Article Text

Download PDFPDF

Original research
Physician training protocol within the WEB Intrasaccular Therapy (WEB-IT) study
  1. Adam Arthur1,2,
  2. Daniel Hoit1,2,
  3. Alexander Coon3,
  4. Josser E Delgado Almandoz4,
  5. Lucas Elijovich1,5,
  6. Saruhan Cekirge6,
  7. David Fiorella7
  8. WEB-IT Study Investigators
  1. 1 Semmes-Murphey Neurologic and Spine Institute, Memphis, Tennessee, USA
  2. 2 Department of Neurosurgery, University of Tennessee Health Science Center, Memphis, Tennessee, USA
  3. 3 Johns Hopkins University School of Medicine, The Johns Hopkins Hospital, Baltimore, Maryland, USA
  4. 4 Department of Neurointerventional Radiology, Abbott Northwestern Hospital, Minneapolis, Minnesota, USA
  5. 5 Department of Neurology, University of Tennessee Health Science Center, Memphis, Tennessee, USA
  6. 6 Marmara University Faculty of Medicine, Pendik Training and Research Hospital, Istanbul, Turkey
  7. 7 Department of Neurosurgery, State University of New York, Stony Brook, New York, USA
  1. Correspondence to Dr Adam Arthur, Semmes-Murphey Neurologic and Spine Institute, 6325 Humphreys Blvd, Memphis, TN 38120, USA; aarthur{at}


Introduction The WEB Intra-saccular Therapy (WEB-IT) trial is an investigational device exemption study to demonstrate the safety and effectiveness of the WEB device for the treatment of wide-neck bifurcation aneurysms. The neurovascular replicator (Vascular Simulations, Stony Brook, New York, USA) creates a physical environment that replicates patient-specific neurovascular anatomy and hemodynamic physiology, and allows devices to be implanted under fluoroscopic guidance.

Objective To report the results of a unique neurovascular replicator-based training program, which was incorporated into the WEB-IT study to optimize technical performance and patient safety.

Methods US investigators participated in a new training program that incorporated full surgical rehearsals on a neurovascular replicator. No roll-in cases were permitted within the trial. Custom replicas of patient-specific neurovascular anatomy were created for the initial cases treated at each center, as well as for cases expected to be challenging. On-site surgical rehearsals were performed before these procedures.

Results A total of 48 participating investigators at 25 US centers trained using the replicator. Sessions included centralized introductory training, on-site training, and patient-specific full surgical rehearsal. Fluoroscopy and procedure times in the WEB-IT study were not significantly different from those seen in two European trials where participating physicians had significant WEB procedure experience before study initiation.

Conclusions A new program of neurovascular-replicator-based physician training was employed within the WEB-IT study. This represents a new methodology for education and training that may be an effective means to optimize technical success and patient safety during the introduction of a new technology.

  • aneurysm
  • blood flow
  • stroke

Statistics from

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.


New neuroendovascular devices for the treatment of cerebral aneurysms are being introduced at a rapid pace and offer potential benefits for patients. Adequate training of physician investigators to allow for accurate evaluation of the safety and efficacy of novel devices within clinical trials is an increasingly important challenge. The woven endobridge (WEB) device is a woven intrasaccular flow disruptor that is used to treat intracranial wide-necked bifurcation aneurysms (WNBAs). Successful deployment within WNBAs depends on accurate sizing in order to ensure that the device is under compression and presents a barrier to inflow at the neck of the aneurysm.

We report the design and implementation of a unique neurovascular-replicator-based training program, which was incorporated into the Web Intrasaccular Therapy (WEB-IT) study, to optimize technical performance and patient safety.


WEB-IT study

The WEB-IT study is a prospective, multicenter, single-arm, interventional study conducted at 25 centers in the United States (USA) and six international centers. The study enrolled 150 adults with WNBAs of the anterior and posterior intracranial circulations. All candidates were screened and approved by the study principal investigators on the basis of inclusion and exclusion criteria, which have been previously published.1 No roll-in cases were permitted within the trial. The device was approved in the European Union (EU) before the initiation of the US investigational device exemption trial. Thus, only EU investigators had experience in treating patients with the WEB device before treating patients within the trial.

The study protocol was approved by each center’s institutional review board and all patients submitted written informed consent. The study was conducted under good clinical practices and included independent adjudication of all adverse events. An independent core laboratory adjudicated effectiveness outcomes. A data monitoring committee conducted study safety reviews.

