MT with a final recanalization result of TICI 0/1 was observed in 72 patients (figures 1 and 2). Median patient age was 77 years (IQR 66–84 years). Patients presented with a median initial NIHSS score of 15 (IQR 13–20). There was a slight female predominance in this patient cohort (women: n=39, (55%)). Baseline clinical characteristics, thrombus etiology, and pre-interventional imaging information are depicted in table 1. Thirty-eight of 72 (53%) patients had received intravenous thrombolysis.

    View this table:
    Table 1

    Demographics and clinical characteristics of patients with a recanalization result of Thrombolysis in Cerebral Infarction (TICI) 0+1 or TICI 2–3

    The majority of patients had an occlusion of the middle cerebral artery (n=37, 51%) while the remaining patients had an occlusion of the internal carotid artery (n=35, 49%). Tandem occlusions were present in 27% of patients. Stenosis of the extracranial carotid was found in 32% of patients.

    All patients received non-contrast CT prior to MT, with scoring of early ischemic changes (Alberta Stroke  Program CT  Early Score (ASPECTS) median (IQR) 7 (6–8)).

    The characteristics of patients with an unfavorable (TICI 0/1) and favorable (TICI 2–3) recanalization result are depicted in table 1.

    Procedural aspects

    Cases in which the occluded artery could not be reached made up 21% (n=15/72) of all patients. This was explained in 53% (n=8/15) by elongated cervical vessels while in 20% (3/15) the femoral puncture was not possible (figure 1). In the remaining 27% (n=4/15) of patients, not reaching the occlusion was due to stenosis of the extracranial arterial vessels.

    In 79% (n=57/72) of all patients the thrombus was reached. When the occlusion was reached but not passed (n=15/72, 21%), in 80% (n=12/15) of these cases it was not possible to pass the thrombus with the microwire or the microcatheter. In the remaining 20% (n=3/15), passage of the occlusion was possible with the microwire but not with the microcatheter (figure 1).

    Overall, passage of the occlusion with a microcatheter was achieved in 58% (n=42/72) of patients (figure 1). In the case of successful passage, the vessel remained occluded in 81% (n=34/42) of patients after the first recanalization attempt. Only 19% (n=8/42) showed a minor degree of initial recanalization that re-occluded during the further course of the procedure (see online supplementary figure 1).

    After the first recanalization attempt failed, the procedure was stopped in 26% (n=11/42) of patients in whom the occlusion was passed, whereas in 74% (n=31/42) more recanalization attempts were undertaken, either with the same retriever (31%, n=13/42) or a different one (36%, n=15/42). In 7% (n=3/42) of cases, local intra-arterial fibrinolytic therapy was given (figure 1). A higher number of passes with a single retriever device led to significant prolongation of the intervention (see online supplementary figure 2). For this evaluation, the MT procedures were divided into two groups; one group with 0–3 retrieval attempts and a second group with >3 attempts. In 11 of 42 (26%) cases with successful passage, the intervention was terminated after a first unsuccessful retrieval attempt due to prolonged time since symptom onset (n=6), subarachnoid hemorrhage (n=2), or clinical deterioration of the patient during the Intervention (n=3).

    Figure 2
    Figure 2

    Selection process of patients for the study. TICI, Thrombolysis in Cerebral Infarction Scale.

    For all patients with failed recanalization attempts, the median time from groin puncture to time of final angiographic run was 85 min (IQR 55–122). The median time from symptom onset to groin puncture was 275 min (IQR 200–320 min) in the TICI 0/1 group compared with 235 min (IQR 173–295) in the TICI 2-3 group. Symptom onset was known in 52% or 64% of patients (TICI 0/1 vs TICI 2–3.

    Subgroup analysis

    Seventy-two patients with a recanalization result of TICI 0/1 were grouped according to whether the occlusion could not be reached, could be reached but not passed, or could be passed. Clinical and demographic characteristics as well as procedural parameters and thrombus etiology were compared between the groups (table 2).

    View this table:
    Table 2

    Demographics and clinical characteristics in subgroups of Thrombolysis in Cerebral Infarction (TICI) 1+0 patients

    In the group in which the occlusion could not be reached (n=15), median patient age was 79 years (IQR 69–84 years), patients presented with a median NIHSS score of 14 (IQR 13–20), and median ASPECTS score was 7.6–8 There was a male predominance in this patient cohort (women: n=3, (20%)).

