We would like to thank our colleagues for reading our publication and for the thorough review of the paper. We greatly look forward to reading their own experience and believe that it will add substantially to the literature on this new and interesting device.

At the outset we would like to make it clear that we stand by our initial comment in saying that we believe the combination of a MED and other devices, in particular intra-luminal FDS may have excellent and extremely rapid results. To that end we are currently in the process of collating our cases of this nature in preparation to submit for publication but we have seen extremely rapid aneurysm exclusion.

The use of adjunctive devices in some instances was not planned. In one case, which was discussed in the original article, after deployment of the MED it was noted that the residual flow into the aneurysm had been inadvertently redirected into an aneurysm bleb. We do not believe this was due to a sizing issue but rather an effect that can occur from the position of the petals which can in our experience be relatively difficult to determine even after deployment of the MED. We did not feel this was a safe situation to leave the aneurysm and therefore treated the bleb with coils. A similar situation was encountered with the basilar tip aneurysm. In both scenarios we felt uncomfortable leaving parts of the aneurysm unprotected. The use of FDS was to treat the aneurysm neck in certain cases e.g. the patients represented in Fig. 8 and 9. However, in another case the use of the FDS was not to treat the aneurysm with MED's but an adjacent blister like aneurysm (Fig. 10). Furthermore, in patient 9, again there were multiple aneurysms on the right with only one suitable for treatment with MED and therefore the placement of the FDS subsequently served as an additional treatment strategy for the other smaller aneurysms as well as the aneurysm that contained the MED. This patient represented one of our first patients in whom we used the MED and hence our understanding of its pro's and con's was not fully developed.

Our reasoning for the deployment of MED and FDS was covered in the discussion section of our publication. However, in our literature search for our upcoming paper we have come across evidence which we believe supports our initial assertion of the effectiveness of MED+FDS. Jing et al [1] recently published their work on the heamodynamic effect of flow diverters in combination with coils in large and giant aneurysms. The FDS used in this study was the Tubridge FDS. Their computational fluid dynamics (CFD) study showed that the combination of coils and FDS resulted in significantly decreased intra-aneurysmal flow velocity, wall shear stress and increase in low wall shear area with a decrease of 54%, 66% and increase of 1001% respectively. Damiano et al [2] recently showed that the principle action of FDS, in the early stages of treatment, are flow redirection away from the aneurysm with 2 overlapping FDS compounding the effect but no significant increased effect seen with the addition of a third FDS. In this study the role of coils was to disturb aneurysmal flow, reduce intra-aneurysmal flow velocity and WSS and disturb aneurysmal flow patterns when the coil packing density exceeds 11%. Below this level the coils may have a scaffolding effect for intra-aneurysmal thrombus but do not appear to have a significant heamodynamic effect. It is likely that the MED has a heamodynamic effect greater than 11% coil packing density, although we are unaware of any studies that have looked at this. Therefore, the combination of MED and FDS is likely similar to that of FDS and coils with high packing density, a strategy which has shown a higher occlusion rate and lower rate of retreatment in a recent study by Lin et al. [3].

Whilst it is indeed true that the MED was not used as the sole device in many of our aneurysms our primary aim when using this device was to determine its effectiveness in a 'real-world' setting. As the device is designed to adopt a spherical shape, and if it was only used in spherical aneurysms, it would only be suitable for use in approximately 1/3 of unruptured aneurysms according to the study of You et al [4]. This problem was covered in the discussion. We chose to use the MED in non- spherical aneurysms because we believed that the flow disruption caused by the device inside the aneurysm would nevertheless promote thrombosis. Furthermore, in the first publication regarding the MED, by Turk et al[5], 1/3 of the aneurysms were not 'spherical'. For example, patient 1 in the series of Turk et al. had an aneurysm with dimensions 12.1 x 7.24 x 5.47mm and yet at 1 month follow-up occlusion was reported as 95%. In addition this group also used a 'sequential' filling of an ACOM aneurysm that is represented in Fig. 3 and 4 in their publication. This was a strategy that we have used that takes advantage of the flow disrupting effect but not really of the shape the MED adopts.