Neurovascular replication

The vascular simulations neurovascular replicator (figure 1) consists of a mechanical heart pump connected to a silicone aorta that bifurcates distally into an iliofemoral system in which access ports are available for sheath placement. The heart pump duplicates the cardiac cycle with a functional left atrium and ventricle with mechanical mitral and aortic valves. The aortic compliance recreates the clinical Windkessel effect, resulting in pressure waveforms in the aorta of the replicator that are equivalent to the waveforms of a human aorta. A ‘homeostatic’ systemic pressure of 120/80 mm Hg is maintained via automatic feedback control of resistance valves and compliance chambers.

Figure 1

The replicator system containing basic arterial access vessels up to the intracranial circulation.

Proximally, the aortic arch anatomy gives rise to a standardized configuration of great vessels. The great vessels may be attached to standardized silicone cerebrovascular models with or without incorporated vascular pathology (figure 2), as well as patient-specific customized cerebrovascular anatomies (see ’Methods' section below). The silicone cerebrovascular models are housed in a clear, polyurethane ‘head’ and suspended with a gel such that the passage of wires and catheters into the cerebrovasculature mimics the straightening of vessels that would be experienced clinically.

Figure 2

Sample intracranial vessels that can be mounted within the replicator system.

The replicator vasculature is filled with a clear blood analog that has viscosity matched to human blood, which helps to duplicate the friction coefficients of endovascular devices, wires, and catheters as they traverse the vascular system. The blood analog fluid is automatically warmed up to body temperature (98.6 °F) by the replicator to allow normal expansion of temperature-dependent nitinol devices.

The whole system can be placed into any standard fluoroscopic/angiographic imaging environment and imaging performed. Standard fluoroscopy, roadmap imaging, digital subtraction angiography, and rotational angiography can be performed with any contrast agent and acquisition sequences employed in routine clinical practice. The physiological nature of the heart pump-recreated hemodynamics and the viscosity properties of the fluid allow for contrast density, dilution, and flow characteristics that would be expected within standard neuroangiographic procedures.

Patient-specific cerebrovascular replicas are based on cross-sectional vascular (magnetic resonance angiography or computed tomographic angiography) or, more optimally, 3D rotational neuroangiographic imaging studies. Digital imaging and communications in medicine (DICOM) data from these angiographic imaging studies were submitted for conversion to stereolithographic files for three-dimensional printing of a mold, which was then cast into a silicone replica of the cerebral vasculature. The patient-specific vascular model was then inserted into the replicator head shell and attached to the circulatory module.

Physician training

A novel training program was devised with the intention of maximizing physician interaction and minimizing didactic content. This program was divided into three principal segments:

  1. Centralized introductory training. Twelve introductory training programs were conducted at a centralized location. These 1-day programs were designed for small groups of physicians. Initial didactic material was presented regarding the medical device technology, terminology, and techniques. This was followed by a 2–3-hour session of hands-on surgical simulation using a library of increasingly difficult aneurysms. Participating investigators were confronted with progressive challenges of access anatomy, sizing decisions, and deployment techniques. Proctors worked directly alongside the training investigators often switching places in order to demonstrate technique. During a break, investigators discussed the challenges and solutions with proctors and specialists and reviewed additional cases together. Afterwards a second hands-on surgical simulation session was held with yet more cases.

  2. On-site training. Thirty-five on-site training sessions were held at participating study centers to refresh the principles and techniques before cases. During these sessions, investigators had the opportunity to use their own neurointerventional suites and personnel. This afforded centers the opportunity to ensure that radiology technologists and nurses were familiar with the surgical procedure. A checklist of drugs, devices, and bailout procedures that might be helpful in the event of a complication was reviewed.

  3. Patient-specific surgical rehearsals. Sixty-eight patient-specific surgical rehearsals in 65 sessions were held at participating centers using custom-created models of the participating patients to be treated at the center. The location of aneurysms treated during these rehearsals is depicted in figure 3. These rehearsals were usually held the day before initial cases at each center, but were also employed if cases were thought to be challenging or significant time had elapsed since there had been a case at that center. WEB device size selection was often a key consideration, and multiple candidate devices were often tested. These rehearsals served to better inform a stepwise strategy for the sequence of devices applied during the actual cases. In some cases, investigators decided not to move forward with WEB treatment based on experience gained during the surgical rehearsal.

