Predictors and Outcomes Associated with Rescue Therapy in SWIFT

Introduction: In the Solitaire With the Intention For Thrombectomy (SWIFT) trial, rescue therapy was used when the Solitaire or Merci device was unable to restore vessel patency. Markers for nonrecanalization in acute stroke have been reported for intravenous tissue plasminogen activator; however, similar predictors are not known for endovascular therapy. We sought to identify predictors and outcomes associated with rescue therapy in the SWIFT trial.

Methods: Rescue therapy included the use of an alternative device, agent, or maneuver following failure to recanalize with three retrieval attempts using the initial device. Clinical, angiographic, and demographic data was reviewed.

Results: Among a total of 144 patients enrolled, 43 (29.9%) required rescue therapy. We used the same baseline demographics for patients with and without rescue therapy. Rescue therapy was used in a higher percentage of patients randomized to the Merci group compared with the Solitaire group (43 vs. 21%, p = 0.009). Patients with rescue therapy experienced a longer recanalization time (p < 0.001), a lower percentage of successful recanalization (p < 0.001), and a lower percentage of good outcome (p = 0.009). In multivariate analysis, patients randomized to the Merci group (OR 3.99, 95% CI 1.58, 10.10) and age >80 years (OR 3.51, 95% CI 1.06, 11.64) were predictors of rescue therapy.

Conclusions: Merci treatment group and age were predictors of rescue therapy, while a trend toward an increased need of rescue therapy was observed with hypertension and proximal clot location. Rescue therapy was associated with fewer good outcomes. These findings may reflect targets for improvement in endovascular therapy.