    Discussion

    To our knowledge, this is the first study investigating the technical and procedural aspects of the subgroup of approximately 10% of LVO stroke patients with failed recanalization attempts. In our cohort, the thrombus could be passed in 58% of patients, reached without passage of the occlusion in 21% and not reached at all in the remaining 21% of cases. Major reasons for failed recanalization were difficult anatomical access (eg, cervical artery stenosis or elongation) and hard or resistant occlusion reflecting hard clots or pre-existing intracranial atherosclerotic stenosis.

    Anatomical reasons were elongated cervical vessels and extracranial carotid stenosis or occlusions. In cases in which the intracranial occluded vessel was not reached, arterial tortuosity was responsible for failure in 55% of cases. Direct puncture of the carotid artery could have been an alternative approach but was not pursued due to individual chance/risk considerations.7–9 Consequently, tandem occlusions and stenosis of the extracranial carotid artery were more frequent in patients with poor recanalization.

    To continue or to terminate treatment is one of the most difficult questions in endovascular stroke therapy. The most frequently stated reason (by interventionalists) for terminating treatment was the extended time since symptom onset. During the period of this study (2010–2017) several randomized clinical trials were published,10 23changing endovascular stroke therapy in many ways.

    The randomized studies published in 2015 in particular, that provided strong evidence for the treatment effect of endovascular therapy, had some impact on the decision to terminate the intervention due to time delay reasons. Four out of 46 interventions were stopped for this reason since 2015, whereas during the years 2010–2015, 10 out of 26 interventions were discontinued for time reasons.

    In our subgroup analyses, apart from localization of the occlusion, there were no other differences in characteristics between the group of patients with an unfavorable recanalization result (TICI 0/1) and those with a favorable recanalization result (TICI 2–3). The same was true for the subgroups of patients with failed recanalization (table 2). Interestingly, we did not observe any substantial differences with regard to thrombus etiology between the two main groups (TICI 0/1 vs TICI 2–3) and subgroups of patients with unsuccessful recanalization (table 2). This is surprising as thrombus etiology influences the composition, rigidity, and friction of the thrombsus and has been demonstrated to impact on the chances of successful recanalization.11–14 24-25One explanation could be the small sample size of the subgroups of patients with a frustrane recanalization, and therefore this aspect should be analyzed in a larger study population.

    In cases refractory to mechanical as well as aspiration thrombectomy, there are only a few treatment options. One is local infusion of thrombolytics and/or glycoprotein IIb/IIIa inhibitors. Another potential approach is permanent intracranial stenting.15–17 Several recent studies suggest a benefit of stent assisted recanalization in patients who would otherwise be left non-recanalized.16–21 It is of interest that more than 50% of patients in this study could have been candidates for intracranial stenting. However, intracranial stent placement normally requires general anesthesia and an aggressive antiplatelet regimen.20 Increased mortality in elderly and polymorbid patients that are intubated21 and the fear of an increased number of intracranial hemorrhages might have been reasons to withhold rescue stenting in this patient group.

    Our study has several limitations, most notably based on its retrospective single center design, representing only a limited view in terms of experiences and techniques. In addition, patients were treated over a period of several years, in which the material used and the experience of the interventionalists with this material, as well as guideline recommendations based on the data from clinical trials, underwent significant changes. However, the rate of TICI 0/1 was comparable with previous studies that ranged from 8% to 18%.1–5 Although angiographic images and written reports allowed a reliable understanding of the reasons to stop the procedure by the interventionalist, our analysis gives little insight into the biological and technical reasons for failed recanalization, particularly when the occlusion was successfully passed. Individual re-assessment of the chances of success during the procedure is assumed to be influenced by a multitude of external and individual factors. Unfortunately, identification of these was beyond the scope of this study due to its retrospective design. As this is a question of clinical relevance, a prospective study would be best suited to identify the factors influencing the decision when to stop the endovascular treatment in case of unsuccessful recanalization attempts.

    Conclusion

    In stroke patients with failed intracranial recanalization attempts, approximately 60% of occlusions can be crossed. In such cases, rescue therapy (eg, improved stent retrievers or permanent intracranial stent placement) might be considered to improve recanalization and clinical outcome. Further improvement of access devices would help in the remaining patients where the microcatheter could not be manipulated to or through the occlusion.

