Time to Treatment with Intravenous Thrombolysis Before Thrombectomy and Functional Outcomes in Acute Ischemic Stroke
Johannes Kaesmacher,Fabiano Cavalcante,Manon Kappelhof,Kilian M. Treurniet,Leon Rinkel,Jianmin Liu,Bernard Yan,Wenjie Zi,Kazumi Kimura,Omer F. Eker,Yongwei Zhang,Eike I. Piechowiak,Wim van Zwam,Sheng Liu,Daniel Strbian,Maarten Uyttenboogaart,Tomas Dobrocky,Zhongrong Miao,Kentaro Suzuki,Lei Zhang,Robert van Oostenbrugge,Thomas R. Meinel,Changwei Guo,David Seiffge,Congguo Yin,Lukas Buetikofer,Hester Lingsma,Daan Nieboer,Pengfei Yang,Peter Mitchell,Charles Majoie,Urs Fischer,Yvo Roos,Jan Gralla
DOI: https://doi.org/10.1001/jama.2024.0589
2024-01-01
Abstract:Importance The benefit of intravenous thrombolysis (IVT) for acute ischemic stroke declines with longer time from symptom onset, but it is not known whether a similar time dependency exists for IVT followed by thrombectomy. Objective To determine whether the benefit associated with IVT plus thrombectomy vs thrombectomy alone decreases with treatment time from symptom onset. Design, Setting, and Participants Individual participant data meta-analysis from 6 randomized clinical trials comparing IVT plus thrombectomy vs thrombectomy alone. Enrollment was between January 2017 and July 2021 at 190 sites in 15 countries. All participants were eligible for IVT and thrombectomy and presented directly at thrombectomy-capable stroke centers (n = 2334). For this meta-analysis, only patients with an anterior circulation large-vessel occlusion were included (n = 2313). Exposure Interval from stroke symptom onset to expected administration of IVT and treatment with IVT plus thrombectomy vs thrombectomy alone. Main Outcomes and Measures The primary outcome analysis tested whether the association between the allocated treatment (IVT plus thrombectomy vs thrombectomy alone) and disability at 90 days (7-level modified Rankin Scale [mRS] score range, 0 [no symptoms] to 6 [death]; minimal clinically important difference for the rates of mRS scores of 0-2: 1.3%) varied with times from symptom onset to expected administration of IVT. Results In 2313 participants (1160 in IVT plus thrombectomy group vs 1153 in thrombectomy alone group; median age, 71 [IQR, 62 to 78] years; 44.3% were female), the median time from symptom onset to expected administration of IVT was 2 hours 28 minutes (IQR, 1 hour 46 minutes to 3 hours 17 minutes). There was a statistically significant interaction between the time from symptom onset to expected administration of IVT and the association of allocated treatment with functional outcomes (ratio of adjusted common odds ratio [OR] per 1-hour delay, 0.84 [95% CI, 0.72 to 0.97], P = .02 for interaction). The benefit of IVT plus thrombectomy decreased with longer times from symptom onset to expected administration of IVT (adjusted common OR for a 1-step mRS score shift toward improvement, 1.49 [95% CI, 1.13 to 1.96] at 1 hour, 1.25 [95% CI, 1.04 to 1.49] at 2 hours, and 1.04 [95% CI, 0.88 to 1.23] at 3 hours). For a mRS score of 0, 1, or 2, the predicted absolute risk difference was 9% (95% CI, 3% to 16%) at 1 hour, 5% (95% CI, 1% to 9%) at 2 hours, and 1% (95% CI, −3% to 5%) at 3 hours. After 2 hours 20 minutes, the benefit associated with IVT plus thrombectomy was not statistically significant and the point estimate crossed the null association at 3 hours 14 minutes. Conclusions and Relevance In patients presenting at thrombectomy-capable stroke centers, the benefit associated with IVT plus thrombectomy vs thrombectomy alone was time dependent and statistically significant only if the time from symptom onset to expected administration of IVT was short.