Familial Thoracic Aortic Aneurysms and Dissections Incidence, Modes of Inheritance, and Phenotypic Patterns

Familial Thoracic Aortic Aneurysms and Dissections Incidence, Modes of Inheritance, and Phenotypic Patterns Gonzalo Albornoz, MD, Michael A. Coady, MD, Michele Roberts, MD, PhD, Ryan R. Davies, MD, Maryann Tranquilli, RN, John A. Rizzo, PhD, and John A. Elefteriades, MD Section of Cardiothoracic Surgery and Department of Genetics, Yale University School of Medicine, New Haven, Connecticut Background. We examined the genetic nature and phenotypic features of thoracic aortic aneurysms (TAAs) and dissections in a large cohort of patients. Methods. Interviews were conducted with 520 patients with TAAs and their pedigrees were compiled to identify family members with aneurysms. Study patients were divided into three groups: 101 non-marfan patients, in 88 pedigrees, had a family pattern for TAA (familial group), 369 had no family pattern (sporadic group), and 50 had Marfan syndrome (MFS). We determined incidence of familial clustering, age at presentation, rate of aneurysm growth, incidence of hypertension, correlation of aneurysm sites among kindred, and pedigree inheritance patterns. Results. An inherited pattern for TAA was present in 21.5% of non-mfs patients. The predominant inheritance pattern was autosomal dominant (76.9%), with varying degrees of penetrance and expressivity. The familial TAA group was significantly younger than the sporadic group (p < 0.0001), but not as young as the MFS group (p < 0.0001) (mean ages, 58.2 versus 65.7 versus 27.4 years). Among all 197 probands and kindred with aneurysm, 131 (66.5%) had TAA, 49 (24.9%) had abdominal aortic aneurysm (AAA), and 17 (8.6%) had cerebral or other aneurysms. Ascending aneurysm paired most commonly with ascending, and descending with abdominal. Abdominal aortic aneurysms (AAAs) and hypertension were more often associated with descending than with ascending TAAs (p < 0.001). Aortic growth rate was highest for the familial group (0.21cm/y), intermediate for the sporadic group (0.16 cm/y), and lowest for the Marfan group (0.1 cm/y; p < 0.01). Conclusions. TAAs are frequently familial diseases. The predominant mode of inheritance is autosomal dominant. Familial TAAs have a relatively early age of onset. Aneurysms in relatives may be seen in the thoracic aorta, the abdominal aorta, or the cerebral circulation. Screening of first-order relatives of probands with TAA is essential. Familial TAAs tend to grow at a higher rate, exemplifying a more aggressive clinical entity. (Ann Thorac Surg 2006;82:1400 6) 2006 by The Society of Thoracic Surgeons Our group [1] and others [2 4] have previously reported family patterns of transmission of thoracic aortic aneurysm (TAA) and dissection. The present study looked at a large number of family pedigrees of patients with TAA or dissection seen at the Yale Center for Thoracic Aortic Disease. Our goals were to confirm the genetic nature of TAAs in a large population of affected patients and families and to describe patterns of inheritance and phenotypic features among familial clusters. Rapid advances are being made in the understanding of TAA disease at the molecular genetic level. In pedigrees with several generations of multiply affected family members, chromosomal loci have been identified that relate to the TAA phenotype by using the methods of linkage analysis and gene sequencing. Thus far, these Accepted for publication April 19, 2006. Presented at the Poster Session of the Forty-second Annual Meeting of The Society of Thoracic Surgeons, Chicago, IL, Jan 30 Feb 1, 2006. Address correspondence to Dr Elefteriades, 121 FMB, 333 Cedar St, New Haven, CT 06510; e-mail: john.elefteriades@yale.edu. loci have been mapped to the 5q13-14, 11q 23.2-24, and 3p24-25 chromosome sites [5 7]. Most recently, important work by Pannu and colleagues [8] has localized the mutation on the 3p24-25 chromosome to the transforming growth factor- receptor type II. As these advances in molecular genetics continue, it remains important to evaluate the clinical patterns of genetic transmission of thoracic aortic aneurysm in large populations. That is the goal of the present investigation. Material and Methods From nearly 3000 patients presenting since 1996 to the Yale Center for Aortic Disease with TAAs or dissections for operative or nonoperative management, 520 patients were interviewed to obtain a thorough medical history and a full pedigree analysis and to permit determination of whether other family members had known TAAs. Interviews were conducted in person or by phone with prior consent. This study was approved by the Yale University Human Investigation Committee. 2006 by The Society of Thoracic Surgeons 0003-4975/06/$32.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2006.04.098

