SGEM#349: Can tPA Be A Bridge Over Trouble Waters to Mechanical Thrombectomy?

Date: November 1st, 2021 Reference: Katsanos et al. Utility of Intravenous Alteplase Prior to Endovascular Stroke Treatment: A Systematic Review and Meta-analysis of RCTs. Neurology 2021 Guest Skeptic: Dr. Michal Krawczyk is in his fifth year of neurology residency at Western University in London, Ontario, Canada. He is interested in acute neurological illness, including cerebrovascular disease and epilepsy. Next year he will be beginning a Neurohospitalist fellowship at the University of Texas at Houston. Case: A 70-year-old male with a past medical history of hypertension and peripheral artery disease, last seen normal 1.5 hours ago, presenting with acute onset of aphasia and right sided face and arm weakness. He has a National Institute of Health Stroke Scale (NIHSS) score of 7. At 1am a CT angiogram is obtained that demonstrated a left M2 occlusion, and an Alberta Stroke Program Early CT Score (ASPECTS) of 10. Given the recent publications of trials assessing if mechanical thrombectomy alone is non-inferior to a bridging approach with tPA in addition to mechanical thrombectomy, you wonder whether these trials apply to your patient and what is the best course of action. Background: There are two treatments for acute ischemic stroke, systemic tPA and mechanical thrombectomy (MT). We have covered some studies looking at both treatment modalities on the SGEM. SGEM#29: Stroke Me, Stroke Me SGEM#70: The Secret of NINDS (Thrombolysis for Acute Stroke) SGEM#85: Won’t Get Fooled Again (tPA for AIS) SGEM#137: A Foggy Day – Endovascular Treatment for Acute Ischemic Stroke SGEM#292: With or Without You – Endovascular Treatment with or without tPA for Large Vessel Occlusions SGEM#297: tPA Advocates Be Like – Never Gonna Give You Up SGEM#333: Do you Gotta Be Starting Something – Like tPA before EVT? Mechanical thrombectomy is indicated only for patients with large vessel occlusions (LVOs) on imaging. There were a few earlier studies on MT that failed to demonstrate superiority, but it was the study MR CLEAN published in NEJM 2015 that really changed practice. It was a multicenter, randomized, unblinded trial treating 500 patients with an anterior circulation LVO within six hours of symptom onset. The primary outcome was mRS 0-2 at 90 days and it showed an absolute difference of 14% favoring MT. This gives a NNT of 7. Six RCTs have been published since MR CLEAN.  All supported MT and all were stopped early (SWIFT PRIME, EXTEND-IA, REVASCAT, ESCAPE, DAWN, and DEFUSE). For patients with LVOs it is unclear whether there is any additional benefit with administering tPA before thrombectomy, also known as a bridging approach, in contrast to skipping tPA and directly proceeding with MT. There are several theoretical advantages of a bridging approach. These potential advantages include thrombus debulking allowing easier clot retrieval, distal emboli lysis, recanalization prior to MT, and it may be beneficial in cases of unsuccessful MT. Conversely, a direct to MT approach may lead to fewer intracerebral hemorrhages (ICH) and quicker initiation of endovascular thrombectomy. Recently, three randomized control non-inferior trials on this topic have been published, two from China (DIRECT-MT, and DEVT) and one from Japan (SKIP). Two trials demonstrated non-inferiority while one trial failed to show that direct MT was non-inferior. Clinical Question: What is the best strategy for treating patients with an acute large vessel occlusion stroke, direct to mechanical thrombectomy or a bridging approach with tPa followed by mechanical thrombectomy? Reference: Katsanos et al. Utility of Intravenous Alteplase Prior to Endovascular Stroke Treatment: A Systematic Review and Meta-analysis of RCTs. Neurology 2021 Population: Randomized controlled trials of patients with acute large vessel occlusion stroke qualifying for MT Exclusions: Observational studies and non-randomized trials Intervention: MT alone Comparison: MT bridged with tPA Outcome: Primary Outcome: mRS score 0-2 at three months Secondary Outcomes: mRS 0-1 and ordinal shift at three months, successful recanalization before MT, successful recanalization after MT, randomization to puncture time, symptomatic intracranial hemorrhage (sICH), any ICH and all-cause mortality Authors’ Conclusions: “We detected no differences in functional outcomes of IV thrombolysis–eligible patients with an acute LVO receiving dEVT compared to BT. Because uncertainty for most endpoints remainslarge and the available data are not able to exclude the possibility of overall benefit or harm, further RCTs are needed.” Quality Checklist for Therapeutic Systematic Reviews: The clinical question is sensible and answerable.  