The use of adjunctive devices such as the pCONus may or may not hamper the deployment of the device but as of yet there is no evidence to suggest a conclusion. The assertion that the pCONus somehow interfered with the deployment of the MED appears to be nothing more than conjecture. Our standard procedure is to mechanically detach the pCONus as the last step in any procedure involving it. As such we may reposition it in during the coiling or MED deployment procedure and hence we may reposition it to allow maximum neck protection and minimum 'hampering' with any intra -saccular devices be they coils or MED's.

We agree with the authors that the follow-up period is short and we hope to publish subsequent studies in the future with longer-term follow- up. However, the follow-up period in the original Turk et al. [5] paper was also short with 1 month angiographic follow-up available for 5/9 aneurysms and no occlusion seen in 2 of the 5 with early follow-up. Our publication was not meant to be a study with long term follow-up, rather our initial experience and insights that we believed may be useful to our colleagues should they wish to use the MED. We are therefore slightly confused as to why short-term follow-up is a problem in our paper but not in the paper of Turk et al.

With regards to sizing, this is perhaps also a matter of debate. It is important to recall that very recently the sizing recommendations of the WEB were altered based on clinical experience and observation. As discussed earlier the problem with sizing relates to the spherical shape of the MED. In our cases that are said to be undersized the MED was actually sized to fit the most 'spherical' part of the aneurysms. By oversizing the MED one cannot predict how the leaflets will position themselves and then one cannot predict the effect on flow created by the device. This was described for one of our cases (Fig. 5c) despite using our sizing technique and this inability to accurately predict the position of the leaflets has been one of our main observations.

As suggested by the authors coils would be an alternative, and in fact would be an excellent alternative in all of these cases, although the additional uses of stents, neck bridging devices etc. may still be needed. The issue of cost is of course important however, with the current cost of the device relative to coils, the use of 2 MED's is approximately the equivalent cost to 40coils - and very few aneurysms would require this many coils. In fact very few aneurysms require 20 coils and therefore, the issue reverts to the speed and security of any occlusion obtained. This then reverts back to our original supposition that the combination of intra-saccular and intra-luminal flow diversion. We believe that a rapid and durable aneurysm exclusion can be achieved through a single intra- luminal flow diverter and in all likelihood only one or two MED's.

In conclusion we thank the authors for their comments and look forward to reading their publication in the coming months. We hope they will also enjoy our second publication dedicated to the use of intra- luminal and intra-saccular flow diversion.

Funding Statement This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors

Author Contribution MAP, PB, RMM - data gathering, manuscript preparation, OG, HB - review, editing HH - guarantor, overall review, study design

Conflict of interest MAP, PB and RMM serve as proctors and consultants for phenox GmbH, with moderate financial compensation. HH is a co-founder and shareholder of phenox GmbH. The other authors have no potential conflict of interest.

References: 1 Jing L, Zhong J, Liu J, et al. Hemodynamic Effect of Flow Diverter and Coils in Treatment of Large and Giant Intracranial Aneurysms. World Neurosurg 2016;89:199-207. doi:10.1016/j.wneu.2016.01.079 2 Damiano RJ, Ma D, Xiang J, et al. Finite Element Modeling of Endovascular Intervention Enables Hemodynamic Prediction of Complex Treatment Strategies for Coiling and Flow Diversion. J Biomech 2015;48:3332-40. doi:10.1016/j.jbiomech.2015.06.018 3 Lin N, Brouillard AM, Krishna C, et al. Use of coils in conjunction with the pipeline embolization device for treatment of intracranial aneurysms. Neurosurgery 2015;76:142-9. doi:10.1227/NEU.0000000000000579 4 You S-H, Kong D-S, Kim J-S, et al. Characteristic features of unruptured intracranial aneurysms: predictive risk factors for aneurysm rupture. J Neurol Neurosurg Psychiatry 2010;81:479-84. doi:10.1136/jnnp.2008.169573 5 Turk AS, Maia O, Ferreira CC, et al. Periprocedural safety of aneurysm embolization with the Medina Coil System: the early human experience. J Neurointerventional Surg 2016;8:168-72. doi:10.1136/neurintsurg-2014- 011585

Conflict of Interest:

MAP, PB and RMM serve as proctors and consultants for phenox GmbH, with moderate financial compensation. HH is a co-founder and shareholder of phenox GmbH. The other authors have no potential conflict of interest.