Ann Thorac Surg ALBORNOZ ET AL 2006;82:1400 6 FAMILIAL THORACIC AORTIC ANEURYSMS 1401 Fig 1. Flow diagram of pedigrees created. The interviewees consented to be interviewed. Selection of patients for interview was essentially random and not based on any clinical patient characteristics. Interviewers tended to select for contact, from the overall lists they were provided, patients recently seen rather than remotely seen, with regards to the time of interview. The labor-intensive interview process was conducted episodically over nearly 10 years, however, so we did not anticipate significant selection bias, temporal or other. Among all patients in all groups, 22% had presented to Yale or to another institution with a complication (rupture or dissection), and the rest were diagnosed after an imaging study. When a proband reported a positive family history for aneurysm disease or sudden cardiac death, detailed investigations were undertaken to determine if the suspect family member(s) indeed harbored aneurysms. With few exceptions, the kindred s having an aortic aneurysm was confirmed either by direct interview of the kindred, by obtaining a corroborating imaging study, or by verifying surgery for correction of an aortic aneurysm. Merely anecdotal cases were excluded. Family screening was strongly recommended but was not performed as a formal part of this study. When family members were screened at their local centers, the results were forwarded to us by the proband and incorporated into our spreadsheets. No evidence of a named syndromic connective tissue disorder was found in 470 of the 520 patients, and 50 were classified as having Marfan syndrome (MFS) in accordance with the Gent criteria [9]. The evaluation for Marfan disease was done by the surgical team, with formal genetic consultation only when the clinical picture appeared ambiguous. Of the non-marfan patients, 101 exhibited familial clustering of TAAs (familial group), and 369 had no family member with TAA (sporadic group). In the familial group, 23 patients had a relative who was also a proband in the group, resulting in 88 separate familial pedigrees. See Figure 1 for a patient flow diagram. For each group, the age at presentation, site of aneurysm or dissection, and rate of aneurysm growth were determined. The two-tailed, unpaired Student t test was used to evaluate for difference in the mean age at presentation. The rate of aneurysm growth was determined according to methods previously established at this institution [10]. In 73 of the 88 familial pedigrees, a correlation was possible between the site of the TAA in the proband, either ascending or descending, and the site of the arterial aneurysms among their kindred, including TAAs, abdominal aortic aneurysms (AAAs), and cerebral arterial aneurysms. In the remaining 15 pedigrees, a family member was known to have had an aneurysm, but the aneurysm site was unknown and could not be determined. The presence of antecedent hypertension was determined in 74 familial pedigrees, permitting analysis of the difference in prevalence between ascending and descending TAA involvement by using the nondirectional Yates 2 analysis. Rates of aneurysm growth for 160 non-marfan probands (31 familials and 129 sporadics) were calculated from the change in the measured aneurysm size on successive imaging studies. The method for calculating this rate of growth was published earlier [10]. The rate of growth of 39 MFS patients analyzed in our prior study is included in the analysis. The mode of inheritance is described for the 88 familial pedigrees and confirmed by a clinical geneticist (MR). Results Incidence of Familial Thoracic Aortic Aneurysm Familial clustering of TAA was evident in 101 (21.5%) of the 470 non-mfs patients. Sex of Probands Males predominated in the three groups. This was most pronounced in the familial group (2.5:1) and least in the MFS group (1.6:1). Age at Presentation A statistically significant difference was noted in the mean age at presentation among the three groups Fig 2. Rate of aneurysm growth in different groups.

