Outcomes were mixed among ischemic stroke patients with large vessel occlusion who underwent thrombectomy by an interventional cardiologist as part of a multidisciplinary stroke team, in a single-center, prospective study from Poland.
Results from the 2-year experience show mechanical thrombectomy took longer when carried out by interventional cardiologists than by vascular surgeons and neuroradiologists (120 minutes vs. 105 minutes; P = .020).
The procedures were also less likely to achieve angiographic success, defined as a Thrombolysis in Cerebral Infarction (TICI) scale score of 2b or 3 (55.7% vs. 71.7%; P = .013), reported Krystian Wita, MD, PhD, Medical University of Silesia, Katowice, Poland, and colleagues.
The differences in duration and recanalization require further attention, they noted, and are related to a learning curve, more time dedicated to decision-making and, in some cases, needing a second opinion. Cardiologists performed 80 procedures compared with 116 for vascular surgeons and 52 for neuroradiologists, and treated twice as many patients with a previous stroke (13.9% vs. 6.5%).
Still, the interventional cardiologist- and noncardiologist-treated groups had similar functional independence at 3 months, defined by a modified Rankin Scale (mRS) score of 0 to 2 (44.4% vs. 54.8%; P = .275). Mortality was also similar at 3 months (31.3% vs. 28.0%; P = .595).
“This is the first analysis to prove the noninferiority of the cardiology services in the treatment of stroke with mechanical thrombectomy,” the authors reported in JACC: Cardiovascular Interventions.
But commenting for this news organization, J Mocco, MD, senior vice chair of neurosurgery and director of the Cerebrovascular Center at Mount Sinai Health System, New York, said this study isn’t designed as a noninferiority trial, is “grossly underpowered,” and the comparator cohort is not a gold standard comparator cohort.
“More importantly, they show that the cardiologists got significantly worse technical results and took longer, and we know that technical outcomes and the time to treatment are the two strongest predictors of outcome, which completely correlates with the fact that patients had 11% worse outcomes overall,” he said.
“It’s dumbfounding to me that this has been presented as evidence [that] an interventional cardiologist should be performing thrombectomy,” added Dr. Mocco, president-elect of the Society of NeuroInterventional Surgery.
Dr. Wita and coauthor Andrzej Kulach, MD, PhD, also with the Medical University of Silesia, told this news organization that timing is critical in mechanical thrombectomy (MT) and the sooner it’s performed, the better. But it cannot be performed by just any interventional cardiologist (IC).
“The IC must be trained in the procedure and cooperate with the neurologist to get good results,” they said. “We would like to stress that it is not a procedure for any cath lab and any cardiologist on duty. A network of cardiologists trained in MT must be organized and the stroke teams developed for the local unit to make the strategy reasonable and safe.”
The study was conducted from 2019 to 2020 and to participate, interventional cardiologists had to have performed a minimum of 700 angioplasties and 1,500 coronary angiographies and undergone complex training in thrombectomy, including 14-day training in a reference center and certified courses on a phantom and an animal model. They were also experienced in carotid angioplasty and participated as the second operators in neurointerventions.
“Considering the cardiologists are acting here in a multidisciplinary team led by neurologists, the findings are not surprising,” Dr. Wita and Dr. Kulach said. “What was surprising, is a certain level of skepticism among neurologists when cardiologists are to be involved in the procedure. We hope the quality of cardiology services will help to get over it.”
Major thrombectomy trials such as PRAGUE-16 have supported a role for interventional cardiologists to help meet demand for stroke thrombectomy. Dr. Wita and Dr. Kulach said there’s a lack of trained neuroradiologists and developed infrastructure for thrombectomy, whereas there’s a sufficient network of catheterization laboratories and trained cardiologists who could be involved.
The take-home message from the study, they said, is to “use the existing infrastructure to optimize the treatment of stroke. Building one from the very beginning is more time and resources-consuming.”
Dr. Mocco said a physician’s training is not a factor in the pathway to neurointerventional expertise, as long as they’re willing to put in the appropriate amount of specialization and training.
“There’s no way this represents a turf war or the neurology community somehow protecting its space, which is often used as a distraction, just like the idea that there’s not enough people,” he said. “It’s just not the case. Neurointervention is the most multispecialty space that I’m aware of.”
In the United States, at least, the problem isn’t a lack of resources or people to provide the service, but in getting patients to the correct hospitals, Dr. Mocco said. “We don’t have regionalized stroke care in the United States for the most part, so patients go to any hospital that says they provide stroke care rather than necessarily being triaged to capable centers that can provide the care.”
A 2021 Medicare analysis by Dr. Mocco and colleagues found that higher physician and hospital stroke thrombectomy volumes were associated with lower inpatient mortality and better outcomes.
Efforts are underway to regionalize care and delivery of patients in Los Angeles County and New York City, for example, where ambulances preferentially take patients with suspected large vessel occlusion to thrombectomy-capable stroke centers certified by independent organizations, Dr. Mocco said. In New York, “they’ve shown it has improved outcomes.”
Estêvão Carvalho de Campos Martins, MD, Hospital de Força Aérea do Galeão, Rio de Janeiro, and Fernando Luiz de Melo Bernardi, MD, Hospital Regional do Oeste, Chapecó, Brazil, noted in an accompanying editorial that the observational study is “hypothesis-generating only” and that the disconnect between technical and clinical outcomes is due to a type II error of low power.
They suggest that collaboration between specialties will be “essential for defining the optimal training program, so that ICs can reach solid procedural results.
“The accumulated experience with virtual simulation-based training for stroke could act as an educational accelerator but should be inserted in a prespecified program,” the editorialists said. “How to train and how to insert ICs into [an] MT interdisciplinary team is the current debate; meanwhile ICs are here, and many of them already contributing.”
Dr. Mocco is the principal investigator on research trials funded by Stryker Neurovascular, Microvention, and Penumbra; and is an investor in Cerebrotech, Imperative Care, Endostream, Viseon, BlinkTBI, Myra Medical, Serenity, Vastrax, NTI, RIST, Viz.ai , Synchron, Radical, and Truvic. He serves, or has recently served, as a consultant for: Cerebrotech, Viseon, Endostream, Vastrax, RIST, Synchron, Viz.ai , Perflow, and CVAid. Dr. Carvalho de Campos Martins and Dr. Luiz de Melo Bernardi have disclosed no relevant financial relationships.
A version of this article first appeared on Medscape.com.