(Stroke. 1999;30:2238-2248.)
© 1999 American Heart Association, Inc.
Letters to the Editor |
Should We Screen for Familial Intracranial Aneurysm?
Departments of Neurosurgery and Neuroradiology, Institute of Neurological Sciences, Southern General Hospital, Glasgow, UK
Department of Clinical Neurosciences, Western General Hospital, Edinburgh, UK
Key Words: cerebral aneurysm • screening
 |
Introduction |
|---|
 Top Introduction ReferencesIntroduction References |
|---|
To the Editor:
The recent article by Crawley et al1 understates the increasing case against screening for familial intracranial aneurysms by overestimating the risk of aneurysm rupture and the accuracy of MR angiography (MRA) in a screening context, as well as underestimating the costs and risks of screening.
The calculations of Crawley et al are based on an annual rupture risk of 0.4% to 1.5%.2 This figure is derived from a systematic review, but since then the International Study of Unruptured Intracranial Aneurysms,3 the largest study of unruptured aneurysms to date, found an annual rupture risk of only 0.05% for aneurysms <10 mm in size and 0.5% for aneurysms >10 mm (or aneurysms in patients with a previous aneurysmal subarachnoid hemorrhage). The figure of 0.05% should be regarded as the more applicable to screening for aneurysms in asymptomatic relatives of subarachnoid haemorrhage patients.
The value of MRA as a screening tool for the detection of intracranial
aneurysms is still controversial, and the sensitivity and
specificity of 90% quoted may be optimistic. We have systematically
reviewed the world literature and identified 20 prospective "blinded
reader" studies (of 
10 subjects) comparing MRA with digital
subtraction angiography and published between 1988 and 1997 that met
quality criteria.4 The sensitivity of MRA ranged from 56%
to 97% (median 88%) and specificity from 75% to 100% (median 95%),
although not all papers provided sufficient data to calculate
specificity. However, most intracranial aneurysms detected by a
screening program would be <10 mm in size and more than a third
would be <5 mm.5 MRA is much less accurate for small
aneurysms (<5 mm), with a sensitivity as low as
54%.6 Crucially, 19 of 20 studies were performed in
populations in which the prevalence of aneurysms was >50%
(and it was 10% in the remaining study), whereas a very low prevalence
would be expected in a screening context. While it had been thought
that prevalence did not influence sensitivity or
specificity,7 more recent evidence indicates that a high
disease prevalence leads to an increase in the calculated sensitivity
and specificity of a diagnostic test.8
Therefore, if MRA is used as a screening tool in a low-prevalence
population, the sensitivity will be less, possibly much less, than
90%.
The costs of screening may be significantly higher than those used in
the model. The quoted cost for MRA of $290 (
274) is
conservative. For a full screening study incorporating MRI of the
brain, MRA plus targeted maximum intensity projection
reconstructions and reported by a neuroradiologist, a figure
approaching $450 (425) is more realistic. No evidence is
quoted to support the assertion that screening would need to be
repeated at least every 10 years. This is a very long time interval,
and de novo intracranial aneurysm formation and rupture within
3 years has been observed in familial intracranial
aneurysms.9
It is also important not to understate the risk of surgery for an unruptured intracranial aneurysm. The estimate of death or dependence of 8% used by Crawley et al excludes less-severe morbidity of 5.5%.10 The prospective International Study of Unruptured Intracranial Aaneurysms data give the even higher rate for combined morbidity and mortality of 15.8%.3 People identified through a screening program for familial asymptomatic unruptured aneurysms would, in general, be healthy, therefore all morbidity after surgery should be included in the cost-benefit analysis of screening. There is a case for taking into account the benefit from reduction of anxiety from screening, but the effect of this has not been established.
