In light of recently presented and generally positive findings from the COAST study, which deemed coiling to be safe in the treatment of small (<5mm) intracranial aneurysms, Victor Volovici (Rotterdam, The Netherlands) outlines existing clinical data, and weighs up the real-world pros and cons of this approach.
During a well-represented, high-stakes meeting on the treatment of intracranial aneurysms, a US physician once stated: “Coiling aneurysms under 5mm is like voting for Trump—nobody says they do it, and everybody does it”.
The question of whether aneurysms under 5mm should undergo treatment at all, and whether it is safe to do it, has been lingering in the minds of researchers and clinicians for at least the past two decades. The COAST study has made an important step towards a better understanding of the consequences of such a treatment. In order to better put these results in perspective, however, we should take into consideration the natural history of intracranial aneurysms, look at ruptured and unruptured aneurysms separately, and discuss blood blister-like aneurysms as well. One of the most often-heard arguments in favour of treatment when discussing small aneurysms is the discrepancy between natural history studies, showing a low longitudinal risk of haemorrhage from small aneurysms, on one hand, and retrospective analyses showing a preponderance of small aneurysms in large subarachnoid haemorrhage (SAH) series on the other.
These facts make researchers conclude that, ‘small aneurysms do bleed, and the risk of haemorrhage may be much higher than expected’. On the one hand, natural history studies are, primarily, highly selected longitudinal registries of Japanese patients not selected for treatment. The aneurysms perceived morphologically to pose a higher risk were treated and were thus not eligible for follow-up. This state of affairs makes natural history data heavily confounded by indication and likely an underestimation of the true risk in a non-representative population. On the other hand, epidemiologically, smaller aneurysms are the larger group—their prevalence is much higher. Therefore, their higher prevalence in SAH studies should not come as a surprise and may purely reflect the mathematical distribution of aneurysm size in general. The truth probably lies somewhere in the middle, with natural history studies likely underestimating the true risk of rupture, but not to the extent suggested by these retrospective studies.
The COAST study included 300 patients from 15 centres in the USA over a timespan of about five years. In terms of the ruptured aneurysms included, these were ‘ISAT-type aneurysms’, in which coiling was proven beneficial—and it is good to see that, even for the smallest aneurysms on this spectrum, ‘simple’ coiling is feasible and should remain the go-to treatment. In terms of the unruptured aneurysms, this is where the discussion makes a turn into a greyer area. The abstract and the presentation of COAST do not give sufficient detail about the types of patients and the characteristics of the aneurysms included. Such a treatment can only be considered safe when its safety profile outweighs the lifelong bleeding risk. As such, whether a disability and mortality rate of ‘just’ 2.4% makes the treatment safe is debatable. The treatment is preventive; therefore, it can only be justified in aneurysms in which the lifelong bleeding risk is higher. Treatment is perhaps also justifiable for relatively young patients without comorbidities, and less so for older patients. One important aspect of this treatment, nevertheless, is patient preference. Often—especially in the USA—patients have a very strong treatment preference, substantiated by the wealth of online information found on reputable websites. Many patients may be more than willing to accept the 49-to-1 odds of a negative outcome.
In addition, no mention is made in the abstract or the presentation about blood blister-like aneurysms. While certain people do not agree with this terminology, and there are debates over whether these aneurysms truly ‘exist’ as a separate pathological entity, some reports suggest a more fragile aneurysmal wall and thus a greater risk of periprocedural rupture. The optimal strategy regarding blood blister-like aneurysms remains incompletely assessed. All aspects considered, the COAST is not clear yet—but most rooms have been confirmed to be enemy-free by the interventional team.
Victor Volovici is a cerebrovascular and skull-base neurosurgeon, methodologist, and clinical epidemiologist—and co-director of the Center for Complex Microvascular Surgery—at Erasmus University Medical Center in Rotterdam, The Netherlands.
