Risk Factors for Aneurysmal Subarachnoid Hemorrhage
To the Editor:
I read with interest a recent article concerning risk factors for subarachnoid hemorrhage (SAH).1 The results of this study confirm the results of previous studies that showed for the first time that cigarette smoking is an independent risk factor for SAH irrespective of alcohol consumption, hypertension, or use of nonsteroidal anti-inflammatory drugs or narcotics.2,3⇓ Cigarette smoking seems even today to be the most important risk factor with a similar attributable risk for SAH despite a well-known decreased prevalence in smoking in Western countries during the past 10 to 15 years. In addition, cigarette smoking increases the risk for SAH in a dose-response manner. Particularly, those smoking more than 20 cigarettes per day have an increased risk.2,3⇓ The mechanism behind this increased risk is likely that cigarette smoking increases risk for rupture of an existing aneurysm by increasing its growth rate and, to a lesser degree, by causing aneurysm formation.4
Alcohol consumption seems to have a relatively short-term risk for SAH given that recent alcohol intake during the few days before SAH is a more important risk factor than long-term heavy consumption, likely because of alcohol-induced transient changes in blood pressure or other mechanisms, and because alcohol consumption seems not to cause aneurysm formation or to increase the growth rate of an existing aneurysm.3–5⇓⇓
Recently, family history of intracranial aneurysms has been offered as evidence for genetic causality of cerebral aneurysms.6,7⇓ In the present study, cigarette smoking was for the first time analyzed as a confounding factor for this association. This is important because correlation of smoking habits between family members is higher than in the general population.7,8⇓ Furthermore, 38% to 48% of SAHs can be attributed to current smoking,2,3⇓ whereas risk attributable to affected first-degree relatives accounts for only 5%.7
In the present study, there may be 2 important sources of bias, which may overestimate the significance of family history as a risk factor. Firstly, familial aneurysm cases do not necessarily mean evidence of hereditary risk for aneurysm formation independent of environmental factors. Smoking behavior (number of cigarettes smoked per day or smoking intensity) seems also to be influenced by genetic factors.8 Previously, the number of cigarettes smoked per day was a more significant risk factor than current smoking, and <10 cigarettes smoked per day seems not to increase risk.2,3⇓ This same seems to hold true for aneurysm growth.4 Thus, family history in the present study may be partly a proxy of smoking intensity or heavy smoking as a risk factor independent of current smoking.
Secondly, family history association with SAH was based only on interview data without confirmation of familial aneurysm cases. This confirmation may be a formidable task but it is necessary because even one third of cases giving a positive family history of SAH or cerebral aneurysm either in the interview or in review of medical records may be false-positive cases.6,7⇓ A high number of family history cases in the present study was comparable to that obtained in eastern Finland, where the incidence of SAH is the highest in the Western countries.6,9⇓ This also suggests that there may be a significant number of false-positive family history cases in the present study.
Our purpose was also to test family history as a risk factor for SAH in our previous study of 278 patients with a verified aneurysmal SAH (30% of whom died of SAH) and 314 controls.3 However, I gave up from coding this variable because there were several cases with spontaneous intracerebral hemorrhage, acute brain infarction, or even myocardial infarction or brain tumor who were considered as intracranial aneurysm cases during the interview of only first-degree relatives. This possibility is likely higher among SAH cases than among controls. So, reliability of coding family history of intracranial aneurysms without its verification may be even more difficult than coding alcohol consumption or hypertension.
It is recommended to use smoking intensity and verification of family history cases in future analyses to obtain valid estimates of relative risk of family history of SAH. Only these methods can show true hereditary proportion of risk. All we know is that there are families with a significant history of SAH cases. Fortunately, these cases are very rare in clinical practice, and SAH seems to a great degree to be a modifiable disease.
- ↵Kissela BM, Sauerbeck L, Woo D, Khoury J, Carrozzella J, Pancioli A, Jauch E, Moomaw CJ, Shukla R, Gebel J, Fontaine R, Broderick J. Subarachnoid hemorrhage: a preventable disease with a heritable component. Stroke. 2002; 33: 1321–1326.
