Background: The Treatment Of Brain Arteriovenous malformations Study is a pragmatic study comprising 2 randomized trials and prospective registries of patients with brain arteriovenous malformations (AVMs). We report the results of the stereotactic radiosurgery (SRS) registry. Methods: TOBAS uses an algorithm integrating clinical judgment and randomization. When SRS was selected as the preferred curative modality and randomized allocation of SRS versus observation was judged unreasonable, patients were included in the SRS registry (with/without pre-SRS embolization). The primary outcome was modified Rankin Scale > 2 at last follow-up. Secondary outcomes included serious adverse events, AVM rupture, radiation-induced changes, and magnetic resonance angiographic results at 3 years. Analyses were intent-to-treat. Results: From June 2014 to May 2021, 1010 patients were recruited; 105 (10.4%) were offered participation in the SRS registry. After exclusions, 94 patients were analyzed; most had ruptured (n = 64/94 [68%]), small AVMs (0-3 cm) (84 [89%]), within eloquent brain (57 [61%]). Pre-SRS embolization was performed in 19 (20%); 4 were cured and 3 had endovascular complications, leaving 87/94 (93%) patients who received SRS. The median SRS dose was 22 Gy (interquartile range: 20-24). During a mean follow-up of 51 months, the primary outcome (modified Rankin Scale > 2) occurred in 7/94 (7%; 95% confidence interval [CI]: 4%-15%), with serious adverse events in 16/94 (17%; 95% CI: 11%-26%), including AVM rupture in 5/94 (5%; 95% CI: 2%-12%), and symptomatic radiation-induced changes in 6/87 (7%; 95% CI: 3%-14%). Three-year imaging follow-up, available in 79/94 (84%) confirmed AVM occlusion in 51/79 (65%; 95% CI: 54%-74%) or 54% (95% CI: 44%-64%) of all patients. Conclusions: In this prospective registry, radiosurgery with or without embolization was relatively safe, but of moderate efficacy, with 3-year AVM occlusion in 54%-65% patients.

Background A widespread practice is to label trials as positive or negative based on whether the P value for the primary outcome is less than 0.05. A more reasonable approach is to examine the totality of evidence, particularly when results approach the threshold of statistical significance. Methods We review the CREST-2 study which included two parallel trials on ≥70% asymptomatic carotid stenoses: one trial (n = 1245) compared stenting with an intensive medical management group (IMM1); a second trial (n = 1240) compared endarterectomy with another IMM2 group. Results A significantly lower rate of the 4-year primary composite endpoint of stroke or death (2.8%) was reported with stenting compared to IMM1 (6.0%; ARR 3.2%; P = 0.02), while results of endarterectomy (3.7%) vs IMM2 (5.3%) were not significantly better (ARR 1.6%; P = 0.24). Closer examination shows that overall, only 104 infarctions occurred in 2485 patients (4.2%) over 4 years, and most (70 or 67%) were minor strokes. A ‘tipping point analysis’ revealed that if 3 more primary outcome events had occurred in the stenting group, significance would have been lost; if 6 more events had occurred in the IMM2 group, endarterectomy would have attained significance. Stenting resulted in as many major strokes (n = 4) as IMM1 (n = 5) over 4 years. Major strokes after endarterectomy (n = 3) and IMM2 (n = 5) were similar. Conclusion Stenting or endarterectomy provide similar results but should be reserved for patients who fail intensive medical management. A global interpretation of all the evidence is required when interpreting results that are close to statistical significance.

