What do studies tell us about the risk of small unruptured aneurysms bleeding?
The natural history of unruptured intracranial aneurysms is of critical importance when coming to an informed decision regarding treatment. However, ascertaining the risk of aneurysm rupture is very difficult. There are a number of studies in this area but they each produce different results. At present there does appear to be a consensus opinion that treatment is normally recommended for:
aneurysms larger than 7 mm,
aneurysms that are responsible for symptoms (such as compression of nerves to the eye and headaches),
aneurysms that have increased in size or changed shape and
aneurysms that may have been responsible for a past subarachnoid haemorrhage.
Small unruptured aneurysms offer a particular challenge. What is the risk that they will rupture? Will they change or increase in size over time? For small unruptured intracranial aneurysms (less than 7mm) there are six papers that I consider of importance in understanding the outcomes and risks associated with these aneurysms. These are:
Juvela S, Porras M, Poussa K. Natural history of unruptured intracranial aneurysms: probability of and risk factors for aneurysm rupture. J Neurosurgery 93:379-387, 2000
Morita A, et al. The natural course of unruptured cerebral aneurysms in a Japanese cohort. NEJM 366:2474-2482, 2012 (UCAS)
Anonymous. Unruptured intracranial aneurysms: Natural history, clinical outcome, and risks of surgical and endovascular treatment. The Lancet 362:103-110, 2003 (ISUIA II)
Ishibashi T, Murayama Y, Urashima M, Saguchi T, Ebara M, Arakawa H, Irie K, Takao H, Abe T. Unruptured intracranial aneurysms. Incidence of rupture and risk factors. Stroke 40:313-316, 2009
Sonobe M, Yamazaki T, Yonekura M, Kikuchi H. Small Unruptured Aneurysm Verification Study SUAVe, Japan. Stroke 41:1969-1977, 2010
Broderick JP, Brown RD Jr, Sauerbeck L, Hornung R, Huston J III, Woo D, Anderson C, Rouleau G, Kleindorfer D, Flaherty ML, Meissner I, Foroud T, Moomaw ECJ, Connolly ES for FIA Study Investigators. Greater rupture risk for familial as compared to spontaneous unruptured intracranial aneurysms. Stroke 40:1952-1957, 2009
Each of these studies contributes to our understanding. However, when asking to whom do the results apply, it is important to analyse each study in great detail. These studies are not randomised control trials. This is of major importance when interpreting their results as patients undergoing treatment for small aneurysms were excluded from the study group during the same period. The only study that included all the patients seen, that is whether or not they were treated by surgery, was that of Juvela and colleagues in Helsinki who collected information on all patients from 1956 until 1978 for a median of about 20 years (the longest of any study). They concluded that the annual risk of rupture was 1.1% with a cumulative rupture rate of 9% by 10 years and 18% by 20 years from the time of diagnosis. There is good reason to believe that patients selected for conservative (non surgical) management in most studies (except Juvela) were not selected as a random control but was influenced by clinical decision at the time. Cliniciansknow of some factors that might predispose to aneurysm rupture, such as: aneurysm size; aneurysm shape; young patients with many years to follow; aneurysms that change in size or shape with time; the ratio of the aneurysm size relative to the parent artery; and aneurysms that run in the family.
The most important factors in assessing the quality of the scientific literature for long-term outcomes from an unruptured aneurysm are:
A large number of patients in the study
Excellent capture of as many patients as possible
A long period of follow-up
Very few patients with unruptured aneurysms undergoing treatment
Only the one study by Juvela et al to date follows these four inclusions. Furthermore, due to the current identification of risk factors and the difficulty and ethical case of withholding treatment when these risk factors are present, it is unlikely that a similar study could be performed to accurately answer this question today. This study was from Finland of 142 patients followed for a median of 20 years (Juvela et al: Natural history of unruptured aneurysms: probability of and risk factors for aneurysm rupture. J Neurosurgery 93:379-387, 2000). The policy at the time at their institute (during the years 1956 to 1978) was to conservatively manage (that is, no surgery) for all unruptured asymptomatic aneurysms. The majority (116 patients) had aneurysms no greater than 6 mm in diameter. 33 aneurysms subsequently ruptured, 23 of which were no greater than 6 mm in diameter. This gave a risk of rupture greater than 20% by 20 years of follow-up (1.3% per year). Larger size and smoking increased the risk of rupture. A family history did not increase the risk of rupture. The context from which the study population was identified was in patients from Helsinki that mostly had been diagnosed with a subarachnoid haemorrhage from another aneurysm (that has been treated).
