Subarachnoid haemorrhage in first and second degree relatives of patients with subarachnoid haemorrhage

BMJ 1995; 311 doi: (Published 29 July 1995) Cite this as: BMJ 1995;311:288
  1. Jacoline E C Bromberg, neurologista,
  2. Gabriel J E Rinkel, neurologista,
  3. Ale Algra, epidemiologist,a,
  4. Paut Greebe, research nursea,
  5. Cornelia M van Duyn, genetic epidemiologist,e,
  6. Djo Hasan, neurologistsb,
  7. Martien Limburg, clinical investigator of the Netherlands Heart Foundationc,
  8. Hans W M ter Berg, neurologist,d,
  9. Eelco F M Wijdicks, neurologista,
  10. Jan van Gijna
  1. aUniversity Department of Neurology, PO Box 85500, 3508 GA Utrecht, Netherlands
  2. bUniversity Department of Neurology, Rotterdam
  3. cUniversity Department of Neurology, Academic Medical Centre, Amsterdam
  4. dDepartment of Neurology, Sittard Hospital, Sittard, Netherlands
  5. eDepartment of Epidemiology and Biostatistics, Erasmus University, Rotterdam
  1. Correspondence to: Dr Bromberg.
  • Accepted 9 May 1995

Aggregation of subarachnoid haemorrhage has been described in numerous families,1 but whether relativxes of patients are at increased risk is unknown. If they are, they might benefit from screening for unruptured intracranial aneurysms since the outcome of subarachnoid haemorrhage is poor and asymptomatic aneurysms can now be repaired with low morbidity and mortality. We therefore studied the cumulative incidence of subarachnoid haemorrhage among first and second degree relatives of patients with recent haemorrhage.

Subjects, methods, and results

We prospectively collected a series of 163 patients with subarachnoid haemorrhage verified by computed tomography from the University Hospitals of Rotterdam, Utrecht, and Amsterdam, and for every patient we constructed a pedigree including all first and second degree relatives. All these relatives were interviewed by telephone in a standardised manner; they were asked about episodes of subarachnoid haemorrhage, sudden severe headache, stroke, and sudden death. For deceased relatives a next of kin was interviewed about the cause of death. When stroke or any other brain disease was reported, medical records were obtained if available. All histories and medical documents were reviewed according to strict criteria, defined in advance, for the diagnosis of subarachnoid haemorrhage. The Cox proportional hazards model was used to compare the incidence of subarachnoid haemorrhage in first and second degree relatives.

The 163 patients had 1290 first degree relatives and 3588 second degree relatives. History or cause of death was known in 1259 (98%) of the first degree relatives and in 3038 (85%) of the second degree relatives. Ten first degree relatives (of nine index patients) and four second degree relatives had subarachnoid haemorrhage (hazard ratio 6.6 (95% confidence interval 2.0 to 21); P=0.001). In addition, seven first degree relatives and 12 second degree relatives met criteria for possible subarachnoid haemorrhage (hazard ratio 2.7 (1.4 to 5.5); P=0.004). The cumulative incidence of subarachnoid haemorrhage is shown in the figure.


We found that subarachnoid haemorrhage occurs almost seven times more often in first degree than in second degree relatives. Even when possible episodes were included the risk was still significantly higher, despite the dilution effect caused by including patients without subarachnoid haemorrhage in both groups.

To our knowledge our study is the first to show this increased risk in first degree relatives. Three previous studies addressed the incidence of familial subarachnoid haemorrhage.2 3 4 In a study from Sweden the incidence of intracranial aneurysms among siblings of patients was similar to that in the general population but data were collected by means of a written questionnaire sent to survivors of subarachnoid haemorrhage.2 In a study from Finland no distinction was made between first and second degree relatives.3 Case finding was probably less complete in both studies.2 3 In a casecontrol study from the United States no significant difference was found in the frequency of affected first degree relatives, but the family history was not verified.4

An important consideration in assessing the risk of subarachnoid haemorrhage in relatives of patients is the incidence expected from population studies; this comparison was not performed in two of the previous studies.2 3 We compared our results with those of the Oxfordshire community stroke project, which is a recent and reliable study of the incidence of stroke in Western Europe,5 and found similar rates for cumulative incidence in the second degree relatives (figure). This supports the notion that first degree relatives in particular are at increased risk of subarachnoid haemorrhage.

We conclude that a familiar factor is important in the development of subarachnoid haemorrhage. First degree relatives of patients run at least a three to seven times greater risk than the general population. This means that the lifetime risk of subarachnoid haemorrhage is between 2% and 5% in first degree relatives. Therefore, screening for unruptured aneurysms should at least be considered in first degree relatives of patients with subarachnoid haemorrhage.


Top: Kaplan-Meier curves for cumulative incidence of subarachnoid haemorrhage in first and second degree relatives of patients with subarachnoid haemorrhage. Bottom: Kaplan-Meier curves for cumulative incidence of all subarachnoid haemorrhage including possible episodes in first (10 definite and seven possible cases) and second degree relatives (four definite and 12 possible cases) of patients with subarachnoid haemorrhage. Cumulative incidence in general population (Oxfordshire community stroke project5; 19 definite and 14 possible cases according to our criteria) is shown for comparison

The complete definitions of our different diagnostic categories of subarachnoid haemorrhage and the references from which they were derived are available on request.


  • Funding This study was partially supported by the Netherlands Heart Foundation (grant No 90.321.)

  • Conflict of interest None.


View Abstract