Prevalence of Brain Arteriovenous Malformations in First-Degree Relatives of Patients With a Brain Arteriovenous Malformation Janneke van Beijnum, MD, PhD; H. Bart van der Worp, MD, PhD; Ale Algra, MD, PhD; W. Peter Vandertop, MD, PhD; René van den Berg, MD, PhD; Patrick A. Brouwer, MD; Jan Willem Berkelbach van der Sprenkel, MD, PhD; L. Jaap Kappelle, MD, PhD; Gabriël J.E. Rinkel, MD, FRCPE; Catharina J.M. Klijn, MD, PhD Background and Purpose—It is uncertain whether familial occurrence of brain arteriovenous malformations (BAVMs) represents coincidental aggregation or a shared familial risk factor. We aimed to compare the prevalence of BAVMs in first-degree relatives (FDRs) of patients with BAVM and the prevalence in the general population. Methods—We sent a postal questionnaire to 682 patients diagnosed with a BAVM in 1 of 4 university hospitals to retrieve information about the occurrence of BAVMs among their FDRs. We calculated a prevalence ratio using the BAVM prevalence among FDRs and the prevalence from a Scottish population-based study (93 per 628 788 adults). A prevalence ratio of ≥9 with a lower limit of the 95% confidence interval of 3 was considered indicative of a shared familial risk factor. Results—Informed consent was given by 460 (67%) patients, who had 2992 FDRs. We identified 3 patients with a FDR with a BAVM, yielding a prevalence ratio of 6.8 (95% CI, 2.2–21). Conclusions—The prevalence of BAVMs in FDRs of patients with a BAVM was increased but did not meet our prespecified criterion for a shared familial risk factor. In combination with the low absolute risk of a BAVM in FDRs, our results do not support screening of FDRs for BAVMs. (Stroke. 2014;45:3231-3235.) Key Words: family ◼ intracranial arteriovenous malformations
I
t has long been presumed that brain arteriovenous malformations (BAVMs) are congenital developmental anomalies, resulting from a persisting direct connection between arterial and venous portions of the primitive vascular plexus in the embryo, which differentiates from the mesoderm in the third week of gestation.1 More recently, it has been suggested that BAVMs are caused by an abnormal response to injury in patients with a genetic predisposition.2 Familial occurrence of BAVMs has been reported in hereditary hemorrhagic telangiectasia3 and in hereditary neurocutaneous angiomatous malformations.4 In the absence of these disorders, BAVMs are thought to occur sporadically. However, several reports of families with ≥2 affected relatives have been described.5,6 It is unclear whether the described families represent coincidental aggregation or share familial risk factors.5,6 Therefore, we aimed to assess the prevalence of BAVMs in first-degree relatives (FDRs) of patients with a BAVM and to
assess whether this prevalence is higher than the prevalence in the general population.7
Methods Patients All patients with BAVMs who were seen in 1 of 4 Dutch university centers (VUmc Amsterdam, AMC Amsterdam, LUMC Leiden, and UMC Utrecht) from 1990 up to and including 2006 were eligible for this study. Patients were identified from multiple and partly overlapping sources: stroke databases, diagnosis registration systems, pathology registrations, operation registrations, registrations of embolized or radiosurgically treated patients, and reports of neurovascular meetings. The diagnosis of BAVM in these patients was confirmed with radiological (digital subtraction angiography, magnetic resonance imaging, or computed tomography angiography) or pathological (operation specimen or autopsy) examination. Patients were excluded if they had a diagnosis of hereditary hemorrhagic telangiectasia, Von Hippel–Lindau disease, Wyburn–Mason syndrome, Sturge–Weber– Dimitri disease, or other hereditary neurocutaneous malformations.
