EVT indicates endovascular treatment.
mRS scores ranged from 0 to 6, with 0 indicating no symptoms and 6 indicating death. There were no patients with mRS 5. No difference in median (interquartile range) scores was found between the intervention group and the control group (1 [0-2] vs 1 [0-2]; crude common odds ratio, 0.90 [95% CI, 0.38-2.14]).
eTable 1. Details of Endovascular Procedure
eTable 2. Baseline Characteristics Stratified According to Prognostic Variables
eTable 3. Primary Outcome Stratified According to Baseline Prognostic Variables
eTable 4. Experience of Neurointerventionalists Prior to Start of the TO-ACT Trial per Participating Hospital
eTable 5. Screening Log
eFigure 1. Subgroup Analyses
eFigure 2. Recanalization Grades
eFigure 3. Comparison of Functional Outcome Between Patients Included vs Excluded From TO-ACT
Data Sharing Statement
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Coutinho JM, Zuurbier SM, Bousser M, et al. Effect of Endovascular Treatment With Medical Management vs Standard Care on Severe Cerebral Venous Thrombosis: The TO-ACT Randomized Clinical Trial. JAMA Neurol. 2020;77(8):966–973. doi:10.1001/jamaneurol.2020.1022
Does endovascular treatment with guideline-based standard medical care improve the functional outcome of patients with severe cerebral venous thrombosis?
In this multicenter, open-label, blinded end point randomized clinical trial involving 67 patients with severe cerebral venous thrombosis, no difference in the degree of disability at 12 months was found between patients who underwent endovascular treatment with standard medical care and those who received standard medical care alone.
Findings of this study suggest that endovascular treatment may not improve the functional outcome of patients with cerebral venous thrombosis.
To date, only uncontrolled studies have evaluated the efficacy and safety of endovascular treatment (EVT) in patients with cerebral venous thrombosis (CVT), leading to the lack of recommendations on EVT for CVT.
To evaluate the efficacy and safety of EVT in patients with a severe form of CVT.
Design, Setting, and Participants
TO-ACT (Thrombolysis or Anticoagulation for Cerebral Venous Thrombosis) was a multicenter, open-label, blinded end point, randomized clinical trial conducted in 8 hospitals in 3 countries (the Netherlands, China, and Portugal). Patients were recruited from September 2011 to October 2016, and follow-up began in March 2012 and was completed in December 2017. Adult patients with radiologically confirmed CVT who had at least 1 risk factor for a poor outcome (mental status disorder, coma state, intracerebral hemorrhage, or thrombosis of the deep venous system) were included. Data were analyzed according to the intention-to-treat principle from March 2018 to February 2019. The trial was halted after the first interim analysis for reasons of futility.
Patients were randomized to receive either EVT with standard medical care (intervention group) or guideline-based standard medical care only (control group). The EVT consisted of mechanical thrombectomy, local intrasinus application of alteplase or urokinase, or a combination of both strategies. Patients in the intervention group underwent EVT as soon as possible but no later than 24 hours after randomization.
Main Outcomes and Measures
Primary end point was the proportion of patients with a good outcome at 12 months (recovered without a disability; modified Rankin Scale [mRS] score of 0-1). Secondary end points were the proportion of patients with an mRS score of 0 to 1 at 6 months and an mRS score of 0 to 2 at 6 and 12 months, outcome on the mRS across the ordinal continuum at 12 months, recanalization rate, and surgical interventions in relation to CVT. Safety end points included symptomatic intracranial hemorrhage.
Of the 67 patients enrolled and randomized, 33 (49%) were randomized to the intervention group and 34 (51%) were randomized to the control group. Patients in the intervention group vs those in the control group were slightly older (median [interquartile range (IQR)] age, 43 [33-50] years vs 38 [23-48] years) and comprised fewer women (23 women [70%] vs 27 women [79%]). The median (IQR) baseline National Institutes of Health Stroke Scale score was 12 (7-20) in the EVT group and 12 (5-20) in the standard care group. At the 12-month follow-up, 22 intervention patients (67%) had an mRS score of 0 to 1 compared with 23 control patients (68%) (relative risk ratio, 0.99; 95% CI, 0.71-1.38). Mortality was not statistically significantly higher in the EVT group (12% [n = 4] vs 3% [n = 1]; P = .20). The frequency of symptomatic intracerebral hemorrhage was not statistically significantly lower in the intervention group (3% [n = 1] vs 9% [n = 3]; P = .61).
