Article Text

Download PDFPDF
SNIS 9th annual meeting oral abstracts
O-018 Pooled analysis of roll-in and randomized patients in SWIFT confirms benefit of solitaire thrombectomy in patients with acute ischemic stroke
  1. J Saver1,
  2. R Jahan2,
  3. E Levy3,
  4. T Jovin4,
  5. B Baxter5,
  6. R Nogueira6,
  7. W Clark7,
  8. R Budzik8,
  9. O Zaidat9,
  10. For SWIFT Investigators2
  1. 1Neurology, UCLA Medical Center, Los Angeles, California, USA
  2. 2Division of Interventional Neuroradiology, UCLA Medical Center, Los Angeles, California, USA
  3. 3Neurosurgery, Millard Fillmore Hospital, Buffalo, New York, USA
  4. 4Neurology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
  5. 5Radiology, Erlanger Hospital, Chattanooga, Tennessee, USA
  6. 6Neurology, Emory University, Atlanta, Georgia, USA
  7. 7Neurology, Oregon Health Science University, Portland, Oregon, USA
  8. 8Interventional Neuroradiology, Riverside Methodist Hospital, Columbus, Ohio, USA
  9. 9Neurology, Medical College of Wisconsin, Milwaukee, Wisconsin, USA


Background and Objective The Solitaire With Intention For Thrombectomy (SWIFT) study demonstrated superiority of Solitaire FR device compared with the Merci Retrieval System®. Because the study stopped early due to overwhelming efficacy, number of patients in the randomized Solitaire FR arm was modest. The study included a roll-in phase where centers treated the first two patients with Solitaire FR prior to beginning randomization. The purpose of this presentation is to provide a comprehensive analysis of Solitaire FR outcomes, roll-in and randomized Solitaire FR patients inclusive, and to compare against outcome data from patients randomized to Merci®.

Methods Multicenter randomized trial with blinded primary endpoint analysis. Inclusion criteria: age 22–85; NIHSS 8–29; within 8 h of onset; ineligible or failed IV TPA; intracranial ICA, M1, M2, BA, or VA occlusion. Following treatment of the two roll-in patients with the Solitaire FR, sites randomized patients to initial neurothrombectomy with Solitaire FR device or Merci. The primary efficacy endpoint was core lab (blinded)-assessed TIMI 2 or 3 flow in all treatable vessels without symptomatic intracranial hemorrhage, after up to three passes of the assigned device. Rescue therapy with intra-arterial lytic or an FDA approved device was permitted after primary endpoint reperfusion ascertainment.

Results The study population, enrolled at 18 sites, comprised 31 roll-in phase Solitaire FR, 58 randomized Solitaire FR, 55 randomized Merci for a total of 89 Solitaire FR and 55 Merci treated patients. Baseline characteristics and outcomes in roll-in Solitaire FR patients were similar to randomized Solitaire FR patients, satisfying statistical criteria for homogeneity permitting combination for analysis. Patients in the combined Solitaire FR and the randomized Merci treatment groups were comparable in all demographic and medical history variables collected, but differed in history of atrial fibrillation, which was higher in the Merci arm (42.7% vs 67.3%, p=0.006). Procedure-related serious adverse events were similar in the two treatment groups. The primary efficacy outcome, successful recanalization without SICH, was achieved more often in Solitaire FR vs Merci patients (58.8% vs 24.1%), non-inferiority p<0.0001, superiority p=0.0001. Additional efficacy outcomes are shown in the Abstract O-018 table 1.

Abstract O-018 Table 1

Conclusions The Solitaire flow restoration device shows persistent superiority when comparing the entire population of patients (roll-in plus randomized) receiving Solitaire FR therapy to those randomized to the Merci device. Initial treatment with Solitaire FR rather than Merci is associated with more frequent reperfusion, less symptomatic intracranial hemorrhage, reduced mortality, and increased achievement of good final neurologic outcome 3 months after stroke.

Competing interests J Saver: Covidien. R Jahan: Covidien. E Levy: Covidien. T Jovin: Covidien. B Baxter: Covidien. R Nogueira: Covidien. W Clark: None. R Budzik: None. O Zaidat: None. For SWIFT Investigators: None.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.