LOS ANGELES – A postmarketing study of the Wingspan stent shows that the safety of the device in the treatment of intracranial atherosclerotic disease (ICAD) is good enough to be a reasonable alternative to medical management in these patients, but only if the device is used on label.
The results may reassure some interventionalists who were alarmed by results from the Stenting versus Aggressive Medical Management for Preventing Recurrent Stroke in Intracranial Stenosis (SAMMPRIS) trial (N Engl J Med. 2011;365:993-1003), which showed a 30-day rate of stroke or death of 14.7%. The new study showed a frequency of 2.6%, so long as the device was used on label. Off-label use yielded a frequency of 23.9%.
pixologicstudio/Thinkstock.com
Up to 10% of strokes in the United States result from ICAD, and in China the rate is an estimated 20%-46%. The condition can also be treated medically. Early trials of the Wingspan device showed initial success with complication rates of 4.5%-6.2%, but the SAMMPRIS trial, which directly compared stenting to aggressive medical management, showed superior outcomes with medical treatment. The 30-day rate of stroke or death of 14.7% was too high to compete with medical therapy, which included aspirin 325 mg per day, clopidogrel 75 mg per day for 90 days after enrollment, and management of primary and secondary risk factors.
Dr. Alexander believes that the SAMMPRIS trial did not employ favorable patient selection. “ICAD is variable. Some patients present with hemodynamic compromise, where stenting is probably beneficial. Some present with embolic stroke, and some present with small-vessel perforator strokes that are unlikely to be responsive to stenting and better treated with medical therapy. All these patients were grouped together” in SAMMPRIS, said Dr. Alexander, who is director of the Neurovascular Center and endovascular surgery at Cedars-Sinai in Los Angeles.
The SAMMPRIS findings put a damper on stenting, and use of the procedure has dropped at many U.S. hospitals. But studies conducted prior to SAMMPRIS had shown much lower periprocedural morbidity, and those studies looked at on-label use of stenting. SAMMPRIS was off label.
Now the WEAVE study, which was an Food and Drug Administration–mandated, postmarketing surveillance study of Stryker’s Wingspan stent, suggests that the off-label use of the system in the SAMMPRIS trial may explain the poor results. SAMMPRIS had attempted to extend the approved boundaries of stenting by treating patients who presented with transient ischemic attacks only, patients who had failed medical therapy, and patients who had experienced a stroke in the past 7 days. In fact, about half of the patients were treated within 7 days of the previous event, sometimes within 24 hours.
Previous studies had shown that risk factors for poor outcomes included stenting within 10 days of the last event, stenting a posterior circulation target lesion, stenting presentations other than stroke, and sites with a low patient volume for stenting. Patient selection is vital to success, according to Dr. Alexander. Patients with hemodynamic compromise are good candidates for stenting, while those with perforator stroke alone are better off with medical therapy. Embolic stroke patients are candidates for either approach.
WEAVE looked at 152 consecutive patients treated on label at 24 institutions. The primary analysis group consisted of patients with a 70%-99% stenotic intracranial atherosclerotic lesion who were refractory to medical treatment, 22-80 years of age, and had a modified Rankin Scale (mRS) score of 3 or less at baseline. The treatment was performed at least 7 days after the last stroke. Finally, patients had to have experienced two or more strokes. This last requirement presented a problem for recruitment, according to Dr. Alexander. “That was never a criterion for any of the [previous] trials, so it’s not clear why FDA added that. That made it very difficult to enroll for this trial – to have patients who had two or more strokes and still had a functional mRS score,” he said.
The study protocol aimed for a frequency of 6.6% for periprocedural morbidity, defined as a stroke or death within 72 hours.
An interim analysis at 100 patients showed that the periprocedural morbidity frequency was below 4%, which met the agency’s requirement and allowed the trial to be halted once the trial enrolled 150 on-label patients. The total number of on-label patients reached 152, and the researchers analyzed the results from another 46 patients who were treated with stenting despite not meeting the study’s inclusion criteria, and these patients were considered to be off-label use. The final analysis showed that the on-label group had a periprocedural morbidity of 2.6%, compared with 23.9% in the off-label group (P = .0001).
The most glaring difference in the patient populations was that half of the off-label group received the stent within 7 days of experiencing a stroke. What might be the reason for worse outcomes when stenting is performed within 7 days? “There’s speculation that the plaques might be hot, and those patients might have a higher thrombotic risk with putting a foreign body in the vessel, or there’s capillary instability, so reperfusing a vessel that has a 99% stenosis has a higher risk for reperfusion hemorrhage,” Dr. Alexander said.
Experience may also be a factor. Interventionalists participating in the WEAVE trial had performed a stent using Wingspan an average of 37 times before the study began, compared with a mean of 10 cases for physicians in the SAMMPRIS trial. Those who had performed over 50 procedures had no periprocedural morbidity outcomes at all.
The study was funded by Stryker Neurovascular. Dr. Alexander has consulted for Stryker.
SOURCE: Alexander M et al. ISC 2018 Abstract LB13.