BACKGROUND AND PURPOSE:There are few randomized trials comparing coiling (with or without stent placement) and flow diversion (FD) for the treatment of wide-neck unruptured intracranial aneurysms. MATERIALS AND METHODS:EVIDENCE was an investigator-led randomized (1:1) trial conducted in 7 French university hospitals. Patients with 7 to 20-mm intradural unruptured intracranial aneurysms with a 4- to 10-mm neck and a "dome/neck" ratio of ≥1 were randomly allocated to coiling with or without adjunctive stent placement or FD alone. The composite primary efficacy outcome was "treatment failure," defined as initial failure to treat the aneurysm; aneurysm rupture or retreatment during follow-up; death or dependency (mRS> 2); or an angiographic residual aneurysm adjudicated by an independent core laboratory at 12 months. The primary hypothesis (revised for slow accrual) was that FD would decrease treatment failures from 35% to 10%, requiring 90 patients. Primary analyses were intent to treat. RESULTS:Among the 91 enrolled patients, four (two in each group) withdrew consent; 87 patients were included in the analysis: 43 in the FD group and 44 in the control group. Most patients had <10 mm (57/87; 65.5%) asymptomatic ophthalmic aneurysms (75/87, 86.2%). A poor primary outcome, ascertainable in 86 patients, was reached in 8/43 patients with FD (18.6%; 95% CI, 9.7%-32.6%) compared with 10/43 coiling patients (23.3%; 95 CI, 13.1%-37.7%) (RR = 0.80; 95% CI, 0.35-1.83; P = .60). Serious adverse events were similar. CONCLUSIONS:For patients with mostly unruptured, wide-neck ophthalmic aneurysms of <10 mm, FD was not superior to coiling with or without stent placement.

OBJECTIVE Many patients recruited in the Treatment of Brain Arteriovenous Malformations Study (TOBAS) are managed conservatively. The aim of this study was to monitor what happened to those patients. METHODS TOBAS comprises two randomized controlled trials and multiple prospective registries. All patients with brain arteriovenous malformations (AVMs) can participate. This report concerns patients selected for conservative management. The primary trial outcome measure is related death or dependency (modified Rankin Scale [mRS] score > 2) at 10 years. Secondary outcomes include intracranial hemorrhages, nonhemorrhagic neurological events, and serious adverse events (SAEs). For this report, outcome results are presented using patient-years, Kaplan-Meier survival curves, and Cox log-rank tests. There was no blinding. RESULTS From June 2014 to May 2021, 1010 patients were recruited, of whom 498 (49%) were proposed the prospective observation registry. After exclusions, 434 (87%) patients remained for analysis. The majority of patients had unruptured AVMs (378/434 [87%]), of which 195 (52%) were low grade (Spetzler-Martin grade I or II). During a mean follow-up period of 3.2 years (total 1368 patient-years), the primary outcome occurred in 23 of 434 (5%) patients, corresponding to an incidence of 1.7 (95% CI 1.1-2.5) per 100 patient-years. For unruptured AVMs the incidence was 1.1 (95% CI 0.7-1.9) per 100 patient-years, and for low-grade unruptured AVMs it was 0.6 (95% CI 0.2-1.7) per 100 patient-years. Poor outcomes were more frequent in patients with a history of rupture (HR 5.6 [95% CI 2.4-13.0], p < 0.001), infratentorial AVMs (HR 2.9 [95% CI 1.1-7.3], p = 0.027), and age >= 55 years (HR 3.2 [95% CI 1.4-7.6], p = 0.007). Major intracranial hemorrhage occurred in 35 of 434 (8%) patients (incidence of 2.6 [95% CI 1.9-3.6] per 100 patient-years; 2.0 [95% CI 1.3-2.9] per 100 patient-years for unruptured AVMs and 1.3 [95% CI 0.6-2.6] per 100 patient-years for low-grade unruptured AVMs). Major AVM hemorrhages were more frequent in ruptured (HR 4.4 [95% CI 2.1-8.9], p < 0.001), large (HR 2.6 [95% CI 1.1-6.6], p = 0.039), and high-grade (HR 2.5 [95% CI 1.2-5.3], p = 0.013) AVMs and those with deep venous drainage (HR 2.1 [95% CI 1.1-4.2], p = 0.032). SAEs occurred in 48 of 434 (11%) patients (incidence of 3.6 [95% CI 2.7-4.8] per 100 patient-years). For unruptured AVMs the incidence was 2.8 (95% CI 2.0-4.0) per 100 patient-years, and for low-grade unruptured AVMs it was 1.8 (95% CI 1.0-3.2) per 100 patient-years. CONCLUSIONS Nearly half of TOBAS participants were observed. Rates of untoward neurological events were within expected boundaries.