Received April 4, 2014; final revision received August 5, 2014; accepted August 26, 2014. From the Department of Neurology and Neurosurgery, Brain Center Rudolf Magnus (J.v.B., H.B.v.d.W., A.A., J.W.B.v.d.S., L.J.K., G.J.E.R., C.J.M.K.), and the Julius Centre for Health Science and Primary Care (A.A.), University Medical Center Utrecht, Utrecht, The Netherlands; Departments of Neurosurgery (W.P.V.) and Radiology (R.v.d.B.), Neurosurgical Center Amsterdam, VU University Medical Center and Amsterdam Medical Center, Amsterdam, The Netherlands; and Departments of Neurosurgery (J.v.B.) and Radiology (P.A.B.), Leiden University Medical Center, Leiden, The Netherlands. The sponsors had no role in the study design, data collection, data analyses, data interpretation, or writing of the report. The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA. 114.005442/-/DC1. Correspondence to Janneke van Beijnum, MD, PhD, Department of Neurosurgery, Leiden University Medical Center, PO Box 9600, 2300 RC, Leiden, The Netherlands. E-mail [email protected] © 2014 American Heart Association, Inc. Stroke is available at http://stroke.ahajournals.org
DOI: 10.1161/STROKEAHA.114.005442
Downloaded from http://stroke.ahajournals.org/ at Michigan State University on March 10, 2015 3231
3232 Stroke November 2014 Before the start of the study, 2 patients with BAVMs at one of the centers were known to have 1 FDR with a BAVM.6 Each patient was approached by postal questionnaire to ascertain information about the occurrence in FDRs (ie, parents, sibs, and children) of BAVMs (question 1), intracranial vascular malformations (question 2), hemorrhagic strokes (question 3), seizures (question 4), neurological deficits (question 5), sudden deaths (question 6), and performed neuroimaging (question 7). We resent the questionnaire to nonresponders after 1 month. If an index patient had died, we attempted to approach a next-of-kin to complete the questionnaire on behalf of the deceased index patient after contacting the patient’s general practitioner. We applied several strategies to improve response, such as provision of a stamped return envelope, personation of letters, signing letters by hand, using a short questionnaire, and using reminders.8 Each FDR with a possible BAVM was approached by a letter about the study sent to this relative by the index patient to ask for participation in the study. Only after obtaining the relative’s approval and informed consent, we received additional information from the FDR by telephone and by retrieval of medical information when indicated. For deceased relatives, a next-of-kin was interviewed about the cause of death. In case of brain-related death, medical records were retrieved to clarify the cause of death with approval and informed consent of a legal representative. The study was approved by the institutional ethical committee of the University Medical Center Utrecht.
Data Collection For all FDRs, we obtained information on whether they had a diagnosis of BAVM or an alternative diagnosis in case of vascular malformations, hemorrhagic stroke, seizure(s), neurological deficit(s), sudden death, or previous neuroimaging if the index patient had given a positive answer to ≥1 of the questions. In all index patients and all FDRs with a proven BAVM, we reviewed medical files for age at presentation, sex, mode of presentation, location of the BAVM (lobar, deep, and infratentorial), and diagnostic modality (digital subtraction angiography, magnetic resonance imaging, computed tomography angiography, or pathological examination).