- ↵Longstreth WT Jr, Nelson LM, Koepsell TD, van Belle G. Cigarette smoking, alcohol use, and subarachnoid hemorrhage. Stroke. 1992; 23: 1242–1249.
- ↵Juvela S, Hillbom M, Numminen H, Koskinen P. Cigarette smoking and alcohol consumption as risk factors for aneurysmal subarachnoid hemorrhage. Stroke. 1993; 24: 639–646.
- ↵Juvela S, Poussa K, Porras M. Factors affecting formation and growth of intracranial aneurysms: a long-term follow-up study. Stroke. 2001; 32: 485–491.
We appreciate Dr Juvela’s relevant concerns regarding the potential for bias or error in our retrospective case-control study.1 He points out that family history of subarachnoid hemorrhage (SAH) may be unreliable and that family history of smoking may be a confounder in our SAH cases. It is highly likely that SAH due to intracranial aneurysm rupture is a complex trait with important environmental and genetic risk factors. While the purpose of our study was to demonstrate the impact of risk factors on the incidence of SAH, Dr Juvela raises important concerns about such retrospective family histories.
With regard to a bias resulting from false-positive family history, it is true that we did not verify the family history by review of family members’ medical records or imaging studies. In a retrospective case-control study, the most commonly encountered bias regarding family history would be recall bias, in which affected families are more likely to be aware of their family history of SAH. As a measure of the degree of recall bias, we did ask both cases and controls about a family history of dementia and Alzheimer’s disease. We found no difference of positive responses among cases and controls, nor did we find a difference in reported family history of other stroke subtypes such as ischemic stroke or intracranial hemorrhage. Nevertheless, we agree that verification of the subtype of a family history of stroke is the next step in this analysis.
While family histories of SAH may be unreliable, it is not known if they are any more or less reliable in cases as compared with controls. Dr Juvela is correct that data about reliability of family history must be collected in future studies, but in a retrospective study such as ours, validation would be a formidable task requiring much time and expense. Such data will be collected systematically in a future study that will soon be starting (the Familial Intracranial Aneurysm or FIA Study). Our results are consistent with other published results in which odds ratios for family history ranged between 1.8 and 6.6.2–4⇓⇓
It is possible that family history of smoking may be a confounder for family history of aneurysmal SAH. We would first point out that many of our SAH cases with a family history of SAH did not smoke. It may be helpful to consider a pedigree from the FIA study in which 3 of 6 siblings had SAH as a result of ruptured intracranial aneurysm by age 40. While 2 of the 3 were smokers, the risk for SAH at this young age is extremely low (a cumulative absolute risk of <1% for smokers up to age 40). Given that 50% of siblings were affected by this age, it is unlikely that SAH was due solely to smoking. Rather, it is more likely that smoking increases the penetrance of a genetic predisposition to aneurysmal rupture.
Smoking is consistently recognized as the most important risk factor for SAH and may aggregate in families due to environmental/behavioral reasons as stated by Dr Juvela or due to genetic causes (ie, a gene for smoking or addiction). It is thus possible that the familial aggregation of SAH may indeed be a genetic susceptibility to smoking, which leads to SAH.
Given that we did not ask about family history of smoking in our interviews, we cannot quantify this effect. Any such information obtained would also be subject to recall bias unless all first- and second-degree family members among cases and controls were interviewed directly. In our study,1 family history of subarachnoid hemorrhage was a risk factor for SAH independent of both current and former smoking (see Table 3, matched logistic regression). Our report was a preplanned midpoint analysis and there was insufficient power to consider interactions. In the final logistic regression analysis, an interaction between smoking and family history may lend further indirect evidence to the effect of smoking among families. Our upcoming Familial Intracranial Aneurysm study will examine the importance of a family history of smoking and the potential susceptibility to SAH.
Dr Juvela’s points are well taken. Future epidemiologic studies of SAH must take these factors into account as much as can be done practically.