Interventional Radiology · Neurointervention · Primary publication (2014) with final follow-up (2020)

ARUBA (2014): Medical Management versus Intervention for Unruptured Brain AVMs

ARUBA compared medical management alone with medical management plus interventional therapy for adults with unruptured brain arteriovenous malformations. It found a lower short-to-medium-term risk of death or symptomatic stroke with medical management.

Key takeaway

In ARUBA-eligible adults with an unruptured brain AVM, medical management alone produced fewer deaths or symptomatic strokes than medical management plus interventional therapy over a mean 33.3 months; the advantage persisted during extended follow-up.

At a glance

Study design: International, multicentre, open-label, parallel-group randomised controlled trial conducted at 39 centres in nine countries. Participants were allocated 1:1, stratified by centre, and analysed by intention to treat; recruitment was stopped early after a prespecified interim analysis.
Population: Adults aged 18 years or older with an unruptured brain arteriovenous malformation, no previous interventional therapy, and a lesion judged by the participating centre to be suitable for an intervention intended to eradicate it.
Sample size: 223 participants in the 2014 interim-analysis dataset; 226 participants in the final follow-up cohort
Intervention: Medical management plus lesion-directed intervention selected by the treating team: neurosurgery, embolisation or stereotactic radiotherapy, used alone or in any combination, sequence or number.
Comparator: Medical management alone, including pharmacological treatment of symptoms and coexisting medical disorders, without an intervention intended to eradicate the AVM.
Primary endpoint: Time to death from any cause or symptomatic stroke.

Key results

At the 2014 analysis, after a mean follow-up of 33.3 months, the primary endpoint occurred in 11 of 109 participants (10.1%) assigned to medical management alone versus 35 of 114 (30.7%) assigned to interventional therapy; hazard ratio 0.27 (95% CI 0.14–0.54) in favour of medical management.
In the 2020 final follow-up report, after a mean follow-up of 50.4 months, death or symptomatic stroke occurred in 15 of 110 participants receiving medical management alone versus 41 of 116 receiving interventional therapy; incidence rates were 3.39 versus 12.32 per 100 patient-years, respectively, with hazard ratio 0.31 (95% CI 0.17–0.56).

Limitations

Recruitment was stopped early for benefit at an interim analysis, leaving a much smaller cohort and shorter observation period than originally intended.
The interventional arm pooled neurosurgery, embolisation and stereotactic radiotherapy, alone or in combinations; the trial was not powered to determine the efficacy of any individual modality or a specific low-risk subgroup.
Eligibility required that a centre consider the lesion treatable and that the patient accept randomisation, limiting generalisability across the heterogeneous spectrum of brain AVMs and specialist-centre practice.
Even the extended mean follow-up of 50.4 months is short relative to the lifetime haemorrhage risk of an unruptured AVM and the latency to benefit after stereotactic radiotherapy.

Why it matters

ARUBA provided randomised evidence challenging the assumption that prophylactic eradication of every unruptured brain AVM improves patient outcomes. It shifted counselling towards explicitly weighing the immediate procedural hazard against the natural-history risk of an untreated lesion.

Practice implication

For adults resembling the ARUBA population, routine prophylactic intervention should not be assumed to improve outcomes: medical management is supported over the first several years. Decisions about intervention remain individualised in a multidisciplinary cerebrovascular team because ARUBA did not establish the value of a particular modality in carefully selected subgroups or resolve lifetime risk.

Exam pearl

ARUBA studied unruptured brain AVMs: its primary endpoint was death or symptomatic stroke, and medical management was superior at the 2014 analysis with the advantage persisting in the 2020 final follow-up. The key caveat is the heterogeneous composite interventional arm.

Viva questions

What does ARUBA stand for?: A Randomized Trial of Unruptured Brain Arteriovenous Malformations.
What was the design of ARUBA?: It was an international, multicentre, open-label, parallel-group randomised controlled trial comparing medical management alone with medical management plus interventional therapy.
Which patients were studied in ARUBA?: Adults with an unruptured brain AVM, no previous interventional treatment, and a lesion considered suitable for eradication by the participating centre.
What treatments constituted the ARUBA interventional arm?: Neurosurgery, embolisation or stereotactic radiotherapy, used alone or in any combination, sequence or number, together with medical management.
What was the primary endpoint in ARUBA?: Time to death from any cause or symptomatic stroke.
What was the primary 2014 result of ARUBA?: At a mean 33.3 months, death or symptomatic stroke occurred in 10.1% with medical management versus 30.7% with intervention; the hazard ratio was 0.27 in favour of medical management.
Did the ARUBA result persist on final follow-up?: Yes. At a mean 50.4 months, death or symptomatic stroke remained less frequent with medical management, with a hazard ratio of 0.31.
What is the central management implication of ARUBA?: In patients resembling the trial population, routine prophylactic intervention should not be presumed beneficial; conservative medical management should be the reference strategy during the first several years.
What is the main interpretive limitation of ARUBA for an interventional radiologist?: It pooled embolisation, surgery and stereotactic radiotherapy into one heterogeneous arm and was not powered to judge any individual technique or carefully selected subgroup.
Does ARUBA prove that no unruptured brain AVM should ever be treated?: No. It supports medical management in the studied population over the observed period, but it does not resolve modality-specific benefit, selected low-risk lesions or lifetime haemorrhage risk.