Mohammad Anadani, MD
Since the publication of the pivotal trials in 2015, mechanical thrombectomy (MT) has become the standard of care for patients with large vessel occlusions. Intravenous thrombolysis (IVT) before MT (bridging therapy) remains the standard of care in eligible patients; however, there is an ongoing debate on the benefit of IVT prior to MT, especially that IVT administration can delay MT, which could be harmful.
In this study, PREDICT-RECANAL investigators conducted a retrospective multicenter study to: 1) determine the incidence and predictors of early recanalization (ER); and 2) develop and validate an ER prediction score.
The derivation cohort included patients from 4 MT-capable centers. Inclusion criteria were: 1) acute stroke due to large vessel occlusion (LVO) of the anterior circulation; 2) patients received IVT; and 3) ER evaluation before MT. The validation cohort consisted of 4 prospectively collected MT databases. MRI was used as first line imaging in this study, and CT/CT angiography was used as a second line in patients with contraindication to MRI. MRIs were reviewed to collect the following variables: 1) occlusion site; 2) length of the susceptibility vessel sign (SVS); and 3) DWI lesion extent using the DWI-ASPECTS. ER was evaluated <=3 hours after initiation of IVT. ER was defined as mTICI (modified Thrombolysis In Cerebral Infarction) 2b-3 for ICA T/L or M1 occlusion and Arterial Occlusive Lesion scale (AOL) 3 for the remaining occlusion sites.
Six hundred and thirty-three patients and 474 patients were included for the present study in the derivation and validation cohorts, respectively. ER occurred in 19.6 % and 17.9% of patients in the derivation and validation cohorts, respectively (P=0.49). ER occurred in the drip-and-ship more frequently when compared to the mothership patients in both cohorts (derivation cohort: 25.9% versus 10.4%; validation cohort: 22.0% versus 12.0%). ER rate was lower with proximal occlusions than distal occlusions.
Univariate analysis identified mothership paradigm, higher baseline NIHSS, shorter IVT-to-ER evaluation (EReval) time, proximal occlusions, and longer SVS as predictors of no-ER. In multivariate analysis, SVS length, occlusion site, and mothership paradigm emerged as predictors of no-ER. ER rate decreased with increasing SVS length ,and with higher baseline NIHSS, plateauing for NIHSS³16 (Figure 2). Conversely, ER increased with IVT-to-EReval, plateauing for IVT-to EReval time > 90 minutes (Figure 2).
Based on univariate and multivariate analysis, the authors created two scores: FIRE6 (For Intravenous thrombolysis REsistance) score, which included SVS characteristics and FIRE4, which did not include SVS characteristics (Table 3).
FIRE6 and FIRE4 scores were validated in the validation cohort with acceptable C-statistics (0.888[ 95%CI, 0.848-0.928] and 0.752 [95% CI, 0.704–0.800], respectively). Both scores performed well in predicting no-ER with high scores. On the contrary, lower scores for both scores performed poorly in predicting no-ER (Figure 3). Therefore, the authors concluded that neither score can be used to support decisions to withhold referral for MT in clinical practice.
The two key findings of this study are: 1) early recanalization occurs in almost 20% of patients with LVO who received IVT prior to MT; and 2) longer clot, and occlusion site are important predictors of ER.
This study supports the current clinical practice which consist of bridging therapy in patients eligible for IVT. The main limitation of this study is that MRI was the first line imaging study, which limits the generalizability of this study.