Figure 3

Anatomical location of replicated aneurysms. Acom, anterior communicating artery; Basilar, basilar artery; ICAt, intraluminal carotid artery thrombus; MCA, middle carotid artery.


Comparison of WEB-IT procedural characteristics with those of other WEB trials

Thirty-day safety results within the WEB-IT trial have been published and efficacy results will not be available until the end of the investigational device exemption trial.1 Safety results within WEB-IT were comparable to those seen in European WEB trials, even though physicians in Europe generally had a range of case experience with the WEB device before enrolling in EU WEB studies while the US WEB-IT investigators did not.2–5

Additionally, we can compare procedure time and total fluoroscopy time within WEB-IT with those associated with European trials where these data are available. In 148 WEB-IT cases, the mean (SD) procedure time was 20.1 (21.2) min. This is not significantly different from procedure times within the WEB Clinical Assessment of Intrasaccular Aneurysm Therapy (WEBCAST) trial (20.2 min) or the French Observatory study (26.7 min). In 140 WEB-IT cases, total fluoroscopy time averaged 30.1 min, which was not significantly different from average fluoroscopy times in WEBCAST (37.0 min) and the French Observatory (38.7 min). These metrics are displayed in table 1.1 3 5 6

Table 1

Procedure and fluoroscopy times within WEB-IT, WEBCAST, and French Observatory studies

Surgical rehearsal with the replicator enabled physicians to practice device deployment and communication with proctors. The WEB devices used within the trial were available in a range of different dimensions, and in two distinct shapes (ie, cylindrical and round). An additional potential benefit was the opportunity to carry out trials of different devices within a given aneurysm. This enabled physician investigators to choose a device size for the clinical case. The identical device size was chosen for the clinical case in 60% of cases and a size within 1 mm of the size chosen in the replicator session was used in 79% of cases as depicted in figure 4. Figure 5 shows anteroposterior digital subtraction images of a WEB-IT patient with a basilar apex aneurysm immediately before treatment (A) and in the corresponding replicator model (B); an oblique view of a second WEB-IT patient’s left middle cerebral artery aneurysm (C) and the analogous image from that patient’s replicator session before treatment (D); a third WEB-IT patient’s basilar aneurysm immediately after deployment of the WEB device (E) and the corresponding image in the replicator (F); and a similar treated patient (G) and replicator (H) image for a fourth WEB-IT patient with a middle cerebral artery aneurysm.

Figure 4

Percentage of times device size tested either matched or was within 1 mm of clinical case.

Figure 5

(A) Anteroposterior digital subtraction angiogram of a WEB-IT participant’s basilar apex aneurysm and intracranial access before treatment. (B) Analogous image from that patient’s replicator session. (C) Anteroposterior digital subtraction angiogram of a second WEB-IT patient’s left middle carotid artery aneurysm before treatment. (D) Corresponding replicator image from the same patient. (E) Third WEB-IT patient’s basilar apex aneurysm immediately after treatment with early contrast stagnation at the apex of the aneurysm. (F) Corresponding replicator image from that patient’s session with the WEB in place deployed within the aneurysm. (G) Fourth WEB-IT patient’s right middle carotid artery aneurysm immediately after treatment. (H) Corresponding image from that patient’s replicator session with the WEB also deployed within the aneurysm.


Neurovascular devices and techniques are evolving rapidly. These new technologies must be evaluated within clinical trials to ensure that they are safe and effective, as well as to gain appropriate regulatory clearances. Operating physicians must be trained as efficiently as possible so that new devices and techniques may be implemented safely during clinical trials. Training programs also are required to facilitate the safe introduction of new devices into the broader physician community in clinical practice. In addition, optimized training of investigators is critical to minimize the impact of the ‘early learning curve’ on trial results, thus allowing the trial to reflect the true performance of the new technology. We describe a new program using the neurovascular replicator to augment physician training within the context of a clinical research trial.

The present experience indicates that it was feasible to train physicians to use the new WEB technology with a level of safety that was directly comparable to that observed within a group of experienced European physicians. The implementation of the replicator-based training program involved an initial skill building session consisting of practice on a library of aneurysms with varying levels of complexity under the guidance of a proctor. This was followed by the replication of patient specific anatomy and pre-case rehearsal under the guidance of both proctors and experienced clinical specialists. Experience obtained during the surgical rehearsal on the patient-specific anatomy helped operators to anticipate potential challenges with respect to access, device sizing, and device deployment. These sessions also afforded proctors and investigators a chance to build a rapport before working together on a human subject. The ability to perform patient-specific replicated cases within the operator’s angiography suite—on their own angiography equipment and with their assistants and full surgical team participating—provided a unique opportunity for the entire team to gain experience and familiarity with the WEB device and procedure. One particularly helpful aspect of this interaction was that it allowed the team to establish a lexicon of terms to optimize communication during the actual cases.

