Having tackled operator and institutional experience in carotid artery stenting at previous conferences, Sumaira MacDonald presented a scoring system developed to aid novices in selecting patients for the procedure at the CX Symposium on Monday 6 April 2009.
“How do we make carotid stenting safer? Well, of course there are a number of considerations and strategies, which include the use of dedicated equipment, appropriate pharmacological support, and judicious control of patient’s blood pressure. But I’d like to focus on case selection. I would argue that a learning curve is about more than the acquisition of technical ability. It is learning which patients to stent, and which to avoid stenting.”
Through the Delphi consensus method – in which panellists’ decisions are made in isolation and subsequently reviewed by an experienced statistician – the Carotid Stenting Delphi Consensus Panel sought to identify and evaluate all patient anatomic features considered to increase technical difficulty and therefore potentially increase the procedural stroke rate with a view to the generation of a scoring system intended to grade expected difficulty during stenting.
“We used private decision making, emailed questionnaires, so that we avoided the undue influence of the more charismatic, or perhaps more aggressive panellists. And we used explicit methods of combing judgments according to strict mathematical rules.”
The panelists were Alberto Cremonesi, Robert Fathi, Peter Gaines, Sumaira Macdonald, Claudio Schoenholz, Luc Stockx, Jos Van Den Berg, Jean-Pierre Becquemin, Marc Bosiers, Michel Makaroun, Jon Matsumura, and Peter Schneider: a group selected to incorporate representation from interventional cardiology, interventional radiology and vascular surgery.
Panellists were asked to propose individual anatomic criteria that were thought to be important considerations during carotid stenting, and these were subdivided by consensus according to anatomic level. Each criterion was marked on a scale of one to nine, representing the level of difficulty that the panellists thought each would present to an inexperienced operator.
It was considered that in real life patients were likely to have some combination of adverse anatomic features. The list of individual criteria was therefore reduced by consensus in order to limit the total number of subsequent combinations – as over 2,000 combination anatomies would have resulted from a full factorial design on all 12 original features, and this was considered impractical.
Tortuous common carotid artery was excluded from composite anatomy evaluations as it was considered the most difficult anatomic feature when encountered in isolation and it was thought, therefore, that it would be graded as very difficult in combination with other anatomic features. Low bifurcation was excluded as it scored as the least difficult of all the anatomies.
The panellists were then shown 96 slides and were asked to adjudge whether or not a novice should attempt stenting. The slides were scored according to the system previously derived, and a “cutting score” was arrived at by taking a mean of the difficulty score at which panellists changed from a “yes” to a “no” response.
“The literature reveals a steep technical learning curve for carotid stenting, and it is evident that structured training programmes and virtual reality simulation can go some way towards reducing this learning curve,” MacDonald told CX Daily News. “The Delphi consensus was aimed at those involved in carotid stenting programmes with less than 50 cases personal experience.”
“The final scoring system does have some inherent limitations,” she admitted. “Validation of the scoring system will require prospective or retrospective analysis of independently reviewed patient anatomy datasets.”
The intention is to formulate an electronic interactive version of the final results to aid case selection.