    References

    1. ↵
      2. Goyal M ,
      3. Menon BK ,
      4. van Zwam WH , et al
      . Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet 2016;387:1723–31.doi:10.1016/S0140-6736(16)00163-X
      OpenUrlCrossRefPubMed
    2. ↵
      2. Berkhemer OA ,
      3. Fransen PS ,
      4. Beumer D , et al
      . A randomized trial of intraarterial treatment for acute ischemic stroke. N Engl J Med 2015;372:11–20.doi:10.1056/NEJMoa1411587
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      2. Campbell BC ,
      3. Mitchell PJ ,
      4. Kleinig TJ , et al
      . Endovascular therapy for ischemic stroke with perfusion-imaging selection. N Engl J Med 2015;372:1009–18.doi:10.1056/NEJMoa1414792
      OpenUrlCrossRefPubMed
    4. ↵
      2. Saver JL ,
      3. Goyal M ,
      4. Bonafe A , et al
      . Stent-retriever thrombectomy after intravenous t-PA vs. t-PA alone in stroke. N Engl J Med 2015;372:2285–95.doi:10.1056/NEJMoa1415061
      OpenUrlCrossRefPubMed
    5. ↵
      2. Goyal M ,
      3. Demchuk AM ,
      4. Menon BK , et al
      . Randomized assessment of rapid endovascular treatment of ischemic stroke. N Engl J Med 2015;372:1019–30.doi:10.1056/NEJMoa1414905
      OpenUrlCrossRefPubMed
    6. ↵
      2. Jovin TG ,
      3. Chamorro A ,
      4. Cobo E , et al
      . Thrombectomy within 8 hours after symptom onset in ischemic stroke. N Engl J Med 2015;372:2296–306.doi:10.1056/NEJMoa1503780
      OpenUrlCrossRefPubMed
    7. ↵
      2. Ribo M ,
      3. Flores A ,
      4. Rubiera M , et al
      . Difficult catheter access to the occluded vessel during endovascular treatment of acute ischemic stroke is associated with worse clinical outcome. J Neurointerv Surg 2013;5(Suppl 1):i70–i73.doi:10.1136/neurintsurg-2012-010438
      OpenUrlAbstract/FREE Full Text
    8. ↵
      2. Mokin M ,
      3. Snyder KV ,
      4. Levy EI , et al
      . Direct carotid artery puncture access for endovascular treatment of acute ischemic stroke: technical aspects, advantages, and limitations. J Neurointerv Surg 2015;7:108–13.doi:10.1136/neurintsurg-2013-011007
      OpenUrlAbstract/FREE Full Text
    9. ↵
      2. Nii K ,
      3. Kazekawa K ,
      4. Onizuka M , et al
      . Direct carotid puncture for the endovascular treatment of anterior circulation aneurysms. AJNR Am J Neuroradiol 2006;27:1502–4.
      OpenUrlAbstract/FREE Full Text
    10. ↵
      2. Broderick JP ,
      3. Palesch YY ,
      4. Demchuk AM , et al
      . Interventional Management of Stroke (IMS) III Investigators. Endovascular therapy after intravenous t-PA versus t-PA alone for stroke. N Engl J Med 2013;368:893–903.doi:10.1056/NEJMoa1214300
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      2. Yoo AJ ,
      3. Andersson T
      . Thrombectomy in acute ischemic stroke: challenges to procedural success. J Stroke 2017;19:121–30.doi:10.5853/jos.2017.00752
      OpenUrlCrossRefPubMed
    12. ↵
      2. Moftakhar P ,
      3. English JD ,
      4. Cooke DL , et al
      . Density of thrombus on admission CT predicts revascularization efficacy in large vessel occlusion acute ischemic stroke. Stroke 2013;44:243–5.doi:10.1161/STROKEAHA.112.674127
      OpenUrlAbstract/FREE Full Text
    13. ↵
      2. Yan S ,
      3. Liu K ,
      4. Tong L , et al
      . Different risk factors for poor outcome between patients with positive and negative susceptibility vessel sign. J Neurointerv Surg 2016;8:1001–5.doi:10.1136/neurintsurg-2015-011999
      OpenUrlAbstract/FREE Full Text
    14. ↵
      2. Almekhlafi MA ,
      3. Menon B ,
      4. Goyal M , et al
      . Role of biological markers in predicting findings of multimodal thrombus imaging: implications for aetiology and recan-alization. Stroke 2011;42:e599.
      OpenUrl
    15. ↵
      2. Levy EI ,
      3. Mehta R ,
      4. Gupta R , et al
      . Self-expanding stents for recanalization of acute cerebrovascular occlusions. AJNR Am J Neuroradiol 2007;28:816–22.
      OpenUrlAbstract/FREE Full Text
    16. ↵
      2. Brekenfeld C ,
      3. Schroth G ,
      4. Mattle HP , et al
      . Stent placement in acute cerebral artery occlusion: use of a self-expandable intracranial stent for acute stroke treatment. Stroke 2009;40:847–52.doi:10.1161/STROKEAHA.108.533810
      OpenUrlAbstract/FREE Full Text
    17. ↵
      2. Zaidat OO ,
      3. Wolfe T ,
      4. Hussain SI , et al
      . Interventional acute ischemic stroke therapy with intracranial self-expanding stent. Stroke 2008;39:2392–5.doi:10.1161/STROKEAHA.107.510966
      OpenUrlAbstract/FREE Full Text
    18. ↵
      2. Baek JH ,
      3. Kim BM ,
      4. Kim DJ , et al
      . Stenting as a rescue treatment after failure of mechanical thrombectomy for anterior circulation large artery occlusion. Stroke 2016;47:2360–3.doi:10.1161/STROKEAHA.116.014073
      OpenUrlAbstract/FREE Full Text
    19. ↵
      2. Baracchini C ,
      3. Farina F ,
      4. Soso M , et al
      . Stentriever thrombectomy failure: a challenge in stroke management. World Neurosurg 2017;103:57–64.doi:10.1016/j.wneu.2017.03.070
      OpenUrl
    20. ↵
      2. Zhu L ,
      3. Liebeskind DS ,
      4. Jahan R , et al
      . Thrombus branching and vessel curvature are important determinants of middle cerebral artery trunk recanalization with Merci thrombectomy devices. Stroke 2012;43:787–92.doi:10.1161/STROKEAHA.110.612986
      OpenUrlAbstract/FREE Full Text
    21. ↵
      2. Forbrig R ,
      3. Lockau H ,
      4. Flottmann F , et al
      . Intracranial rescue stent angioplasty after stent-retriever thrombectomy : multicenter experience. Clin Neuroradiol 2018.doi:10.1007/s00062-018-0690-4
      2. Hassan AE ,
      3. Chaudhry SA ,
      4. Zacharatos H , et al
      . Increased rate of aspiration pneumonia and poor discharge outcome among acute ischemic stroke patients following intubation for endovascular treatment. Neurocrit Care 2012;16:246–50.doi:10.1007/s12028-011-9638-0
      OpenUrlCrossRefPubMedWeb of Science
      2. Gunning GM ,
      3. McArdle K ,
      4. Mirza M , et al
      . Clot friction variation with fibrin content; implications for resistance to thrombectomy. J Neurointerv Surg 2018;10:34–8.doi:10.1136/neurintsurg-2016-012721
      OpenUrlAbstract/FREE Full Text
      2. Borggrefe J ,
      3. Kottlors J ,
      4. Mirza M , et al
      . Differentiation of clot composition using conventional and dual-energy computed tomography. Clin Neuroradiol 2017.doi:10.1007/s00062-017-0599-3