Magnetic resonance angiography (MRA) is a commonly used non-invasive imaging modality for the follow-up of treated intracranial aneurysms. The 3-category Montreal classification system is widely used to evaluate treatment outcomes, including aneurysms treated with Woven EndoBridge (WEB) devices or coils. This study aimed to assess the reliability and clinical implications of the Montreal scale for aneurysms imaged with MRA. An electronic portfolio of 60 aneurysms (30 treated with WEB devices and 30 with coils) was evaluated by 30 raters from diverse clinical backgrounds. Raters graded aneurysm occlusion using the Montreal classification system and provided management recommendations (delayed follow-up, close follow-up, or re-treatment). Twenty-three raters completed a second evaluation of permuted cases after one month. Reliability was assessed using Gwet’s AC2 (κG) coefficients, and the correlation between occlusion grade and management recommendation was analyzed with Cramer’s V. Inter-rater agreement for occlusion grades was substantial (κG = 0.70; 95

Background Decision making for preventive interventions in asymptomatic patients, such as the treatment of incidental intracranial aneurysms, is eminently uncertain and at risk of over-treatment. One approach suggests that the weighing of the natural risk of the disease against the risk of intervention should be replaced by a comparison of outcomes measured as expected quality-adjusted life-years survival. Methods We review the problems of over-diagnosis and over-treatment and how prognostic studies can help address the clinical uncertainty. We examine and compare the assumptions that underlie the mathematical transformations that are involved in the so-called outcome-based approach with the risk-based approach when they are both derived from observational data. Finally, we propose a more pragmatic approach. Results Both risk-based and outcome-based models depend on two strong assumptions: exchangeability of patients selected to be observed and patients selected to be treated (in other words ignorability of treatment assignment), and ii) dominance of time-to-event data (the only thing pertinent for decision making is the time to the first event in the patient’s history). The outcome-based approach needs an additional assumption: fatality (once a patient suffers a poor outcome from an event, recovery is impossible). These three theoretical assumptions are rarely verified in practice. Conclusion Clinical decision-making based on observational data relies on unrealistic assumptions. Clinical practice should instead be guided by conducting pragmatic clinical trials.

BACKGROUND:Transvenous embolization (TVE) is a new treatment of brain arteriovenous malformations (AVMs). The safety and efficacy of TVE have not been compared with transarterial embolization (TAE). The primary hypothesis of this trial was that TVE would increase the proportion of AVM occlusion from 40% to 80%. METHODS:The TATAM (Transvenous Approach for the Treatment of Cerebral AVMs) was an investigator-initiated, multicenter, prospective, phase 2, open-label, controlled randomized trial comparing TVE versus TAE alone (1:1). Patients with symptomatic or asymptomatic AVMs considered curable with 2 endovascular sessions were reviewed by a case selection committee. Participating centers were experienced (>20 cases) or proctored by experts. The primary outcome was complete occlusion of the AVM, confirmed by catheter angiography at 3 to 6 months. There was no blinding. Primary analyses were intent-to-treat. RESULTS:From May 2019 to April 2023, 77 patients were recruited in 7 French and 2 Canadian centers. After exclusions, results from 69 patients were analyzed; 35 were allocated TVE and 34 TAE. The mean age of participants was 43.3 years; 29 of 69 (42%) were female. There were 2 crossovers. The primary outcome was reached in 29 of 35 patients with TVE (83% [95% CI, 67%-92%]) compared with 20 of 34 patients allocated TAE (59% [95% CI, 42%-74%]; P=0.036). Poor outcomes (modified Rankin Scale score >2 at 3-6 months) occurred in 7 of 35 patients with TVE (20% [95% CI, 10%-36%]) and 9 of 34 patients with TAE (27% [95% CI, 15%-43%]; P=0.578) and related serious adverse events in 12 of 35 patients with TVE (34% [95% CI, 21%-51%]) and 14 of 34 patients with TAE (41% [95% CI, 26%-58%]; P=0.624). CONCLUSIONS:TVE was more effective than TAE in terms of angiographic results at 3 to 6 months. Morbidity was similar but high for both groups. More studies are needed to determine the role of curative embolization in managing patients with brain AVM. REGISTRATION:URL: https://www.clinicaltrials.gov; Unique identifier: NCT03691870.