Statistical Analysis We estimated the risk of having a BAVM among FDRs by calculating a prevalence ratio using the observed BAVM prevalence among the FDRs versus the prevalence in a Scottish population-based study.7 Because population-based cohorts of BAVMs are scarce,7,9 we regarded prevalence data from a nearby European country most appropriate as reference.7 We compared age, sex, and proportion of patients with a ruptured BAVM between the Dutch index patients and the Scottish cohort using parametric statistics when data obeyed a normal distribution and nonparametric statistics when they did not. In the Scottish cohort, the minimum crude BAVM prevalence was 93 per 628 788 adults and the maximum adjusted prevalence after capture-recapture analysis 113 per 628 788 adults.7 We calculated the prevalence ratio with the adjusted prevalence as sensitivity analysis. Because the Scottish population-based cohort reported on residents aged ≥16, we also repeated the analysis excluding FDRs who were aged C and tumor necrosis factor α −238G>A were associated with BAVM hemorrhage.14 Genome-wide association studies in a large cohort of patients have not yet been published. Given the low prevalence of BAVMs in FDRs of patients with BAVMs and the fact that benefits of interventional treatment of asymptomatic BAVMs are at best uncertain on the short term16 and on the long term,17–19 we do not recommend screening of asymptomatic FDRs of patients with a BAVM. Although it is remarkable that 2 of the 3 index patients with a FDR with a BAVM were teenagers at the time of presentation, the number of patients with BAVM in our cohort who presented = 16 years population [372] [92]1 Age at BAVM presentation 34 ± 16* 38 ± 14† 45 ± 16 Mean ± SD, years Male sex (%) 243 (53)‡ 199 (53)§ 49 (53) BAVM Presentation (%) 217 (47)‖ 173 (47)# 42 (46) Ruptured 243 (53) 199 (53) 50 (54) Unruptured 132 (29) 106 (28) 25 (27) Seizure 111 (24) 93 (25) 25 (27) Other BAVM, Brain arteriovenous malformation; SD, standard deviation. Statistics: * mean difference -11 (95% CI -15 to -7.4); † mean difference -7,0 (95% CI -10 to -3.7);‡ difference between proportions 0.00% (95% CI -0.12 to 0.11);§ difference between proportions 0.00% (95% CI -0.11 to 0.12); ‖ difference between proportions 1.5% (95% CI -9.6 to 13); # difference between proportions 0.85% (95% CI -11 to 12).
4
Prevalence of Familial Brain Arteriovenous Malformations
REFERENCES (1) Al Shahi Salman R, Bhattacharya JJ, Currie DG, Papanastassiou V, Ritchie V, Roberts RC, et al. Scottish Intracranial Vascular Malformation Study (SIVMS): evaluation of methods, ICD-10 coding, and potential sources of bias in a prospective, populationbased cohort. Stroke 2003;34:1156-62.
5
Prevalence of Brain Arteriovenous Malformations in First-Degree Relatives of Patients With a Brain Arteriovenous Malformation Janneke van Beijnum, H. Bart van der Worp, Ale Algra, W. Peter Vandertop, René van den Berg, Patrick A. Brouwer, Jan Willem Berkelbach van der Sprenkel, L. Jaap Kappelle, Gabriël J.E. Rinkel and Catharina J.M. Klijn Stroke. 2014;45:3231-3235; originally published online September 18, 2014; doi: 10.1161/STROKEAHA.114.005442 Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 2014 American Heart Association, Inc. All rights reserved. Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located on the World Wide Web at: http://stroke.ahajournals.org/content/45/11/3231
Data Supplement (unedited) at: http://stroke.ahajournals.org/content/suppl/2014/10/23/STROKEAHA.114.005442.DC1.html
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Stroke is online at: http://stroke.ahajournals.org//subscriptions/
Downloaded from http://stroke.ahajournals.org/ at Michigan State University on March 10, 2015
I
t has long been presumed that brain arteriovenous malformations (BAVMs) are congenital developmental anomalies, resulting from a persisting direct connection between arterial and venous portions of the primitive vascular plexus in the embryo, which differentiates from the mesoderm in the third week of gestation.1 More recently, it has been suggested that BAVMs are caused by an abnormal response to injury in patients with a genetic predisposition.2 Familial occurrence of BAVMs has been reported in hereditary hemorrhagic telangiectasia3 and in hereditary neurocutaneous angiomatous malformations.4 In the absence of these disorders, BAVMs are thought to occur sporadically. However, several reports of families with ≥2 affected relatives have been described.5,6 It is unclear whether the described families represent coincidental aggregation or share familial risk factors.5,6 Therefore, we aimed to assess the prevalence of BAVMs in first-degree relatives (FDRs) of patients with a BAVM and to
assess whether this prevalence is higher than the prevalence in the general population.7
Methods Patients All patients with BAVMs who were seen in 1 of 4 Dutch university centers (VUmc Amsterdam, AMC Amsterdam, LUMC Leiden, and UMC Utrecht) from 1990 up to and including 2006 were eligible for this study. Patients were identified from multiple and partly overlapping sources: stroke databases, diagnosis registration systems, pathology registrations, operation registrations, registrations of embolized or radiosurgically treated patients, and reports of neurovascular meetings. The diagnosis of BAVM in these patients was confirmed with radiological (digital subtraction angiography, magnetic resonance imaging, or computed tomography angiography) or pathological (operation specimen or autopsy) examination. Patients were excluded if they had a diagnosis of hereditary hemorrhagic telangiectasia, Von Hippel–Lindau disease, Wyburn–Mason syndrome, Sturge–Weber– Dimitri disease, or other hereditary neurocutaneous malformations.