Simulation versus replication

The present replicator-based training is distinctly different from simulator-based training.7 8 In simulator-based training, a mechanical interface inputs the operator’s physical manipulations into a software program, which translates these movements into images that are displayed on a screen. The displayed results are purely derived from a programer’s ‘impression’ of how the device and physiological system should interact in response to various maneuvers. This characteristic is particularly limiting when the simulation module is based on minimal clinical experience with a new device and a less well-developed understanding of the behavior of the device within the cerebrovasculature, such as in the present situation with the WEB. In this scenario, neurovascular replication provides operators with the ability to fully manipulate, deliver, and deploy any new device within a validated anatomical and physiological model under live fluoroscopy, thus overcoming many of the challenges associated with simulation.

Limitations of replication and this report

The replicated system is only as good as the data from which it is derived. Accurate depiction of the regional vascular anatomy is predicated on having source data of the highest possible spatial resolution. In our experience, conventional catheter-based rotational three-dimensional angiography provided the best possible data from which to construct a stereolithographic file that most accurately depicted the anatomy. When computed tomographic angiography or magnetic resonance angiography data were used, we found that the replicated anatomy was less reliable. While the use of the system provides the operator with an opportunity to use actual devices within a validated physiological system, it is important to acknowledge that this benefit comes at a financial expense in that numerous catheters, wires, and implants are required for the completion of a comprehensive replicator training. Additionally, while the replicator provides the advantage of working under live fluoroscopy within an angiographic laboratory, the process also involves radiation exposure for the operators and trainers. Finally, while basic procedural metrics can be compared with those of similar trials, there was no prospective intent, nor ability, to prove the effect of this replicator-based training program within the US WEB-IT trial. The current assessment was merely structured to show feasibility and report this experience. Future studies will be required to validate the added value of replicator-based training within device trials. Comparative trials will be required to determine their relative value in comparison with simulator-based training.


A new program of neurovascular-replicator-based physician training was employed within the WEB-IT study. This represents a new methodology for education and training that is feasible, and may be an effective means of optimizing technical success and patient safety during the introduction of a new technology. Future prospective studies will be required to quantify the value of replicator-based training within the context of approval trials for new clinical devices and in the subsequent early commercialization of these devices.


The authors thank Andrew J Gienapp, BA (Department of Medical Education, Methodist University Hospital, Memphis, Tennessee, USA and Department of Neurosurgery, University of Tennessee Health Science Center, Memphis, Tennessee, USA) for copy editing, preparation of the manuscript and figure for publishing, and publication assistance.



  • Contributors All authors of this work met ICMJE criteria for authorship and made substantial contributions to the conception and design, acquisition of data, analysis and interpretation of data, drafting, critical revising, and final approval of this manuscript.

  • Funding The WEB-IT study was supported by Sequent Medical Inc.

  • Competing interests The primary Investigators for the WEB-IT trial received institutional salary support for study-related activities. Investigators in the WEB-IT trial also received payment for proctoring cases within the context of the trial. AA received personal fees from Sequent during the conduct of the study; is a consultant for Leica, Medtronic, Microvention, Penumbra, Siemens, and Stryker; receives research support from Microvention, Penumbra, and Siemens; and is a shareholder in Bendit, Cerebrotech, Serenity, and Synchron outside of the submitted work. DH received personal fees from Sequent during the conduct of the study; and serves as a consultant for Covidien outside of the submitted work. AC received personal fees from Sequent during the conduct of the study; and serves as a consultant for Medtronic, Microvention, and Stryker Neurovascular outside of the submitted work. JDA received personal fees from Sequent during the conduct of the study; and serves as consultant for Medtronic, Penumbra, and Sequent outside of the submitted work. LE received personal fees from Sequent during the conduct of the study; and serves as a consultant for Codman Neurovascular, Medtronic, MicroVention, Penumbra, Sequent, and Stryker outside of the submitted work. SC received personal fees from Sequent during the conduct of the study; and serves as consultant for Medtronic, MicroVention, Sequent outside of the submitted work. DF received personal fees from Sequent during the conduct of the study; and is a consultant for Medtronic, Stryker, Microvention, Penumbra, and Codman; receives research support from Medtronic, Microvention, Penumbra, and royalties from Codman; and is a stockholder for Vascular Simulations outside of the submitted work.