1402 ALBORNOZ ET AL Ann Thorac Surg FAMILIAL THORACIC AORTIC ANEURYSMS 2006;82:1400 6 Fig 3. Modes of inheritance. Round symbols represent females and square symbols males. Black symbols indicate individuals affected by aneurysm, and open symbols indicate those free of aneurysm. A crossed line through a symbol indicates that the individual represented has died. An arrow indicates the proband. Partial blackening indicates an aneurysm other than thoracic (ie, abdominal or cerebral). The alternate aneurysms sites (abdominal, cerebral, other) are represented by blackening in the different corners of the box (right upper, right lower, left lower, respectively). See the illustrated legend accompanying the illustrations. studied. Age at presentation refers either to the age at which a symptomatic patient presented with a complication or the age at which an asymptomatic patient had an incidental finding of aneurysm on radiographic imaging. The MFS patients were significantly younger than the familial group (27.4 years versus 55.4 years, p 0.0001), and the familial group was significantly younger than the sporadic group (58.2 years versus 65.7 years; p 0.0001). Nature and Anatomic Location of Aneurysm Disease For the 520 patients in the three study groups, the ascending aorta was affected with an aneurysm or dissection in 413 cases (79.4%) and the descending aorta in 107 cases (20.6%). Arch aneurysms cases were grouped with the portion of the aorta, ascending or descending, most closely related anatomically to the arch aneurysm. For the familial group, 63 (79.7%) of 79 ascending

Ann Thorac Surg ALBORNOZ ET AL 2006;82:1400 6 FAMILIAL THORACIC AORTIC ANEURYSMS 1403 Fig 4. Distribution of sites of arterial aneurysms and dissections in kindred of familial probands. AAA abdominal aortic aneurysm; asc ascending; desc descending. aortas affected had aneurysms and 16 (20.3%) had dissections. Of 22 descending aortas, an equal number (11, 50%) were affected with aneurysms as with dissections. For the sporadic group, 235 (82.1%) of 287 ascending aortas affected had aneurysms and 52 (17.9%) had dissections. Of 82 descending aortas affected, 43 (52.5%) had aneurysms and 39 (47.5%) had dissections. For the MFS group, 41 (87.2%) of 47 ascending aortas affected had aneurysms and 6 (12.8%) had dissections. Of 3 descending aortas, all were affected with dissections. For the three groups, the presented data reveals that the proportion of dissections in the descending aorta was significantly greater than in the ascending (familial, p 0.05; sporadic, p 0.0005; MFS, p 0.005) Rate of Aneurysm Growth Growth rate determinations were done using two measurements for each patient, determined a mean of 21 months apart. The weighted average rate of growth was 0.21 cm/y for 31 familial probands, 0.16 cm/y for 129 sporadics, and 0.10 cm/y for 39 Marfan patients (p 0.01; Fig 2). Mode of Inheritance Of 88 familial pedigrees evaluated, 70 (79.5%) had an inheritance pattern that was most consistent with a dominant mode of inheritance: 30 were autosomal dominant, 24 were autosomal dominant versus X-linked dominant, 15 were autosomal dominant with decreased penetrance, and there was one pair of monozygotic probands with a likely autosomal dominant spontaneous mutation. A nondominant pattern could not be excluded in several of these pedigrees (see Fig 3). The other 18 pedigrees (20.5%) were most consistent with a recessive inheritance pattern, eight being autosomal recessive versus X-linked recessive, five autosomal recessive, and five autosomal recessive versus autosomal dominant with decreased penetrance. Distribution of Kindred Aneurysms Among the 88 familial pedigrees, the site of the aneurysm in the family members was known for 73 probands (Fig 4). These 73 probands had 124 kindred with an arterial aneurysm. Fifty-eight of these had a TAA, yielding 131 family members with TAA, 105 with ascending, and 26 with descending aneurysms. Sixty-six family members had aneurysms at sites other than the thorax, yielding a total of 197 probands and kindred with aneurysm in some site; of these, 131 (66.5%) had TAA, 49 (24.9%) had AAA, and 17 (8.6%) had cerebral or other arterial aneurysms. To determine the nature and prevalence of other arterial aneurysms in kindred of family members with TAA, we generated all possible pairings of aneurysm sites for the 131 family members with TAA and their 66 kindred with other arterial aneurysms. This resulted in 193 proband-kindred pairs, 148 pairs involving family members with ascending TAAs and 45 involving family members with descending TAAs. It was seen that the pairing of ascending TAAs with ascending TAAs, 90 (60.8%) of 148 pairs, was significantly more common than for descending TAAs with ascending TAAs, 7 (15.6%) of 45 pairs (p 0.0001). Similarly, it was seen that the pairing of descending TAAs with AAAs, 27 (60.0%) of 45 paired sites, was significantly more common than for ascending TAAs and AAAs, 34 (30.0%) of 148 paired sites (p 0.0001), using the nondirectional Yates 2 method. As can be seen in Figure 4, the kindred could harbor the aneurysm in any site, but ascending paired most commonly with ascending and descending paired most commonly with AAA. Hypertension in Ascending Versus Descending Thoracic Aortic Aneurysms Of 74 probands and kindred in the familial group for whom a history of antecedent hypertension could be ascertained, a statistically higher prevalence was found Figure 5. Hypertension (htn) in ascending (asc) versus descending (desc) familial probands.

