Nohra Chalouhi M.D., Stavropoula Tjoumakaris M.D.,and Pascal Jabbour M.D.

We read with great interest the article by Turk et al[1] assessing the safety and efficacy of endovascular stroke intervention based on CT perfusion (CTP) criteria. The authors are to be congratulated for this excellent report and for achieving remarkably high rates of favorable outcomes (42%) in stroke patients with poor neurological status (mean NIHSS, 18.2). They further categorize their patient population based on time from symptom onset and clearly show that patients treated after 8 h had no difference in outcomes or mortality than those treated before 8 h. The fact that time from symptom onset to intervention was as long as 16.4 h in the late group (>8h) lends further credence to their findings because the extent of salvageable brain tissue and the prospect of improvement significantly decrease with delayed recanalization.[2-4] This report adds substantially to the growing body of literature supporting the use of CTP in stroke patients and reinforces what we have always believed; specifically, CTP can optimize and guide patient selection for intraarterial therapy based on an individual's physiological parameters namely the extent of salvageable ischemic penumbra.

Although we agree with the authors about the merits of CTP and stress the importance of their results, we would like to bring to their attention that their conclusion may not be totally supported by the presented data. The authors have compared functional outcomes in patients treated <8 h and >8 h after symptom onset. Although all patients were selected for endovascular stroke intervention based on CTP criteria, they conclude that "physiologic imaging-guided patient selection rather than time for endovascular reperfusion in ischemic stroke may be effective and safe". We believe the only way to reach such a conclusion is to compare two groups of patients, one selected based on CTP criteria and the other based on time from symptom onset. Indeed, the authors do not discuss the study by Hassan et al[5] published in 2010 that retrospectively compared 69 patients undergoing CTP-guided and 127 patients undergoing time-guided endovascular treatment. The authors of this study found no incremental benefit with the use of CTP, with similar rates of recanalization, intracranial hemorrhage, favorable outcomes, and in-hospital mortality in both groups. It should be noted, however, that only 40% of patients in their study underwent mechanical thrombectomy and that CTP-guided treatment at the primary author's institution was compared to time-guided treatment at a different institution, which could have influenced their results. Still, this is to our knowledge the only study that has compared time-guided to CTP-guide patient selection for acute stroke intervention. Randomized controlled trials comparing the 2 strategies are needed especially that many prominent centers still have not adopted CTP as a screening tool in this setting.

We would also like to highlight some of the potential shortcomings associated with CTP that were not discussed in this study. The inter- observer variability remains a great concern with CTP.[6] Additionally, the optimal post-processing algorithm for defining penumbra and core infarct have yet to be determined, and thresholds for guiding therapy are still under investigation.[7] The variation in reconstruction of CTP images and qualitative interpretation of salvageable tissue may lead to selection of a relatively heterogeneous population, leading to the inclusion of patients with limited salvageable tissue. False negatives and non interpretable imaging can occur with CTP due to low cardiac output, inappropriate slow rate of bolus administration, contrast extravasation in the subcutaneous tissue, patient movement, and operator inexperience.

Finally, despite the potential caveats of CTP imaging , Turk et al[1] have clearly demonstrated that CTP allows effective treatment of many patients who otherwise would be destined for supportive management or end of life care due to presentation outside of the therapeutic window. We commend the authors for their rigorous work and await similar contributions that will help us determine the best approach for patient selection for acute stroke intervention.