Conclusions and Relevance
The TO-ACT trial showed that EVT with standard medical care did not appear to improve functional outcome of patients with CVT. Given the small sample size, the possibility exists that future studies will demonstrate better recovery rates after EVT for this patient population.
ClinicalTrials.gov Identifier: NCT01204333
Cerebral venous thrombosis (CVT) is a distinct cause of stroke that primarily affects young adult and middle-aged patients.1,2 Approximately half of patients with CVT have an intracerebral hemorrhage (ICH) at presentation.3 All major guidelines recommend anticoagulation with heparin as the standard treatment for CVT4,5 regardless of the presence of an ICH. Despite heparin treatment, however, approximately 20% of patients with CVT retain their disability or die.3 Baseline variables associated with an increased risk of poor outcome include mental status disorder, coma state, ICH, and thrombosis of the deep venous system.3
Endovascular treatment (EVT) is increasingly being used to treat patients with CVT.6-9 Heparin treatment predominantly prevents growth or embolization of the existing thrombus, whereas EVT aims to achieve rapid recanalization of the sinuses through local application of a thrombolytic drug, mechanical thrombectomy, or a combination of both. Thus far, only uncontrolled studies have evaluated the efficacy and safety of EVT in patients with CVT.10 Because of the lack of prospective studies, the European Stroke Organisation guideline does not provide a recommendation on EVT for CVT,4 whereas the American Heart Association guideline recommends considering EVT only in patients who deteriorate despite anticoagulant treatment.5 The TO-ACT (Thrombolysis or Anticoagulation for Cerebral Venous Thrombosis) trial was designed to test the hypothesis that EVT in addition to standard medical care improves the clinical outcome of patients with severe CVT compared with standard medical care alone.
The TO-ACT randomized clinical trial was an investigator-initiated, multicenter, open-label trial with blinded end point evaluation (Prospective Randomized Open-label Blinded Endpoint [PROBE] design). It was conducted in 8 hospitals in 3 countries (the Netherlands, China, and Portugal), with patient recruitment taking place from September 2011 to October 2016 and follow-up beginning in March 2012 and ending in December 2017. An article describing the trial protocol was published elsewhere,11 and the trial protocol (Supplement 1) was approved by the institutional review board of the Amsterdam University Medical Centers and by each participating hospital (eAppendix in Supplement 2). Written informed consent was obtained from all patients or their legal representatives. The trial was monitored in the Netherlands by the Clinical Research Unit of the Amsterdam University Medical Center, in China by George Clinical International, and in Portugal by Eurotrials. We followed the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline.
We included adult patients (aged ≥18 years) who had radiologically confirmed CVT with a high probability of poor outcome, defined as the presence of at least 1 of the following risk factors: mental status disorder, coma state (Glasgow Coma Scale score <9; score range: 3-15, with the highest score indicating normal consciousness), ICH, or thrombosis of the deep cerebral venous system. Major exclusion criteria were as follows: duration from diagnosis to randomization of more than 10 days; pregnancy (women in the puerperium were eligible); thrombocytopenia (platelet count, <100 × 109/L); major surgical procedure (excluding lumbar puncture) in the past 2 weeks; clinical and radiological signs of impending transtentorial herniation from large, space-occupying lesions; and associated condition with a poor short-term prognosis independent of CVT. The full list of exclusion criteria can be found in the trial protocol (Supplement 1).
We did not keep a prospective screening log for the trial. However, after database closure, all sites submitted retrospectively collected data on the number of patients with CVT who were admitted to their hospital during the study period as well as data on the number and clinical outcome of patients who underwent EVT outside of the trial.
The first interim analysis was performed in November 2016, when the first 55 patients had completed the 12-month follow-up visit. Based on this interim analysis, the Data and Safety Monitoring Board recommended the termination of the TO-ACT trial because the predefined boundary for futility had been passed (conditional power of 11%). Even if an intermediate effect was anticipated in the remainder of the trial, the conditional power still remained below 20%. The TO-ACT steering committee unanimously voted to adopt the Data and Safety Monitoring Board recommendation and halted patient recruitment on November 30, 2016.
Participants were randomized 1:1 to receive either EVT with guideline-based standard medical care (intervention group) or guideline-based standard medical care only (control group) (Figure 1).