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

  • Data sharing statement There are no additional data to share.

  • Collaborators Methodist University Hospital Neuroscience Institute & Cancer Center, Memphis, TN (21), PI: Adam Arthur, MD, Sub-I: Daniel Hoit, MD, Lucas Elijovich, MD; Stony Brook University Medical Center, Stony Brook, NY (12), PI: David Fiorella, MD, Sub-I: Henry Woo, MD; The Johns Hopkins Hospital, Baltimore, MD (12) PI: Alexander Coon, MD Sub-I: Geoffrey Colby, MD; Koru Hospital, Ankara, Turkey (11) PI: Isil Saatci, MD, Sub-I: Saruhan Cekirge, MD; National Institute of Neurosciences, Budapest, Hungary (10) PI: Istvàn Szikora, MD; Marmara University Faculty of Medicine Pendik Training and Research Hospital, Istanbul, Turkey (9) PI: Feyyaz Baltacioğlu, MD, Sub-I: Ruslan Asadov, MD; Brigham and Women’s Hospital, Boston, MA (8), PI: Ali Sultan, MD Sub-I: Ram Chavali, MD, Kai Frerich, MD; Abbott Northwestern Hospital, Minneapolis, MN (7), PI: Josser Delgado, MD Sub-I: Yasha Kayan, MD; Lyerly Neurosurgery, Baptist Medical Center, Jacksonville, FL (7), PI: Ricardo Hanel, MD Sub-I: Eric Sauvageau, MD; Medical University of South Carolina, Charleston, SC (7), PI: Raymond Turner, MD, Sub-I: Imran Chaudry, MD, Alejandro Spiotta, MD, Aquila Turk, MD; West Virginia University Medical Center, Morgantown, VA (6) PI: Ansaar Rai, MD, Sub-I: Jeffrey Carpenter, MD, Sohyun Boo, MD; Thomas Jefferson University Hospital, Philadelphia, PA (6), PI: Pascal Jabbour, MD, Sub-I Stavropoula Tjoumakaris, MD, Robert Rosenwasser, MD; Fort Sanders Regional Medical Center, Knoxville, TN (5), PI: Keith Woodward; Albany Medical Center, Albany, NY (4) PI: Alan Boulos, MD, Sub-I: John Dalfino, MD, Junichi Yamamoto,MD; Riverside Methodist Hospital, Columbus, OH (4) PI: Nirav Vora, Sub-I: Ronald Budzik,MD. Peter Pema, MD, Thomas Davis, DO; Baptist Memorial Hospital, Memphis, TN (3), PI:Daniel Hoit, MD, Sub-I: Adam Arthur, MD, Lucas Elijovich, MD; Mayo Clinic, Rochester, MN,(3): PI: David Kallmes, MD, Sub-I: Giuseppe Lanzino, MD, Harry Cloft, MD; University at Buffalo Medical Center, Buffalo, NY (2), PI: Adnan Siddiqui, MD, Sub-I: Elad Levy, MD,Kenneth Snyder, MD; Rush University Medical Center, Chicago, IL (2), PI: Demetrius Lopes,MD Sub-I: Michael Chen, MD; Mount Sinai Roosevelt Hospital, New York, NY (2) PI: JDMocco, MD, Sub- I: Johanna Fifi, MD; University of Texas Medical School, Houston, TX (2) PI:Peng Roc Chen; Royal University Hospital, Saskatoon Canada, (2), PI: Michael Kelly, MD, Sub- I: Lissa Peeling, MD; Swedish Medical Center, Englewood, CO (1) PI: Don Frei, MD, Sub-I: Daniel Huddle, DO, Richard Bellon, MD, David Loy, MD; Barrow Neurological Institute, St. Joseph’s Hospital and Medical Center, Phoenix, AZ (1) PI: Felipe Albuquerque, MD Sub-I: Andrew Ducruet, MD; Righospitalet, University of Copenhagen, Copenhagen, Denmark (1), PI: Marcus Holtmannspötter; University of Louisville Hospital, Louisville, KY (1), PI: Robert James, MD, Sub-I: Wei Liu, MD; Helios General Hospital, Erfurt, Germany (1) PI: Joachim Klisch, MD, Sub-I: Christoph Strasilla, MD, Christin Clajus, MD.