Thank you for your technical considerations regarding stent in stent placement without hooking the first stent.

Use of 3 D Roadmap may be helpful. Moreover, following passage of the microwire, reconstructions of a second flat panel angioCT with the microwire in place clearly outlines the relationship between the microwire and the struts of the first stent, especially if reconstructions perpendicular to the orientation of the first stent are performed. This may avoid the passage of the stent with a DAC (Distal Access Catheter) which may demage or displace the first stent, especially at it's proximal end.

Using this technique, even passage through a deformed and fragmented stent can be performed, and the location of the microwire within the central axis of the first stent can be confirmed before introduction of the second stent or flow diverter.

Mordasini P, Al-Senani F, Gralla J, Do D, Schroth G: The use of flat Panel angioCT (DynaCT) for Navigation through a deformed and fractured carotid stent. Neuroradiology 52; 2010:629-632.

Prof. Gerhard Schroth Neurologist and Radiologist gerhard.schroth@insel.ch University of Bern Senior Consultant of the Institute for Diagnostic and Interventional Neuroradiology

3010 Bern - Inselspital Switzerland

Conflict of Interest:

None declared

Frederic Clarencon, MD, PhD 1, 2, Nader-Antoine Sourour, MD 1 *

1. Department of Interventional Neuroradiology. Pitie-Salpetriere Hospital.

APHP. Paris France. 2. Paris VI University Pierre et Marie Curie. Paris. France

* Corresponding author

We read with great interest the case series entitled: "The Medina Embolic Device: early clinical experience from a single center" by Aguilar Perez M et al, recently published online in JNIS. 1 We would like to congratulate the authors for this interesting feedback from their preliminary experience, the largest published to date, with a new device that combines the design of a detachable coil and the one of a intrasaccular flow disrupter device: the Medina Embolization Device (MED) (Medtronic, Irvine, CA). 2 However, we would like to make some comments and raise some disagreements with the Methods used in this retrospective case series. First, in the Abstract, the authors state that their objective was "To report (their) initial experience with the Medina Embolic Device (MED) in unruptured intracranial aneurysms either as sole treatment or in conjunction with additional devices". However, only 2/15 patients (13%) were treated by means of MED alone or MED with filing coils. Although we agree with the authors that the use of adjunctive devices such as balloon remodeling, support devices such as pCONus (Phenox, Bochum, Germany) or PulseRider (Codman Neurovascular, Raynham, MA) or even filing coils deployed in the MED's mesh, may not, by themselves, dramatically affect the angiographic outcome and the occlusion rate, we would like to stress the fact that additional flow diverter stent (FDS) deployed during the same session or after early follow-up, prevents from any conclusion about the angiographic outcome with the MED. Indeed, FDS (p64, Phenox) were deployed in 47% (7/15) of the cases in this series either during the procedure or after early follow-up. Since, FDSs are not adjunctive tools but therapeutic tools by themselves, they may have a dramatic influence on the angiographic outcome. Additionally, one may wonder why using MED for loose intra-saccular packing in large paraclinoid aneurysms treated by FDS (like in Patients # 9 and 14). The use of regular coils would be easier, sufficient for aneurysm occlusion and more cost-effective.

Second, we would like to underline the fact that the use of a support device like the pCONus (20% [3/15] of the cases in this series), with flares deployed inside the aneurysm, may hamper the satisfactory deployment of the MED's petals, especially at the neck. This may lead to an incomplete sealing of the neck and finally to inadequate occlusion of the aneurysm, as shown in the Figure 7 of the article.

Third, the delay for the angiographic follow-up is very short (average: 2.2 months). We believe that it is impossible to evaluate the effect of an endosaccular device with such a short-term follow-up, especially if patients are under dual anti-platelet therapy (in at least 13/15 patients in this series). As observed with other endovascular flow disruption devices such as the WEB (Sequent Medical, Aliso Viejo, CA) or LUNA (NFocus/Covidien, Irvine, CA), the mechanism that leads to aneurysm occlusion is a gradual thrombosis triggered by the flow disruption effect. 3 Consequently, the occlusion of the aneurysm's sac obtained with MED may be delayed and cannot be evaluated with an as short delay as the one provided in this article.