Randomization was accomplished through a web-based randomization system (ALEA Clinical Trial Data Management System; NKIAVL) using permuted blocks and stratified techniques for the presence of ICH and coma (Glasgow Coma Scale score <9). Patients, local investigators, and treating physicians were not masked to treatment randomization. Assessors of the functional outcome (modified Rankin Scale [mRS] score at 6 and 12 months; mRS score range: 0-6, with 0 indicating no symptoms and 6 indicating death) were blinded to treatment randomization. Two radiologists among us (S.D.R. and B.J.E.) scored the recanalization rate and were also masked to randomization.
Patients randomized to the control group received heparin in a therapeutic dose according to international guidelines. Either unfractionated or low-molecular-weight heparin was administered according to local preference.
Patients randomized to the intervention group underwent EVT as soon as possible but no later than 24 hours after randomization. The intervention consisted of mechanical thrombectomy, pharmacological thrombolysis, or a combination of both. The interventionalist made the decision regarding the exact EVT strategy to use. Access to the venous system was through the jugular or femoral vein. Either alteplase or urokinase was allowed for thrombolysis. The choice and dose of the thrombolytic drug was at the discretion of the interventionalist, but the maximum doses were specified in the trial protocol. Continuous infusion with a thrombolytic drug in the affected sinus was allowed after the initial procedure for a maximum of 72 hours. For mechanical thrombectomy, the use of standard devices for clot disruption and removal, such as microcatheters, balloon angioplasty devices, rheolytic catheters (AngioJet; Boston Scientific Corporation), and stent retrievers, was allowed at the discretion of the interventionalist.
All patients randomized to the intervention group had to commence (or continue) heparin in a therapeutic dose after EVT was completed. Heparin therapy during EVT was allowed but was not mandatory. After the acute phase, all patients received vitamin K antagonists for a minimum of 3 months if a transient risk factor was present or for 6 to 12 months in all other cases. The exact duration of long-term anticoagulation was left to the discretion of the local investigator.
Clinical evaluation was performed at baseline, discharge, and 6 and 12 months after randomization. The severity of neurological deficits at baseline was scored using the National Institutes of Health Stroke Scale (score range: 0-42, with higher scores indicating more severe deficit). Functional outcome was scored with the mRS at 6 and 12 months by a physician or research nurse who was not involved in the care of the patient. Following the advice of the Data and Safety Monitoring Board, we added in October 2013 an evaluation at 1 month after randomization. Follow-up magnetic resonance imaging with magnetic resonance venography was performed after 6 to 12 months.
The primary end point was the proportion of patients who recovered without a disability (mRS score 0-1) 12 months after randomization. Secondary end points were the proportion of patients with an mRS score of 0 to 1 at 6 months and an mRS score of 0 to 2 at 6 and 12 months, outcome on the mRS across the ordinal continuum at 12 months, recanalization rate, and surgical interventions in relation to CVT (ventricular shunting procedures or craniotomy). Safety end points were major hemorrhagic complications and symptomatic ICH within 1 week of randomization, all-cause mortality at 6 and 12 months, and all serious adverse events during the follow-up period. Symptomatic ICH was defined as any ICH that was associated with an increase of 4 points or more on the National Institutes of Health Stroke Scale or that resulted in death. Major hemorrhagic complication was defined as clinically overt bleeding associated with a decrease in hemoglobin level of 1.9 g/dL or more (to convert hemoglobin from g/dL to g/L, multiply by 10.0) if it was located retroperitoneal, intracranial, or intraocular; if it required a blood transfusion of 2 or more units; if it required an operation; or if it directly led to the death of the patient.12 Recanalization was determined for each sinus and was scored as complete (uninterrupted blood flow within the venous system disregarding some small residual thrombi adherent to the sinus wall), partial (more extensive thrombi with small interruptions of continuous blood flow or narrowing of the lumen), or absent (no recanalization, interrupted blood flow) (eFigure 2 in Supplement 2).