Received April 4, 2014; final revision received August 5, 2014; accepted August 26, 2014. From the Department of Neurology and Neurosurgery, Brain Center Rudolf Magnus (J.v.B., H.B.v.d.W., A.A., J.W.B.v.d.S., L.J.K., G.J.E.R., C.J.M.K.), and the Julius Centre for Health Science and Primary Care (A.A.), University Medical Center Utrecht, Utrecht, The Netherlands; Departments of Neurosurgery (W.P.V.) and Radiology (R.v.d.B.), Neurosurgical Center Amsterdam, VU University Medical Center and Amsterdam Medical Center, Amsterdam, The Netherlands; and Departments of Neurosurgery (J.v.B.) and Radiology (P.A.B.), Leiden University Medical Center, Leiden, The Netherlands. The sponsors had no role in the study design, data collection, data analyses, data interpretation, or writing of the report. The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA. 114.005442/-/DC1. Correspondence to Janneke van Beijnum, MD, PhD, Department of Neurosurgery, Leiden University Medical Center, PO Box 9600, 2300 RC, Leiden, The Netherlands. E-mail [email protected] © 2014 American Heart Association, Inc. Stroke is available at http://stroke.ahajournals.org
DOI: 10.1161/STROKEAHA.114.005442
Downloaded from http://stroke.ahajournals.org/ at Michigan State University on March 10, 2015 3231
3232 Stroke November 2014 Before the start of the study, 2 patients with BAVMs at one of the centers were known to have 1 FDR with a BAVM.6 Each patient was approached by postal questionnaire to ascertain information about the occurrence in FDRs (ie, parents, sibs, and children) of BAVMs (question 1), intracranial vascular malformations (question 2), hemorrhagic strokes (question 3), seizures (question 4), neurological deficits (question 5), sudden deaths (question 6), and performed neuroimaging (question 7). We resent the questionnaire to nonresponders after 1 month. If an index patient had died, we attempted to approach a next-of-kin to complete the questionnaire on behalf of the deceased index patient after contacting the patient’s general practitioner. We applied several strategies to improve response, such as provision of a stamped return envelope, personation of letters, signing letters by hand, using a short questionnaire, and using reminders.8 Each FDR with a possible BAVM was approached by a letter about the study sent to this relative by the index patient to ask for participation in the study. Only after obtaining the relative’s approval and informed consent, we received additional information from the FDR by telephone and by retrieval of medical information when indicated. For deceased relatives, a next-of-kin was interviewed about the cause of death. In case of brain-related death, medical records were retrieved to clarify the cause of death with approval and informed consent of a legal representative. The study was approved by the institutional ethical committee of the University Medical Center Utrecht.
Data Collection For all FDRs, we obtained information on whether they had a diagnosis of BAVM or an alternative diagnosis in case of vascular malformations, hemorrhagic stroke, seizure(s), neurological deficit(s), sudden death, or previous neuroimaging if the index patient had given a positive answer to ≥1 of the questions. In all index patients and all FDRs with a proven BAVM, we reviewed medical files for age at presentation, sex, mode of presentation, location of the BAVM (lobar, deep, and infratentorial), and diagnostic modality (digital subtraction angiography, magnetic resonance imaging, computed tomography angiography, or pathological examination).