Finally, we have some concerns about the choice of the aneurysms treated by MED in this series and about the sizing. Indeed, the MED is a spherical device that may be suitable mostly for spherical or ovoid aneurysm, especially with wide neck. Probably a conical shape (patient # 6) or an irregular giant aneurysm as presented in Figure 12, are a priori not suitable for the MED. We also tried to find the aneurysms' size as well as necks' size in the article, in vain. For instance, on Figure 8, the aneurysm's largest diameter is obviously over 10 mm. However, the largest diameter available to date in MED in 9 mm. Additionally, why the authors chose to undersize the MED in spherical aneurysms ("In spherical saccular aneurysms with a fundus diameter of 9 mm or less, the size of the first framing MED was purposely undersized by about 1 mm"). We do believe that one should at least use the same size as the aneurysms' maximum diameter for the first MED Framer or even oversize by 1 mm. The rational of our strategy is that it allows for a satisfactory application of the petals along the aneurysm's wall, and across the neck, which is of tremendous importance to obtain a neck sealing and subsequently a flow disruption effect. In the Figures 5, 6 and 8, the MED is obviously undersized, which lead to incomplete covering of the neck by the petals and thus to residual filing of the sac and the bleb. In our early monocenter experience, presented at the ABC-WIN 2016 congress 4 and under consideration for publication in another journal, using a proper sizing strategy, we obtained an 83% (10/12 cases) occlusion rate at angiographic follow-up (average delay: 5.2 months).

Owing to the above-mentioned drawbacks in the Methods of this case series, we think that the only conclusion that can be drawn from this article is that the MED can be safely deployed in an intracranial aneurysm in a human been. Other considerations on the effectiveness in terms of angiographic outcome cannot be supported by the data presented in this series.

References 1. Aguilar Perez M, Bhogal P, Martinez Moreno R, et al. The Medina Embolic Device: early clinical experience from a single center. J Neurointerv Surg 2016. 2. Turk AS, Maia O, Ferreira CC, et al. Periprocedural safety of aneurysm embolization with the Medina Coil System: the early human experience. J Neurointerv Surg 2015. 3. Asnafi S, Rouchaud A, Pierot L, Brinjikji W, Murad MH, Kallmes DF. Efficacy and Safety of the Woven EndoBridge (WEB) Device for the Treatment of Intracranial Aneurysms: A Systematic Review and Meta-Analysis. AJNR Am J Neuroradiol 2016. 4. Sourour N, Di Maria F, Vande Perre S, Gabrieli J, Chiras J, Clarencon F. Medina devices in the treatment of wide neck intracranial aneurysms: single-center preliminary experience. ABC-WIN annual congress, 2016.

Conflict of Interest:

F. Clarencon is consultant for Medtronic and Codman Neurovascular NA Sourour is consultant Medtronic , MicroVention and investor for Medina

Letter by Parthasarathy et al. regarding article, "Unwanted detachment of the Solitaire device during mechanical thrombectomy in acute ischemic stroke ".

Rajsrinivas Parthasarathy MRCP (UK) Neurology, Vipul Gupta MD, Gaurav Goel MD DM

Department of Neurointerventional surgery, Institute of Neuroscience, Medanta, the Medicity, Gurgaon, India.

Title word count: 14

Word Count: Abstract: 157; Manuscript count excluding references: 554

References: 4

Key words:Stentriever, mechanical thrombectomy, detachment, stent based retrieval

Corresponding Author

Dr Vipul Gupta MD, Additional Director and Head, Neurointerventional Surgery, Medanta Institute of Neurosciences Medanta - the Medicity, Gurgaon 122001 Tel: 00919810542372 drvipulgupta25@gmail.com

Abstract:

Stent detachment is a dreaded device complication in the setting of acute stroke management and may result in poor outcome. The principal objective in the event of stent detachment is to establish flow in the occluded territory. Techniques including balloon angioplasty, local infusion of antiplatelet or thrombolytic agent and suction have been described with variable success in achieving meaningful reperfusion. Stent retrieval can potentially restore flow to the occluded territory and negate the need for administering dual antiplatelet. We read with interest the article by Castano et al (2016) on unwanted detachment of the Solitaire device during mechanical thrombectomy in acute ischemic stroke.Stents with type 'A' detachment were retrievable; whereas, attempts at retrieving stents with type B detachment were invariably unsuccessful. Therefore, a rescue strategy, 'stent based retrieval',may prove to be useful when snare retrieval fails. Stent based retrieval could be considered in both type A and type B detachments when snare retrieval is unsuccessful.