The TO-ACT trial was powered to detect an absolute difference of 20% (from 60% to 80%) in the proportion of patients with an mRS score of 0 to 1 in favor of EVT. Using a 2-group χ2 test with an α = .05 2-sided significance level and a power of 80%, we calculated a sample size of 164 patients (82 patients in each group). Two interim analyses were planned after one-third (n = 55) and two-thirds (n = 110) of patients had completed their 12-month follow-up visit. The interim analyses were performed by an independent Data and Safety Monitoring Board assessor, which monitored safety, efficacy, and futility. The stopping rule for futility applied if the conditional power (or probability of observing a statistically significant result in favor of the intervention group given the data obtained so far) decreased below 20%. The conditional power was calculated under the assumption that, in the remaining one-third or two-thirds of the study population, the distributions of the primary end point would be the same as those observed in the interim analysis. Statistical analyses were based on the intention-to-treat principle. In case of missing data for the primary end point, we imputed the mRS score using the last observation carried forward method.
Baseline assessments, details of the endovascular procedure, and outcome parameters were summarized with descriptive statistics. The effect of treatment on the primary and secondary end points was expressed as relative risk (RR) ratios. The precision of effect estimates was described with 95% CIs. We used ordinal logistic regression to analyze the full range of mRS scores at 12 months (secondary end point), and we expressed the effect size in a crude common odds ratio (OR). Differences between proportions were analyzed using the Fisher exact test, and differences between medians were analyzed with the Mann-Whitney test. We performed preplanned subgroup analyses based on ICH (present vs absent), coma state (Glasgow Coma Scale score <9 vs 9-15), and thrombosis of the deep venous system (present vs absent) using logistic regression models to examine the interaction effects between the subgroups and treatment.
To determine recanalization, we calculated the sum score of the total number of sinuses with complete recanalization. One point was given for complete recanalization of each of the following segments: superior sagittal sinus, straight sinus, left transverse sinus, right transverse sinus, left sigmoid sinus, right sigmoid sinus, left jugular vein, and right jugular vein. The proportion of patients with complete recanalization of the superior sagittal sinus and straight sinus was also identified separately.
A 2-sided P < .05 was considered statistically significant. All analyses were performed from March 2018 to February 2019 using IBM SPSS Statistics, version 24 (IBM Corp).
A total of 67 patients were enrolled and randomized. Of these patients, 33 (49%) were randomized to EVT with standard medical care (intervention group) and 34 (51%) were randomized to standard medical care alone (control group) (Figure 1). One patient in the intervention group did not receive EVT because she improved clinically shortly after randomization. Two patients in the control group underwent EVT because of clinical deterioration, with 1 deteriorating at day 0 and the other at day 4 after randomization. The levels of experience of the interventionalists who performed the endovascular procedures are listed in eTable 4 in Supplement 2. In 1 patient in the control group, the diagnosis of CVT was retracted 2 days after randomization. Six-month follow-up data were available for 67 patients (100%), and 12-month data were available for 66 patients (99%). All 67 patients were included in the final analysis. In total, 420 patients with CVT were admitted to the 8 participating hospitals during the study period (eTable 5 and eFigure 3 in Supplement 2). Therefore, approximately 16% of patients (67 of 420) were enrolled in the TO-ACT trial. Twenty-six patients underwent EVT outside of the TO-ACT trial, mostly in 2 of the participating hospitals.
Patients in the intervention group vs those in the control group were slightly older (median [interquartile range (IQR)] age, 43 [33-50] years vs 38 [23-48] years) and comprised fewer women (23 women [70%] vs 27 women [79%]) (Table 1). Median (IQR) National Institute of Health Stroke Scale score at baseline was 12 (7-20) in the intervention group and 12 (5-20) in the control group, and most patients in each group had an ICH at baseline (22 [67%] vs 25 [74%]). In both groups, 79% of patients (n = 53 of 67) received heparin at the time of randomization.
Among patients in the intervention group (n = 33), the median (IQR) time between randomization and performance of EVT was 278 (105-724) minutes (eTable 1 in Supplement 2). The median (IQR) duration of the procedure was 117 (93-157) minutes. Access to the venous system was through the jugular vein in 23 patients (72%), the femoral vein in 8 patients (25%), or both in 1 patient (3%). Mechanical thrombectomy was performed in 30 patients (91%), and chemical thrombolysis was performed in 17 patients (52%). The most frequently used thrombectomy devices were the AngioJet (n = 14) and stent retriever (n = 5). Of the 17 patients treated with chemical thrombolysis, 11 (33%) received urokinase and 6 (18%) received alteplase. Three patients (9%) received continuous infusion of urokinase or alteplase after EVT for a median (range) duration of 7 (1-24) hours.
No differences were found in the proportion of patients who received oral anticoagulation between the intervention and control groups (85% [n = 28 of 33] vs 85% [n = 29 of 34]) nor in the duration of oral anticoagulation use (mean, 9 months for both groups).