Is this study by Juvela relevant to the Australian population of unruptured aneurysms? In one of our studies we looked at the difference between patients in Eastern Finland and those in Northern Sydney that developed subarachnoid haemorrhage (Mahindu A, Koivisto T, Ronkainen A, Rinne J, Assaad N, Morgan MK. Journal of Clinical Neuroscience. 2008;15:617-621). Between 2000 and 2005, Northern Sydney identified 392 cases and Eastern Finland 487 cases. There were differences between the two populations with respect to gender, aneurysm location and multiple aneurysms. There were no differences between the two groups with respect to age at rupture, size of rupture, severity of haemorrhage or size for aneurysm location. We drew the conclusion that there was a greater likelihood of aneurysm formation in the population in Finland. However, when an aneurysm forms, the progress to rupture is not related to racial background. This is of importance as the results from the Juvela data are therefore applicable to our population of patients in Northern Sydney.
A frequently quoted recent study is that of the International Study of Unruptured Intracranial Aneurysms [(Anonymous. Unruptured intracranial aneurysms: Natural history, clinical outcome, and risks of surgical and endovascular treatment. The Lancet 362:103-110, 2003 (ISUIA II)]. This is often erroneously used as the natural history of all unruptured aneurysms when in fact for every 1 aneurysm treated conservatively, nearly 2 aneurysms were selected for treatment in this audit. Selection was not on a randomised basis. At the time of the study, neurosurgeons were not naïve as to the factors that are likely to lead to aneurysm rupture (such as shape – Crawford T. Some observations on the pathogenesis and natural history of intracranial aneurysms. J Neurol Neurosurg Psychiat. 1959;22.259) or factors that increased the risk of surgery (Jamieson KG. Aneurysm of the vertebrobasilar system. 1968;28:544). Therefore, the one in three cases elected to be watched were likely to be those thought by the clinician, to be of lower risk of rupture or have greater risk of surgery. This distorts the results of an audit such as ISUIA in that the patients included in the study may well have a lower risk of rupture than those that were treated.
Of interest from this ISUIA study, for those that had a previous aneurysm rupture and for those who had aneurysms running in the family, there appeared to be no increased risk of subsequent haemorrhage for the period followed.
Subsequent to this study, a trial was performed for those who had aneurysms running in the family. Some of my patients were included in this study. This study (Broderick JP, Brown RD Jr, Saurbeck L, et al. Greater rupture risk for familial as compared to sporadic unruptured intracranial aneurysms. Stroke. 2009;40:1952) looked at 113 subjects diagnosed with intracranial aneurysms on screening. 11 of the 113 cases underwent treatment. Of the untreated patients, five had subsequent subarachnoid haemorrhage (four of these aneurysm were 4mm or less in size). They concluded that there was a 1.2% per year rupture rate. They observed that this was significantly higher than the less than 0.07% rupture rate for ISUIA for equivalent size and location aneurysm. They concluded that those with familial aneurysms had a significantly greater risk of rupture than sporadic aneurysms. Given the closeness of the results to that of Juvela, and given the findings by others that familial aneurysms do not have a greater risk of rupture than sporadic aneurysms, it could be equally concluded that the results of ISUIA significantly under estimate the risk of rupture from small aneurysms. We believe that this is more likely and that the results confirm the findings of Juvela that size is of some importance, but small aneurysms rupture with a rate of approximately 1% per year.
What is the chance of a small aneurysm growing?
Aneurysms do not grow at a uniform rate. We believe they grow in fits and starts. They can remain stable at a particular size and then suddenly increase in size. One study from Melbourne, found that 40% of aneurysms that were followed showed some growth within 2 years of first being monitored (So TY, Dowling R, Mitchell PJ, Laidlaw J, Yan B. Risk of growth in unruptured intracranial aneurysms: A retrospective analysis. J Clin Neurosci. 2010;17:29-33). It is probably during these changes in size or shape that the aneurysm is likely to be most unstable and rupture. Therefore, having being diagnosed with an aneurysm that is small and conservatively managed, it is important to follow the aneurysm with surveillance MRI/MRA or CTA.