    Footnotes

    • Contributors HL, FF, UH, CB, and JF made substantial contributions to the conception and design of the work. Data acquisition was performed by HL, FF, CB, GB, MD-C and TF. HH, FF, and JF performed the data analysis. Interpretation of the data was done by HH, FF, J-HB, MB, CB, GT, CG, and JF. HL and JF drafted the manuscript and all of the other authors revised it critically for important intellectual content. All authors approved the final version to be published. They agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the manuscript are appropriately investigated and resolved.

    • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial, or not-for-profit sectors.

    • Competing interests JF served as a consultant for Acandis, Boehringer Ingelheim, Codman, Microvention, Sequent, and Stryker; speaker for Bayer Healthcare, Bracco, Covidien/ev3, Penumbra, Philips, and Sie­mens; and received grants from Bundesministeriums für Wirtschaft und Energie (BMWi), Bundesministe­rium für Bildung und Forschung (BMBF), Deutsche Forschungsgemeinschaft (DFG), European Union (EU), Covidien, Stryker (THRILL study), and Microvention (ERASER study) GT serves as a consultant or speaker for Acandis, Bayer Healthcare, Boehringer Ingelheim, BristolMyersSquibb/Pfizer, Covidien, and Glaxo Smith Kline; served as a lead investigator of the WAKE-UP study; was principal investigator of the THRILL study; and received grants from the Eu­ropean Union (grant No. 278276 and 634809) and Deutsche Forschungsgemeinschaft (SFB 936, Projekt C2). CG received fees as a consultant or lecture fees from Bayer Vital, Boehringer Ingelheim, EBS Technologies, Glaxo Smith Kline, Lundbeck, Pfizer, Sanofi Aventis, Silk Road Medical, and UCB. The other authors have no conflicts of interest.

    • Patient consent Not required.

    • Ethics approval The study was approved by the ethics committee of the Medical Association of the City of Hamburg.

    • Provenance and peer review Not commissioned; externally peer reviewed.