Letter to the Editor

We read with great interest the article by Casta?o et al (2016) on unwanted detachment of the Solitaire device(Medtronic/Covidien/ev3, Dublin, Ireland) during mechanical thrombectomy in acute ischemic stroke.1 Stent detachment in their series was invariably associated with a poor outcome, higher rates of symptomatic intracranial hemorrhage and higher mortality.

They had classified stent detachment as either type A or type B based on if the separation occurred before or after the proximal radiopaque stent marker. They attempted stent retrieval with an Amplatz GooseNeck snare (Medtronic/Covidien/ev3, Dublin, Ireland) in all irrespective of the type of detachment.A rescue strategy in case of failure of snare retrieval was not described.Stents were retrievable with 'type A' detachments; however, attempts to retrieve stents with 'type B' detachmentwere invariably unsuccessful. The likely explanation is that with type B detachment the 'stent legs' either open or stay together making a 'spearhead' that digs into the wall of the artery making snare retrieval not possible.

The primary aim in such a situation is to re-establish flow in the occluded territory. A number of techniques including balloon angioplasty, local infusion of antiplatelet/ thrombolytic agent, and suction have been described with variable success in achieving meaningful reperfusion.(2,3) Furthermore, leaving a stent in situ necessitates administration of dual antiplateletto a patient in whom the infarct core can increase either due to delayed occlusion or failure to recanalise. Therefore, an attempt to restore flow by retrieving the detached stent is likely to be crucial to achieving a good outcome. Castano et al (2016) were able to achieve meaningful reperfusion in 50% of patients; however, the outcomes were poor despite device retrieval.(1) This further reiterates the importance of time and the need to establish perfusion early. Therefore, a second strategy may be beneficial when snare retrieval fails. Snare retrieval may not be successful in type A detachments and not likely to be the appropriate strategy in type B stent detachments. Therefore, an alternate strategy, 'Solitaire stentectomy' may prove to be effective under these circumstances. (4) 'Stent based retrieval' using a Solitaire device can be performed as a '4 step' process in a patient with stent detachment. Step 1: Partial deployment of an appropriately sized solitaire device proximal to the detached stent [the distal end of microcatheter is positioned proximal to the proximal radiopaque marker/ proximal legs (struts) of the detached stent and then the stent is partly unsheathed to allow for its expansion]. Step 2: Engaging the proximal end of the detached stent by the distal end of the second stent by advancing the microcatheter and the stent together. Step 3: re-sheathing the device in an attempt to capture the proximal legs/ struts of the detached stent. Step 4: retrieval of microcatheter, device and detached stent. This strategy may be useful in both type A and type B detachments when snare retrieval fails.

Clearly, stent detachment is a dreaded device complication for the neurointerventionist in the setting of acute stroke management and may result in poor outcome. Development of methods and techniques is crucial to dealing with stent detachments in a timely and effective manner. "Stent based retrieval" can be a useful alternate technique when snare retrieval fails and may be employed as the primary technique in type B detachments.

References: 1. Castano C, Dorado L, Remollo S, et al. Unwanted detachment of the Solitaire device during mechanical thrombectomy in acute ischemic stroke. J Neurointerv Surg. 2016 Jan 27. [Epub ahead of print]

2. Kim ST, Jin SC, Jeong HW, et al. Unexpected detachment of solitaire stents during mechanical thrombectomy. J Korean Neurosurg Soc 2014;56:463-8.

3. Yub Lee S, Won Youn S, Kyun Kim H, et al. Inadvertent detachment of a retrievable intracranial stent: review of manufacturer and user facility device experience. Neuroradiol J 2015;28:172-6.

4. Chapot R, Stracke P, Nordmeyer H, Heddier M. Stentectomy: Retrieval of stents after stent assisted coiling. Interventional Neuroradiology 2015; 21(1S):160

Conflict of Interest:

None declared