Yes The search for studies was detailed and exhaustive. Yes The primary studies were of high methodological quality. No The assessment of studies were reproducible. Yes The outcomes were clinically relevant. Yes There was low statistical heterogeneity for the primary outcomes. Yes The treatment effect was large enough and precise enough to be clinically significant. No Results: The three RCTs included a total of 1,092 patients. Median age was in the early 70’s and 42% were female. Key Results: No statistical difference in good neurologic outcome Primary Outcome: mRS score 0-2 at three months OR 1.08 (95% CI 0.85 to 1.38) and adjusted OR 1.11 (95% CI 0.76 to 1.63) Secondary Outcomes: mRS score 0-1 at three months OR 1.10 (95% CI 0.84 to 1.43) and adjusted OR 1.16 (95% CI 0.84 to 1.61) Successful recanalization before EVT: OR 0.37 (0.18-0.77) Moderate certainty Successful recanalization after EVT: OR 0.77 (0.54-1.08) Low certainty sICH: OR 0.75 (0.45-1.25) Low certainty Any ICH: OR 0.67 (0.49-0.92) Moderate certainty All-cause mortality: OR 0.93 (0.68-1.29) Low certainty 1. External Validity: All three trials were from Asia and as such may not be directly applicable to North American populations and healthcare systems. In one of the trials, they used 0.6mg/kg of tPA (SKIP) instead of the standard 0.9mg/kg. This could bias the trial to finding non-inferiority. In addition, these studies were all conducted at stroke centres with MT availability and do not address a drip and ship model of care. 2. Non-Inferiority Margins: All three studies included in the SRMA were non-inferiority trial designs. They were asking if direct to MT was non-inferior to the standard bridging with tPA before MT. Two out of three trials (DIRECT-MT and DEVT) the non-inferiority was met, but the non-inferiority margin was set at ≤10% absolute clinical effect in DEVT, and 20% effect size in odds ratio in DIRECT-MT. Even if non-inferiority is demonstrated, it does not mean there is no clinical benefit from a bridging approach if the non-inferiority margin is too large, which may represent a clinically important difference. Many argue that the non-inferiority claim should only be reserved when a less conservative margin of 5% is utilized. None of the trials met this less conservative margin. 3. Performance Bias: We have discussed different forms of bias many times on the SGEM. This is the first time we have mentioned performance bias. This type of bias is defined by Cochrane Risk of Bias (RoB) Tool as the result of “systematic differences between groups in the care that is provided, or in exposure to factors other than the interventions of interest.” As highlighted in this SRMA, there was a performance bias in the DIRECT-MT trial with 9.4% of patients in the bridging group not receiving MT, while only 5.2% in the direct group did not receive MT. This 4.2% difference may have resulted in worse outcomes in the bridging group, favoring direct MT and a finding of non-inferiority. 3. Selection Bias: This is a type of bias we have discussed many times on the SGEM. The Cochrane RoB Tool defines selection bias as the result of “systematic differences between baseline characteristics of the groups that are compared.” Selection bias may affect the estimate of the per-protocol effect and/or the intention-to-treat effect. It depends on the definition that is used for the groups that are being compared. In the DEVT trial, an exclusion criterion was “arterial tortuosity and/or other arterial disease that would prevent the device from reaching the target vessel.”  This exclusion criterion may effectively ‘cherry-pick’ patients, excluding those where thrombectomy would have been difficult, potentially resulting in less favorable outcome in the direct MT group. It is unclear how many patients were excluded from the DEVT trial for this reason. In the DIRECT-MT trial approximately 5.8% (38/654) of patients intended to undergo thrombectomy did not due to technical reasons, highlighting that even in specialized academic centers thrombectomy remains technically challenging. 4. Timing of tPA: In the SKIP trial, 21% of patients in the bridging group had tPA started after groin puncture for MT. It is likely that in a significant proportion of these patients MT was completed even before the tPA infusion was finished. In the DIRECT-MT trial 87% of patients had a tPA infusion ongoing during MT, and 9% of patients in the bridging group did not receive the full dose of tPA. This could have biased the study towards finding non-inferiority for MT alone. 5. Subgroups: Certain subgroups that may benefit more from a bridging approach were underrepresented in the three trials. In the study design of the DEVT and SKIP trials they did not include patients with M2 occlusions. After final adjudication the percentage of M2 occlusions in the DEVT trail was 1.7%, SKIP 19%, and DIRECT-MT 10.1%. It is known that compared to M1/ICA occlusions, tPA is much more effective at lysing M2 clots. In the INTERRSeCT study,