References 1. Turk AS, Magarick JA, Frei D, et al. CT perfusion-guided patient selection for endovascular recanalization in acute ischemic stroke: a multicenter study. J Neurointerv Surg 2012 doi: neurintsurg-2012-010491 [pii] 10.1136/neurintsurg-2012-010491[published Online First: Epub Date]|. 2. Wardlaw JM, Murray V, Berge E, et al. Recombinant tissue plasminogen activator for acute ischaemic stroke: an updated systematic review and meta-analysis. Lancet 2012;379(9834):2364-72 doi: S0140-6736(12)60738-7 [pii] 10.1016/S0140-6736(12)60738-7[published Online First: Epub Date]|. 3. Hesselmann V, Niederstadt T, Dziewas R, et al. Reperfusion by combined thrombolysis and mechanical thrombectomy in acute stroke: effect of collateralization, mismatch, and time to and grade of recanalization on clinical and tissue outcome. AJNR Am J Neuroradiol 2012;33(2):336-42 doi: ajnr.A2746 [pii] 10.3174/ajnr.A2746[published Online First: Epub Date]|. 4. Vergouwen MD, Algra A, Pfefferkorn T, et al. Time Is Brain(stem) in Basilar Artery Occlusion. Stroke 2012 doi: STROKEAHA.112.666867 [pii] 10.1161/STROKEAHA.112.666867[published Online First: Epub Date]|. 5. Hassan AE, Zacharatos H, Rodriguez GJ, et al. A comparison of Computed Tomography perfusion-guided and time-guided endovascular treatments for patients with acute ischemic stroke. Stroke 2010;41(8):1673-8 doi: STROKEAHA.110.586685 [pii] 10.1161/STROKEAHA.110.586685[published Online First: Epub Date]|. 6. Hassan AE, Zacharatos H, Chaudhry SA, et al. Agreement in endovascular thrombolysis patient selection based on interpretation of presenting CT and CT-P changes in ischemic stroke patients. Neurocrit Care 2012;16(1):88 -94 doi: 10.1007/s12028-011-9577-9[published Online First: Epub Date]|. 7. Amenta PS, Ali MS, Dumont AS, et al. Computed tomography perfusion- based selection of patients for endovascular recanalization. Neurosurg Focus 2011;30(6):E6 doi: 10.3171/2011.4.FOCUS10296[published Online First: Epub Date]|.

Conflict of Interest:

None declared

As a participant in the creation of the original ACGME ESNR fellowship training standards, I read the article concerning fellowship suspension with great interest (1). The paper is timely and thorough. It is unfortunate that the Neurointerventional world is dealing with so many unknowns. We do not seem to know how many we are, where we work, what we do, or where we were trained. The problem is very well summarized at the end of the paper. We seem to be spreading our patient population among too many practitioners, which subsequently restricts the number of patients available to each practitioner to stay in practice, make a living, and be proficient.

The first line states: "The purpose of any training program is to provide a supply of skilled workers to address an unmet demand for their services." The authors discuss numbers of procedures and extrapolate implications but do not discuss what the training was that got these people out into the world. Further, there is no clear statement as to what these unmet needs are. These workers might perhaps be skilled in certain areas, but not know anything about other subjects.

Adequacy of numbers is the primary thrust of the article. One endovascular neurosurgeon can possibly cover 5 different hospitals by themselves for aneurysms, AVMs etc., and could do 300 aneurysms per year easily. Most, but not all, neurointerventional fellowships provide an adequate number of aneurysms for training. But as correctly pointed out, the numerous auxiliary procedures that are necessary to be proficient in this profession, such as skull base tumor embolization, epistaxis, vertebroplasty, nerve root block or ablation, are not routinely performed at all institutions. AVMs might be completely ignored at many places. Endovascular spinal procedures are rare.

The gaping hole in fellowship training is acute ischemic stroke. Consistently complete training for all fellows continues to be a void in our profession and a gap in the entire premise of this statement. It is perfectly acceptable to produce specialists in aneurysm intervention as well as vertebroplasty, but we should not count them all in the same bucket and presume we have enough people to cover the nation for stroke. Everyone does not do everything nor do they want to. Many fellowship programs have inadequate caseload and resultant inadequate or poor training in acute stroke treatment. We should not just expect this specialized training to magically appear in fellowships where there is inadequate case volume to begin with. Therefore, what is the true number of interventionists that are available, trained, able, and willing to treat strokes?