The omission of coiling of aneurysms is disappointing. Even though data on the risks and benefits of coiling are more limited, it would have been a useful inclusion in the model for comparison with an earlier study on this subject.11
The available evidence indicates that the case against routine screening for familial intracranial aneurysms is stronger than that stated by Crawley et al. One way forward may be to identify which individuals within an affected family are at most risk. Although risk factors such as female sex, smoking, and heavy alcohol intake are recognized, more information is needed on the genetic basis and patterns of inheritance of familial intracranial aneurysms and subarachnoid hemorrhage.
|
References |
|---|
|
TopIntroductionReferencesIntroduction References |
|---|
1. Crawley F, Clifton A, Brown MM. Should we screen for familial intracranial aneurysm? Stroke. 1999;30:312–316.
2.
Rinkel GJE, Djibuti M, Algra A, van Gijn J. Prevalence
and risk of rupture of intracranial aneurysms: a systematic
review. Stroke. 1998;29:251–256.
3. International Study of Unruptured Intracranial Aneurysms (ISUIA) Investigators. Hemorrhage rates in patients with unruptured intracranial aneurysms. Stroke. 1998;29:273. Abstract.
4. White PM, Wardlaw JM. How reliable is the non-invasive imaging of intracranial aneurysms? An objective assessment of the quality of the published literature. Cerebrovasc Dis. 1999;9(suppl 1):51. Abstract.
5. Kojima M, Nagasawa S, Lee YE, Takeichi Y, Tsuda E, and Mabuchi N. Asymptomatic familial cerebral aneurysms. Neuroimaging Clin North Am. 1998;43:776–781.
6.
Korogi Y, Takahashi M, Mabuchi N, Nakagawa T, Fujiwara
S, Horikawa Y. Intracranial aneurysms: diagnostic
accuracy of MR angiography with evaluation of maximum intensity
projection and source images. Radiology. 1996;199:199–207.
7. Sackett DL, Haynes RB, Guyatt GH, Tugwell P. Clinical Epidemiology: A Basic Science for Clinical Medicine. Boston, Mass: Little, Brown & Co; 1987:69–152.
8. Brenner H, Gefeller O. Variation of sensitivity, specificity, likelihood ratios and predictive values with disease prevalence. Stat Med. 1997;16:981–991.[Medline] [Order article via Infotrieve]
9.
Ronkainen A, Puranen M, Hernesniemi JA, Vanninen R,
Partanen PLK, Saari JT. Intracranial aneurysms: MR angiographic
screening in 400 asymptomatic individuals with increased
familial risk. Radiology. 1995;195:35–40.
10.
Raaymakers TWM, Rinkel GJE, Limburg M, Algra A.
Mortality and morbidity of surgery for unruptured intracranial
aneurysms. Stroke. 1998;29:1531–1538.
11. Kallmes DF, Kallmes MH, Cloft HJ, Dion JE. Guglielmi detachable coil embolization for unruptured aneurysms in nonsurgical candidates: a cost-effectiveness exploration. AJNR Am J Neuroradiol. 1998;19:167–176.[Abstract]
Response
Department of Clinical Neurology, University of London, London, UK
Department of Neurology
Department of Neuroradiology, Atkinson Morley's Hospital, London, UK
Key Words: cerebral aneurysm • screening
|
Introduction |
|---|
|
TopIntroductionReferencesIntroduction References |
|---|
We thank White et al for their letter supporting our conclusions1 that screening for aneurysms is not justified in asymptomatic patients with a family history of subarachnoid hemorrhage. The fact that White et al believe we understated the case against screening while others2 have criticized our model as overstating the case suggests that we got the balance about right at the time we wrote our article. In our model we deliberately chose conservative figures to avoid any possibility of bias against screening. However, many of the figures quoted by White et al were not available when we submitted our original article and we agree that the new data further increase the case against routine screening.
|
References |
|---|
|
TopIntroductionReferencesIntroduction References |
|---|
1. Crawley F, Clifton A, Brown MM. Should we screen for familial intracranial aneurysm? Stroke. 1999;30:312–316.
2.
Solomon RA. Should we screen for familial intracranial
aneurysm? Stroke. 1999;30:1292. Letter.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||