Statistical Analysis We estimated the risk of having a BAVM among FDRs by calculating a prevalence ratio using the observed BAVM prevalence among the FDRs versus the prevalence in a Scottish population-based study.7 Because population-based cohorts of BAVMs are scarce,7,9 we regarded prevalence data from a nearby European country most appropriate as reference.7 We compared age, sex, and proportion of patients with a ruptured BAVM between the Dutch index patients and the Scottish cohort using parametric statistics when data obeyed a normal distribution and nonparametric statistics when they did not. In the Scottish cohort, the minimum crude BAVM prevalence was 93 per 628 788 adults and the maximum adjusted prevalence after capture-recapture analysis 113 per 628 788 adults.7 We calculated the prevalence ratio with the adjusted prevalence as sensitivity analysis. Because the Scottish population-based cohort reported on residents aged ≥16, we also repeated the analysis excluding FDRs who were aged C and tumor necrosis factor α −238G>A were associated with BAVM hemorrhage.14 Genome-wide association studies in a large cohort of patients have not yet been published. Given the low prevalence of BAVMs in FDRs of patients with BAVMs and the fact that benefits of interventional treatment of asymptomatic BAVMs are at best uncertain on the short term16 and on the long term,17–19 we do not recommend screening of asymptomatic FDRs of patients with a BAVM. Although it is remarkable that 2 of the 3 index patients with a FDR with a BAVM were teenagers at the time of presentation, the number of patients with BAVM in our cohort who presented = 16 years population [372] [92]1 Age at BAVM presentation 34 ± 16* 38 ± 14† 45 ± 16 Mean ± SD, years Male sex (%) 243 (53)‡ 199 (53)§ 49 (53) BAVM Presentation (%) 217 (47)‖ 173 (47)# 42 (46) Ruptured 243 (53) 199 (53) 50 (54) Unruptured 132 (29) 106 (28) 25 (27) Seizure 111 (24) 93 (25) 25 (27) Other BAVM, Brain arteriovenous malformation; SD, standard deviation. Statistics: * mean difference -11 (95% CI -15 to -7.4); † mean difference -7,0 (95% CI -10 to -3.7);‡ difference between proportions 0.00% (95% CI -0.12 to 0.11);§ difference between proportions 0.00% (95% CI -0.11 to 0.12); ‖ difference between proportions 1.5% (95% CI -9.6 to 13); # difference between proportions 0.85% (95% CI -11 to 12).
4
Prevalence of Familial Brain Arteriovenous Malformations
REFERENCES (1) Al Shahi Salman R, Bhattacharya JJ, Currie DG, Papanastassiou V, Ritchie V, Roberts RC, et al. Scottish Intracranial Vascular Malformation Study (SIVMS): evaluation of methods, ICD-10 coding, and potential sources of bias in a prospective, populationbased cohort. Stroke 2003;34:1156-62.
5
Prevalence of Brain Arteriovenous Malformations in First-Degree Relatives of Patients With a Brain Arteriovenous Malformation Janneke van Beijnum, H. Bart van der Worp, Ale Algra, W. Peter Vandertop, René van den Berg, Patrick A. Brouwer, Jan Willem Berkelbach van der Sprenkel, L. Jaap Kappelle, Gabriël J.E. Rinkel and Catharina J.M. Klijn Stroke. 2014;45:3231-3235; originally published online September 18, 2014; doi: 10.1161/STROKEAHA.114.005442 Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 2014 American Heart Association, Inc. All rights reserved. Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located on the World Wide Web at: http://stroke.ahajournals.org/content/45/11/3231
Data Supplement (unedited) at: http://stroke.ahajournals.org/content/suppl/2014/10/23/STROKEAHA.114.005442.DC1.html
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Stroke is online at: http://stroke.ahajournals.org//subscriptions/
Downloaded from http://stroke.ahajournals.org/ at Michigan State University on March 10, 2015