The need for trained physicians to treat stroke is not based upon the number of strokes (as might be the case for aneurysms), but rather on the number of hospitals that need stroke coverage. The issue with stroke coverage is that there needs to be a minimal number of practitioners at each hospital to take call 24/7/365, no matter the number of cases: 1 per month or 1 per day. Another paper describes the necessity of stroke coverage as 2-3 interventionists per center for 24/7/365 coverage (2), contradicting the official SNIS stated position as well as EMTALA (Emergency Medical Treatment and Active Labor Act) that states that no requirement can be made for 24/7 coverage with less than 4 persons. If there is a need for 200 hospitals to have 24/7/365 coverage, then there needs to be at least 800 interventionists just at those hospitals alone.

The official SNIS position on stroke coverage is as follows: "Current staffing levels at Comprehensive Stroke Centers will rarely allow continuous immediate coverage for IA stroke therapy at all times, 365 days a year. Instantly available coverage on a continuous basis 24/7/365 should not be expected with fewer than 4 persons. Until adequate staffing levels are achieved, continuously available catheter-directed intra- arterial emergency stroke therapy is considered the "ideal", but is not mandatory in order to be classified as Comprehensive Stroke Center. Therefore, in agreement with federal regulations, IA stroke therapy might be available only a portion of the time as determined by local hospital conditions and staffing levels of qualified and appropriately trained personnel, and the institution still qualify as a Comprehensive Stroke Center." (SNIS executive committee, John Barr president, 2005). As we all know, there are very few hospitals that have 4 neurointerventionists.

There are probably fewer than 20 fellowships in the US that treat by endovascular techniques more than 50 strokes per year. Maybe 10. Even with 50 cases, a "fellow" can only "maybe" attain the SNIS mandated 10 cases as "primary operator" in his last few months. Thus, in the remaining 60-70 fellowships (90%?), the trainees will not get the required experience. There are numerous examples of fellowships that provide a certificate of competency for everything "neurointerventional" where the number of endovascular stroke cases can be counted on one hand... or one finger. How will these fellows get sufficient training, experience, and knowledge to have good outcomes when treating strokes when they are in private practice?

Foremost, I hope that this call for a moratorium in fellowships will be accompanied by an equally vigorous call for augmented and improved stroke training for fellows and our members. A recent SNIS webinar bemoaned the poor operator performance for intracranial stenting during SAMMPRIS. A recent article has documented "death and destruction" in endovascular stroke therapy (3). The failure of SAMMPRIS and the stoppage of IMS 3 only highlights the fact that we are not doing very well at this. As we all used to believe, IMS 3 should have been a slam-dunk. SNIS/SVIN should offer specialized intensive training courses that are, in actual fact, a responsibility of our professional organizations. Our societies need a plan for improved training and continuing education for our fellows and members in addition to a moratorium.

References

1) Fiorella D, Hirsch JA, Woo HH, et al. Should neurointerventional fellowship training be suspended indefinitely? J NeuroIntervent Surg 2012;4:315-318

2) Zaidat OO, Lazzaro M, McGinley E, et al. Demand-supply of neurointerventionalists for endovascular ischemic stroke therapy. Neurology 2012;79 (Suppl 1):S35-S41.

3) Cloft HJ. Death and Destruction in the Intra-Arterial Battle with Acute Ischemic Stroke. AJNR Am J Neuroradiol 32:1767-70.

Conflict of Interest:

None declared

Dear Sir:

I am not a neurosurgeon, neuroradiologist, nor interventionalist. In fact, I am not a regular reader of your journal. However, the article entitled "Should Neurointerventional Fellowship Training Be Suspended Indefinitely?" by David Fiorella, et al., was brought to my attention, unsolicited, by several different medical web sites. I have no objection to the logic and opinions expressed, and actually have no opinion on the issue for or against these fellowships. But I feel I must register my shock at what I see as a glaring omission by the journal.