After 12 months of follow-up, 22 patients (67%) randomized to the intervention group had an mRS score of 0 to 1 vs 23 patients (68%) randomized to the control group (RR ratio, 0.99; 95% CI, 0.71-1.38) (Table 2). No statistically significant shift was found across the median (IQR) mRS scores at 12 months (1 [0-2] vs 1 [0-2]; crude common OR, 0.90 [95% CI, 0.38-2.14]) (Table 2 and Figure 2). The number of patients with an mRS score of 0 to 2 at 12 months was 28 (85%) in the intervention group and 28 (82%) in the control group (RR ratio, 1.03; 95% CI, 0.83-1.27). Complete recanalization of the superior sagittal sinus was more frequent among patients who underwent EVT than those who received standard care alone (22 [79%] vs 15 [52%]; RR ratio, 1.52 [95% CI, 1.02-2.27]) (Table 2), but the recanalization sum score did not differ significantly between the 2 groups (7 vs 6; P = .13). No statistically significant differences were found in any of the other secondary end points.
No differential treatment effects were found across the 3 subgroups of ICH, coma state, and thrombosis of the deep venous system (eFigure 1 in Supplement 2). Baseline characteristics and the primary end point stratified by the presence of each of the 4 baseline prognostic variables (ICH, coma state, thrombosis of the deep venous system, and mental status disorder) are shown in eTables 2 and 3 in Supplement 2. The rate of mRS score of 0 to 1 in the intervention group did not differ between patients randomized in the second half of the study (n = 11 of 16 [69%]) vs the first half of the study (n = 11 of 17 [65%]).
The mortality rate at 6 months (12% [n = 4] vs 3% [n = 1]; P = .20) and 12 months (12% [n = 4] vs 3% [n = 1]; P = .20) was higher in the intervention than control group, but this difference was not statistically significant (Table 3). Four of five deaths occurred within 10 days of randomization owing to progressive brain edema and transtentorial herniation. One of these 4 patients underwent decompressive hemicraniectomy. The fifth fatality occurred in the EVT group 129 days after randomization owing to unknown causes. New symptomatic ICH occurred less frequently in the EVT group than in the standard care group (3% [n = 1] vs 9% [n = 3]; P = .61). Perforation of a sinus or vein during EVT was observed in 3 patients (9%). The procedure was aborted in each case, and no patient suffered a symptomatic ICH from perforation. Seizures after randomization occurred less often in patients randomized to undergo EVT than those in the control group (1 [3%] vs 10 [30%]; P = .006). No significant difference was found in the frequency of status epilepticus between the 2 groups (3% [n = 1] vs 6% [n = 2]; P > .99). No (recurrent) thrombotic or major bleeding events occurred after hospital admission in any of the patients.
To our knowledge, the TO-ACT trial is the first randomized clinical trial on the efficacy and safety of EVT in patients with CVT and at high risk of bad outcome. The study was prematurely terminated because of futility. We found that EVT with medical management did not improve functional outcome in patients. The mortality rate was higher in patients randomized to receive EVT, whereas the frequency of new symptomatic ICH, the most feared complication of EVT, was lower in the intervention group than in the control group, but both differences were not statistically significant.
The neutral results of the TO-ACT trial should not be interpreted as definitive proof that EVT is ineffective in CVT. Although the point estimate of the primary end point in the trial did not point toward a beneficial effect of EVT in CVT (RR ratio, 0.99), the large width of the 95% CI (0.71-1.38) indicated that a clinically meaningful treatment effect cannot be excluded. Similarly, we cannot exclude the possibility that EVT was effective in a subgroup of patients (for instance, to those in a coma) because the number of patients in such groups was small.
We included patients with severe CVT who were at risk of an unfavorable outcome. Based on cohort studies with long-term follow-up, we expected approximately 60% in this population to have a good outcome (mRS score of 0-1). The observed proportion of good outcomes in the control group (68% at 12 months) was slightly better than this expectation.3 In ischemic stroke, an mRS score of 0 to 2 is often considered to be a favorable outcome. The reason we chose an mRS score of 0 to 1 as a primary end point was that, in general, patients with CVT are much younger than patients who had ischemic stroke, and we believed that recovery without disability was more appropriate as a definition of good outcome. The finding that 82% of patients in the control group achieved an mRS score of 0 to 2 at follow-up visits underscores that, overall, the prognosis after CVT is much better than after ischemic stroke. This high rate of good outcome with medical management also makes it difficult to show an additional benefit of EVT.