BACKGROUND:Prognostic studies may inform individuals about the future course of their disease and help clinical decision making, but problems abound. METHODS:We summarize a study on the natural history of unruptured intracranial aneurysms (UIAs) and review the various steps in the construction of prognostic models. We emphasize the fundamental inductive problems of prognostic studies that attempt to use the backward road from the extension of patients suffering an event to create a new intensional definition of classes of patients at risk. RESULTS:The first step in a prognostic model is to identify candidate baseline variables to be entered into the model, according to background knowledge, previous studies, and statistical associations with the event of interest. This is a multivariate task. The modeler already knows the outcomes the model is supposed to 'predict', so that multiple models are tested against the data until a satisfactory fit is obtained. The variables used to construct the model should not be added in an ad hoc fashion to fit heterogeneous studies. They should be selected in such a fashion as to be exportable outside the study to new patients. An infinite number of models can fit the same data. Thus, the most important step is to validate the prognostic value of the model in patients that were not used to construct the model. In the case of UIAs, this has never been done. CONCLUSION:Prognostic studies present multiple problems. Unvalidated models should not be used in clinical practice.

OBJECTIVE:Many patients with ruptured intracranial aneurysms (RIAs) underrepresented or excluded from previous randomized controlled trials (RCTs) comparing surgery with endovascular treatment (EVT) are still considered for surgical clipping, but the best management of these patients remains unknown. METHODS:The International Subarachnoid Aneurysm Trial-2 was a randomized trial comparing surgical versus EVT of RIAs considered for surgical clipping, despite the results of previous RCTs, and also eligible for EVT. The primary endpoint was death or dependency according to the modified Rankin Scale score (mRS score > 2) at 1 year. Secondary endpoints included 1-year angiographic results and length of hospital stay. The primary hypothesis was that endovascular management would decrease the number of poor outcomes (mRS score > 2) from 30% to 23%, necessitating 1896 patients. The trial was interrupted after 10 years because of slow recruitment. Primary analysis was by intent-to-treat. There was no blinding. RESULTS:From November 2012 to December 2022, 270 patients were recruited at 6 North American and European centers. After exclusions, 263 patients were randomly allocated to receive surgery (n = 133) or EVT (n = 130). There were 12 crossovers (9 from surgery to EVT). The primary outcome was reached in 40 of 133 surgical patients (30%, 95% CI 23%-38%) compared with 35 of 130 EVT patients (27%, 95% CI 20%-35%) (p = 0.572). Residual aneurysms at 1 year were less frequent with surgery (10/118 [8%, 95% CI 5%-15%]) than EVT (22/109 [20%, 95% CI 14%-29%]) (p = 0.015). Additional procedures (ventricular drainage and decompressive craniotomy, p < 0.05) and hospitalization > 20 days were more frequent in the surgery group (69/133 [52%, 95% CI 43%-60%]) than in the EVT group (38/130 [29%, 95% CI 22%-38%]) (p < 0.001). CONCLUSIONS:This prematurely interrupted trial showed more frequent additional procedures and longer hospitalizations but better 1-year angiographic results with surgery. The primary clinical outcome, death or dependency at 1 year, was similar for EVT and surgery.