The many authors do indicate their respective affiliations, and cumulatively list in their disclosure, "Competing interests: None." The most important disclosure, however, is not listed anywhere as a conflict of interest. Namely, how many of the authors are practicing, already fellowship-trained neurointerventionalists? In order to get a sense of any potential bias, I think the reader would want to know this fact about each author. Certainly, if most (or all) of the authors currently practice as neurointerventionalists, regardless of their arguments, some readers might feel that their position and recommendations simply represent an effort to restrict competition for their own advantage, and that the journal has given them a platform to advance this protectionist goal. If General Motors found themselves with a surplus of unsold cars and made the argument that there were too many autos already on the road, I might agree with that statement. But if their proposed solution was to ban further manufacture of cars so that they would have less competition unloading their inventories, I might object. Perhaps another company could produce cars that would be better (more efficient, safer, faster, etc.) and preventing them from doing so would hurt consumers, who would have fewer options. Similarly, while the authors may have correctly described the current problem (too many neurointerventionalists with not enough cases) their proposed solution (blocking further entries into the field) could be seen as self-serving. After all, maybe there are potential trainees or training programs who would introduce better practitioners (more efficient, safer, faster, etc.) into the market place, pressuring current (and future) neurointerventionalists to continue to improve the services they now deliver.

I invite each of the authors of the original article (and of any subsequent supportive letters) to make that disclosure: who among you are neurointerventionalists who have already enjoyed the benefits of fellowship training? Like any disclosure, this should not lessen your stated arguments; rather it will just allow the reader to know whence your recommendations are coming from. It only seems fair to be transparent. I'm just saying.

Conflict of Interest:

None declared

Dear Sir, we have read with great interest the paper by Deipolyi and co-workers on the use of desmopressin during inferior petrosal sinus sampling (IPSS) in alternative to corticotropin-releasing hormone (CRH) for the diagnostic work-up of Cushing's disease (1). As the Authors report, evidence collected in small series suggests that desmopressin achieves an equivalent diagnostic accuracy to CRH, thus appears a tempting and less expensive alternative. It is worth recalling, however, that desmopressin is also a potent haemostatic agent and that patients with Cushing's syndrome are at increased risk for thromboembolic events (2), some even occuring during IPSS performed at expert centers (3). In this context, we have recently observed that endothelial factors are released upon administration of desmopressin to patients with Cushing's disease (4), although the dose used for diagnostic purposes is roughly half the haemostatic dosage. The magnitude of the increase in von Willebrand factor, a promoter of platelet adhesion to the subendothelium in the early phases of haemostasis, was comparable to the one observed in healthy subjects and possibly counterbalanced by increased fibrinolytic factors but to what extent these changes might affect general haemostatic processes remains to be ascertained. As a consequence, as long as no further evidence is available on the safety of this haemostatic agent during endovascular damages, procedures which are per se at risk for vascular events, extreme caution should be exercised when using desmopressin instead of CRH during IPSS.

References 1. Deipolyi AR, Hirsch JA, Oklu R. Bilateral inferior petrosal sinus sampling with desmopressin. J.Neurointervent.Surg. 2012;doi10.1136/neurointsurg-2012-010437. 2. Stuijver DJF, van Zaane B, Feelders RA, Debeij J, Cannegieter SC, Hermus AR et al. Incidence of venous thromboembolism in patients with Cushing's syndrome: a multicenter cohort study. J.Clin.Endocrinol.Metab. 2011;96:3525-32. 3. Doppman JL. There is no simple answer to a rare complicaton of inferior petrosal sinus sampling. Am.J.Neuroradiology 1999;20:191-2. 4. Pecori Giraldi F, Ambrogio AG, Fatti LM, Rubini V, Cozzi G, Scacchi M et al. Von Willebrand factor and fibrinolytic parameters during the desmopressin test in patients with Cushing's disease. Br.J.Pharmacol. 2011;71:132-6.

Conflict of Interest:

None declared