A diverse population of patients was included in the TO-ACT trial. For instance, some patients had a mental status disorder, whereas others were in a coma. Given that it is conceivable that the effect of EVT was not consistent across the entire study population, we planned a subgroup analysis. This analysis did not show a statistically significant interaction across the subgroups, but the small sample size precludes us from drawing reliable conclusions regarding subgroups. The not significantly better results of EVT in patients in a coma (eTable 3 in Supplement 2) may warrant further exploration in future studies. In addition, only a selection of patients who were in a coma were eligible for inclusion because a coma state that was attributable to impending transtentorial herniation was an exclusion criterion. This condition explains why the proportion of patients in a coma who achieved good outcomes was higher in the TO-ACT trial than in cohort studies.3
One explanation for the inability of the TO-ACT trial to demonstrate the efficacy of EVT could be associated with the technical aspects of the procedure. With the implementation of EVT for acute ischemic stroke, the number of endovascular procedures performed by neuro-interventionalists has increased in the past few years.13 Endovascular treatment of the venous system, however, differs substantially from that of the arterial system, and it is conceivable that during the inclusion period, available techniques and devices to achieve optimal recanalization in patients with CVT were suboptimal. Given that outcomes among patients randomized to receive EVT did not improve during the course of the trial, a learning effect did not seem to have occurred. Still, novel devices that allow faster and more effective thrombus removal from the cerebral venous system may be developed in the future. Such a development was seen in ischemic stroke, in which studies that used early-generation thrombectomy devices did not demonstrate a benefit.14-16 Future trials on EVT in CVT may need to be more restrictive in device selection and thus base this selection on data from properly designed animal studies, phantom models, and patient studies. Dynamic vascular imaging techniques, such as computed tomography perfusion or dynamic magnetic resonance angiography, may be useful to select patients for treatment, similar to their use in ischemic stroke. Moreover, the heterogeneity of the cerebral venous system and methods to identify the appropriate target occlusion that is safely accessible for EVT should be taken into account in these studies.
Perforations of the cerebral venous system after EVT are rarely reported in the literature,17 but in this study it occurred in 3 of 33 procedures. Although we cannot point to the exact location of the perforation, we believe it likely occurred from the inadvertent advancement of the guidewire into a cortical vein. The American Heart Association guideline on CVT listed perforation as a potential complication but provided no data on its frequency.5 Perforations are likely underreported because the literature on EVT for CVT consists solely of uncontrolled and retrospective case series.18 None of the patients in whom a perforation occurred developed symptomatic ICH after the procedure. A possible explanation for this finding may be that the perforations occurred at the beginning of the procedure when the venous system was still occluded.
The TO-ACT trial was designed as a pragmatic study, and the EVT strategy was largely at the discretion of the interventionalist.19 As a result, various devices were used for mechanical thrombectomy, and only half of the patients randomized to undergo EVT were treated with chemical thrombolysis. We chose not to enforce a strict guideline for the procedure because many of the strategies in the literature offered no clear evidence of their superiority over other approaches. We did not want to interfere with the preferences and experiences of individual radiologists.20 One study did find lower recanalization rates and a lower likelihood of a good outcome in patients treated with the AngioJet device, but this result was based on a comparison of retrospective studies and should be interpreted with caution.9
This trial has several limitations. First, although, to our knowledge, TO-ACT is one of the largest randomized clinical trials in CVT, it included only 67 patients and therefore was underpowered to detect a small difference between the 2 treatment groups. Second, we did not assess early recanalization rate. Magnetic resonance venography was performed after 6 to 12 months, but an assessment of recanalization immediately after EVT would have been useful for identifying the technical success rate of EVT. Overall, we found no difference in the late recanalization sum score, but patency of the superior sagittal sinus was higher in patients randomized to the intervention group. Third, we did not keep a log of patients who were screened for eligibility but were not included in the trial. To partly overcome this oversight, we established a screening log retrospectively after database closure, which indicated that approximately 16% of all patients with CVT who were admitted to the participating hospitals were included in the TO-ACT trial; this figure amounts to an estimated 50% of the potentially eligible cases (based on one-third of all cases meeting the inclusion criteria for severity). Fourth, because of the small sample size and limited number of outcome events, we could not adjust the primary analysis for prognostic variables. Fifth, because of the open-label design, physician choices regarding medical management after the acute phase could have influenced the outcome of patients. However, this scenario was not likely given that the frequency and duration of oral anticoagulation use were similar in both treatment groups and that no thrombotic or major bleeding events occurred after the acute phase.
