Interventional Radiology · Neurointervention · Primary publication with complete one-year and long-term follow-up

ISAT (2002): Endovascular Coiling versus Neurosurgical Clipping for Ruptured Intracranial Aneurysms

Landmark randomised comparison of endovascular coiling and neurosurgical clipping for ruptured intracranial aneurysms judged suitable for either treatment.

Key takeaway

Among selected patients with a ruptured intracranial aneurysm considered suitable for either technique, allocation to endovascular coiling reduced death or dependence at one year compared with neurosurgical clipping; long-term follow-up showed that the disability-free survival advantage persisted despite a small excess risk of late rebleeding after coiling.

At a glance

Study design: International, multicentre, open-label, randomised controlled trial with intention-to-treat analysis; recruitment was stopped early after a planned interim analysis demonstrated a significant difference in the primary outcome.
Population: Patients with aneurysmal subarachnoid haemorrhage from a ruptured intracranial aneurysm for which the treating neurosurgeon and neurointerventionist considered both endovascular coiling and neurosurgical clipping appropriate. The enrolled population was predominantly good grade and had small anterior-circulation aneurysms.
Sample size: 2,143 participants: 1,073 allocated to endovascular coiling and 1,070 allocated to neurosurgical clipping
Intervention: A treatment policy of endovascular occlusion using detachable platinum coils.
Comparator: A treatment policy of open neurosurgical craniotomy and aneurysm clipping.
Primary endpoint: Death or dependence at one year, defined as a modified Rankin Scale score of 3–6.

Key results

In the prespecified time-eligible cohort reported in 2002, death or dependence at one year occurred in 190 of 801 participants allocated to coiling (23.7%) versus 243 of 793 allocated to clipping (30.6%): relative risk 0.774 (95% CI 0.658–0.911; p=0.0019), with an absolute risk reduction of 6.9 percentage points (95% CI 2.5–11.3).
The complete one-year analysis published in 2005 confirmed death or dependence in 250 of 1,063 participants allocated to coiling (23.5%) versus 326 of 1,055 allocated to clipping (30.9%), an absolute risk reduction of 7.4 percentage points (95% CI 3.6–11.2; p=0.0001).
In the UK cohort at 10 years, 674 of 809 participants allocated to coiling (83%) and 657 of 835 allocated to clipping (79%) were alive: odds ratio 1.35 (95% CI 1.06–1.73). Coiling also increased the odds of being alive and independent at 10 years: odds ratio 1.34 (95% CI 1.07–1.67).
During follow-up beyond one year in the UK cohort, 33 participants had recurrent subarachnoid haemorrhage; 17 haemorrhages arose from the treated aneurysm, comprising 13 after coiling and 4 after clipping. The late rebleeding risk was higher after coiling but remained small in absolute terms.

Limitations

Randomisation was restricted to aneurysms judged suitable for both techniques; the findings must not be extrapolated to aneurysms for which equipoise did not exist.
The cohort predominantly comprised good-grade patients with small anterior-circulation aneurysms; poor-grade subarachnoid haemorrhage, middle cerebral artery aneurysms and some complex aneurysm morphologies were under-represented.
The trial was open label and recruitment was stopped early after an interim analysis, which can increase the apparent magnitude of benefit.
Participants were treated between 1994 and 2002 using early-generation coil technology, so the trial does not directly compare contemporary microsurgery with modern adjunctive endovascular devices.
The longest follow-up was derived from the UK cohort, with incomplete questionnaire return among long-term survivors; independence estimates therefore came from a subset of the original trial population.

Why it matters

ISAT changed the treatment paradigm for aneurysmal subarachnoid haemorrhage by providing randomised evidence that a less invasive endovascular strategy improved one-year functional outcome in appropriately selected ruptured aneurysms. Its long-term reports also defined the central trade-off: better disability-free survival after coiling versus greater durability and lower late rebleeding risk after clipping.

Practice implication

For a good-grade patient with aneurysmal subarachnoid haemorrhage from an anterior-circulation aneurysm that is equally suitable for primary coiling and clipping, current evidence supports coiling to improve the probability of a favourable one-year functional outcome. Selection should remain multidisciplinary and account for age, aneurysm morphology and location, associated haematoma, treatment durability and the need for imaging surveillance after coiling.

Exam pearl

ISAT does not prove that every ruptured aneurysm should be coiled: its conclusion applies to aneurysms considered suitable for either coiling or clipping, with the primary benefit being reduced death or dependence at one year.

Viva questions

What clinical question did ISAT address?: ISAT asked whether endovascular coiling or neurosurgical clipping produced better outcomes after aneurysmal subarachnoid haemorrhage when the ruptured aneurysm was considered suitable for either treatment.
What was the design and sample size of ISAT?: It was an international multicentre randomised controlled trial of 2,143 participants: 1,073 allocated to coiling and 1,070 to clipping.
What was the primary endpoint of ISAT?: The primary endpoint was death or dependence at one year, defined as a modified Rankin Scale score of 3–6.
What was the complete one-year ISAT result?: Death or dependence occurred in 23.5% after allocation to coiling versus 30.9% after allocation to clipping, an absolute risk reduction of 7.4 percentage points.
Why do the 2002 and 2005 one-year figures differ slightly?: The 2002 paper reported the prespecified cohort eligible for one-year assessment at the interim analysis; the 2005 report provided nearly complete one-year follow-up for the randomised cohort.
What did long-term ISAT follow-up show?: At 10 years, coiling retained an advantage for being alive and independent, but late rebleeding from the treated aneurysm was more frequent after coiling and remained uncommon overall.
What is the major generalisability limitation of ISAT?: Only aneurysms judged suitable for both techniques were randomised, and the cohort mainly contained good-grade patients with small anterior-circulation aneurysms; the result cannot be applied to every ruptured aneurysm.
How should ISAT influence present-day treatment selection?: When a good-grade anterior-circulation ruptured aneurysm is equally suitable for coiling and clipping, coiling is generally preferred for better one-year functional outcome, while final selection remains multidisciplinary and anatomy-specific.