BACKGROUND:Non-inferiority (NI) trials are essential to evaluate whether new treatments which may offer some collateral benefit (such as less invasiveness, fewer side effects, availability, ease of administration, or lower cost) do so without significant loss of efficacy, as compared to treatments that are already established to be beneficial. NI trials pose specific design challenges, such as selecting appropriate NI margins, as well as unique problems of interpretation. METHODS:We examine six NI trials that compared thrombectomy with or without intravenous thrombolysis (IVT). We review fundamental problems with NI trials. RESULTS:Two of six trials reported that thrombectomy alone was non-inferior compared to thrombectomy combined with IVT. These trials used large NI margins or stopped recruitment prematurely after an interim analysis. The remaining four trials failed to demonstrate non-inferiority. In general, the chosen NI margins were unjustified and clinically unacceptable. A meta-analysis showed NI at a 10% margin, but not at a clinically pertinent 5% level. We examine (i) assay sensitivity, (ii) the constancy assumption and (iii) NI margins, three concepts that are crucial to understand NI trials. We question whether NI trials are appropriate in acute stroke, where there is little room to accept inferiority. Assessing superiority regarding surrogate outcome measures that have shown a causal relationship with clinical outcomes may be an alternative approach. CONCLUSION:Thrombectomy alone has not properly been shown non-inferior to thrombectomy with IVT. The NI trial design is poorly adapted for use in acute stroke.

BACKGROUND AND PURPOSE:Pragmatic trial results are intrinsically heterogeneous and the 'average treatment effect', on which the paradigmatic verdict of explanatory trials is based may not suffice to translate trial results into clinically meaningful conclusions applicable in practice. Examining various subgroups is problematic because they are at risk of both false negative and false positive results. METHODS:We summarize FIAT, a pragmatic care trial on flow diversion where multiple subgroups were examined. The notions of average treatment effect and interaction tests are reviewed to better understand their application in pragmatic trials. RESULTS:The trial showed flow diversion to be superior to standard treatments, but the results do not apply to all intracranial aneurysms. The notion of a 'true average treatment effect' can hardly apply when there are multiple comparator interventions and clinical heterogeneity. Various subgroups were examined in spite of negative interaction tests, to help interpret the trial results. Subgroup findings can be credible so long as they are not data-dependent 'fishing expeditions'. Meaningful clinical subgroups that have been pre-specified and integrated into the randomization scheme and power calculation provide the most credible conclusions. CONCLUSION:Non-prespecified data-dependent subgroup analyses are at high risk of being incorrect and should not be used to make clinical decisions in practice. A critical assessment of pre-specified subgroup analyses can nonetheless help interpret heterogeneous pragmatic trial results.

BACKGROUND:The research-care separation encourages clinicians to experiment without methods within the care context and trialists to design studies that may not properly inform practice. Care trials integrated into practice may solve these problems. METHODS:We first discuss clinical decision-making for SAH patients prior to the ISAT trial which compared surgery with coiling, and how the ISAT results changed practices. We then review the ISAT-2 care trial and its impact on practice in the presence of clinical uncertainty. RESULTS:Historically, ruptured aneurysms were treated with surgical clipping, with endovascular treatment limited to patients judged difficult to clip. ISAT was a turning point when it showed that many patients routinely treated with surgery would have better outcomes with coiling. With the proliferation of new endovascular devices, practices evolved and more patients could be treated endovascularly, but uncertainty regarding best management remained for numerous ruptured aneurysm patients that were not part of ISAT. Practicing under uncertainty, outside of a trial, is like performing research without methods within care. Without a scientific method of assessment, the notion of good surgical care is impossible to define, so ISAT-2 was designed. After 10 years, ISAT-2 remained inconclusive, but because the trial balanced risks for each patient, practicing within ISAT-2 provided the opportunity to realize that trial methods optimized care in real-time, long before conclusive results could be shown. CONCLUSION:Care trials are integral to a good clinical practice, whether they provide final results or not. The research-care separation should be revised to encourage care research.