In the TO-ACT trial, EVT along with medical care did not appear to be superior to medical care only in patients with a severe form of CVT. Because of the small sample size, we cannot exclude the possibility that future studies, using other methods of patient selection and endovascular techniques, may identify better recovery rates after EVT for patients with severe CVT.
Accepted for Publication: February 21, 2020.
Corresponding Author: Jonathan M. Coutinho, MD, PhD, Department of Neurology, Amsterdam University Medical Centers, Meibergdreef 9, PO Box 22660, 1105 AZ Amsterdam, the Netherlands (email@example.com).
Published Online: May 18, 2020. doi:10.1001/jamaneurol.2020.1022
Author Contributions: Dr Coutinho had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Drs Coutinho and Zuurbier contributed equally.
Concept and design: Coutinho, Bousser, Ji, Canhão, Houdart, de Haan, Ferro, Stam.
Acquisition, analysis, or interpretation of data: Coutinho, Zuurbier, Canhão, Roos, Crassard, Nunes, Uyttenboogaart, Chen, Emmer, Roosendaal, Reekers, van den Berg, Majoie, Ferro, Stam.
Drafting of the manuscript: Coutinho, Zuurbier, Ji, Ferro.
Critical revision of the manuscript for important intellectual content: Zuurbier, Bousser, Canhão, Roos, Crassard, Nunes, Uyttenboogaart, Chen, Emmer, Roosendaal, Houdart, Reekers, van den Berg, de Haan, Majoie, Ferro, Stam.
Statistical analysis: Coutinho, Zuurbier, Emmer.
Obtained funding: Ferro, Stam.
Administrative, technical, or material support: Zuurbier, Roos, Nunes, Chen, Roosendaal, Reekers, Ferro.
Supervision: Coutinho, Ji, Nunes, Chen, Houdart, Reekers, Ferro, Stam.
Other—Interpretation and acquisition of data, scoring of data: Emmer.
Conflict of Interest Disclosures: Drs Coutinho and Ferro reported being members of the steering committee of the RESPECT-CVT (A Clinical Trial Comparing Efficacy and Safety of Dabigatran Etexilate With Warfarin in Patients With Cerebral Venous and Dural Sinus Thrombosis) trial, a clinical trial sponsored by Boehringer Ingelheim. Dr Coutinho reported receiving grants from Dutch Heart Foundation during the conduct of the study and from Boehringer Ingelheim outside the submitted work. Dr Roos reported receiving research support from Nico-Lab outside the submitted work. Dr van den Berg reported receiving research support from Cerenovus outside the submitted work. Dr Majoie reported receiving grants from the Dutch Heart Foundation, European Commission, Twin Foundation, Dutch Health Evaluation Program, and Stryker outside the submitted work as well as being a shareholder of Nico.lab. Dr Ferro reported receiving personal fees from Boehringer Ingelheim outside the submitted work. No other disclosures were reported.
Funding/Support: This study was funded by grant 2009B016 from the Dutch Heart Foundation (Dr Stam).
Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
TO-ACT (Thrombolysis or Anticoagulation for Cerebral Venous Thrombosis) Investigators: Jan Stam, MD, PhD, Amsterdam University Medical Centers; Diederik W. Dippel, MD, PhD, Erasmus University Medical Center; Korne Jellema, MD, PhD, Medical Center Haaglanden; Maarten Uyttenboogaart, MD, PhD, University Medical Center Groningen; Patricia Canhão, MD, PhD, Hospital Santa Maria; Ana P. Nunes, MD, PhD, Hospital Sao Jose; Gabriela M. Lopes, MD, PhD, Hospital de Santo António; and Xunming Ji, MD, PhD, XuanWu Hospital.
Data Sharing Statement: See Supplement 3.
Additional Contributions: This trial was designed and led by a steering committee that is composed of independent academic investigators. N. E. LeCouffe, MD, and A. E. Groot, MD, Amsterdam University Medical Centers, assisted with statistical analyses. These individuals received no additional compensation, outside of their usual salary, for their contributions.