Background The choice of control group in randomized clinical trials (RCTs) is crucial, from both an ethical and a scientific perspective. Thrombectomy can be performed under general anesthesia (GA) or conscious sedation (CS). Non-randomized studies and the first thrombectomy trials showed worse outcomes with GA, but studies were obviously confounded: more severe strokes required intubation for airway protection. Thrombectomy centers advocating the use of GA had to compare GA and CS in a randomized fashion. But which arm should be ‘experimental’, and which ‘standard care’? Methods We review the design of RCTs comparing GA to CS during thrombectomy for acute stroke, paying particular attention to the trial hypothesis. Results In early trial centers GA was the standard approach, with CS considered ‘experimental’. Thus, most trials tested the potential superiority of CS over GA, but most trials were too small, yielding inconclusive results that were erroneously interpreted as equivalent. In principle, GA had the burden of proof and should have been considered the experimental intervention, as GA is more invasive and associated with worse outcomes. Interventions that introduce greater risk must be justified by evidence of benefit and should be tested as experimental treatments. However, in practice, centers routinely working under GA had a learning curve to use CS, and for them the experimental intervention was CS. Conclusion We need to integrate clinical trials into practice to optimize care but the best way to compare two active treatments remains a work in progress.

BACKGROUND AND PURPOSE:Flow diversion (FD) has expanded beyond initial indications (proximal carotid artery aneurysms) to include distal circulation aneurysms (on the anterior, middle, or posterior cerebral arteries). Our objective was to examine results obtained from aneurysms in these locations in the Flow Diversion in Intracranial Aneurysm Treatment Trial (FIAT), which compared FD with alternative standard management options (ASMO). MATERIALS AND METHODS:FIAT was an all-inclusive parallel-group 1:1 randomized study comparing FD with 1 of 4 ASMOs (coiling +/- stenting, parent vessel occlusion [PVO], clipping, or observation, prespecified by clinical judgment). The primary safety outcome was death or dependency (mRS >2) at 3 months. The composite primary outcome was "treatment failure," defined as initial failure to treat the aneurysm; aneurysm rupture or retreatment during follow-up; death or dependency (mRS >2); or angiographic residual aneurysm adjudicated by an independent core laboratory at 12 months. This subgroup analysis was not prespecified, and there was no blinding. RESULTS:Of the 323 patients in FIAT, 46 (14%) with distal circulation aneurysms were randomly allocated: 23 to FD and 23 to ASMO (coiling +/- stenting 16 patients; PVO, 1 patient; clipping, 3 patients; and observation, 3 patients). Death or dependency at 3 months occurred in 1 patient (allocated ASMO). Treatment failures occurred in 6/23 FD-treated patients (26.1%; 95% CI, 12.6%-46.5%) compared with 11/22 patients treated with ASMO (50.0%; 95% CI, 30.7%-69.3%) (risk ratio = 0.52 [0.23-1.17]; P = .13). Serious adverse events were similar. CONCLUSIONS:Distal circulation aneurysms treated with FDs in FIAT showed an encouraging trend, but this analysis was underpowered. Further randomized trials are needed.

Angiographic results are commonly used to evaluate and compare the treatment of aneurysms. We sought to determine and compare the reliability and clinical implications of a commonly used 3-point angiographic outcome scale for aneurysms treated with WEB or endovascular coils. An electronic portfolio of 60 aneurysms (30 treated with WEB devices and 30 with coils) was evaluated by 32 raters of various training backgrounds, who were asked to grade angiographic results and provide management recommendations (delayed follow-up, close follow-up, or re-treatment). Twenty-five raters performed a second evaluation of permuted cases at least a month later. Agreement was analyzed with Gwet’s AC2 (κG) reliability coefficients and the relationship between angiographic grade and management recommendation using Cramer’s V. Inter-rater agreement was substantial for occlusion grade (κG = 0.78; 95