1404 ALBORNOZ ET AL Ann Thorac Surg FAMILIAL THORACIC AORTIC ANEURYSMS 2006;82:1400 6 among patients with descending TAAs, 13 (86.7%) of 15, compared with patients with ascending TAAs, 15 (25.4%) of 59 (p 0.001), using the nondirectional Yates 2 method (Fig 5). Comment The data in this study has remained remarkably constant as our series of analyzed patients has grown and permits the following conclusions: 1. TAAs and dissections are frequently familial diseases. More than 20% of patients with a TAA and no known vascular connective tissue syndrome have at least one first-order family member with an arterial aneurysm. Our data strongly support the growing appreciation of a genetic role in the causation of TAA. This concept is also emerging strongly in the literature from other investigators [5 8, 11 18], and especially in the pioneering work of Milewicz and colleagues and [13]. 2. The predominant mode of inheritance of TAA is autosomal dominant, with varying degrees of penetrance and variable expressivity. Other forms of inheritance, including recessive patterns, were also noted. 3. Relative to their sporadic TAA cohorts, familial TAA patients tend to be younger at presentation than sporadic aneurysm patients suggesting a more aggressive clinical entity. 4. Relative to sporadic and Marfan TAA cohorts, familial TAAs grow at a higher rate again suggesting a more aggressive clinical entity and an added risk factor for associated complications. The range of growth rates noted in this study is consistent with our prior reports on our overall population of patients with TAA and dissection. That TAAs in the familial patients grew faster than even the Marfan patients is consistent with one of our prior studies [19]. 5. Patterns of aneurysm clustering between probands and their family members are apparent. Probands with ascending TAAs were significantly more likely to have kindred with ascending TAAs, whereas probands with descending TAAs were significantly more likely to have kindred with AAAs. This disparity in association of aneurysm sites suggests that descending TAAs and AAAs have clinical and possibly pathophysiologic features in common and that these features differ from those of ascending TAAs. This finding is supported in the literature [20, 21]. 5. Underlying hypertension frequently characterizes descending aneurysm patients and their kindred, but not ascending. The significantly higher prevalence of hypertension in probands and kindred with descending TAAs compared with those with ascending TAAs suggests that these segments of the aorta may also differ regarding certain associated risk factors, with hypertension also being a known risk factor for AAAs. The thrust of this investigation argues strongly that all first-order family members of patients with aneurysms of any type should be screened for TAA and AAA. We use cardiac echocardiography for younger individuals (age 40) and echocardiography plus computed tomography (CT) scans or magnetic resonance imaging (MRI) of the chest and abdomen for the older group. We screen siblings, parents, grandparents, children, and grandchildren, as well as more distant relatives in highly affected families. Additional commentary is warranted on several of these observations. The predominant mode of inheritance for our TAA pedigrees is autosomal dominant, an inheritance mode that is best explained by the transmission of a gene(s) encoding for a protein that affects the vascular wall integrity. We also found reduced penetrance and variable expressivity as well as multiple anatomic locations of familial arterial aneurysms, consistent with other studies of the inheritance pattern of TAAs [18]. The decreased penetrance, however, gives rise to the possibility of noninclusion of affected family members owing to a lack of overt clinical signs. Indeed, it is quite likely that the true rate of inheritance of TAA is even higher than this study indicates. To be counted as affected in this study, a family member needed to have a known aneurysm. Many family members of our probands may well have harbored an aneurysm but simply not have been diagnosed or aware. An additional cause for noninclusion was the relatively advanced age of onset of aneurysm formation, so that younger patients harboring the preclinically overt TAA phenotype may have also been underdiagnosed [18]. Our finding in this large series that more than 20% of TAAs occur with familial clustering is a result that has remained remarkably constant, because our series of analyzed patients has increased and accords with results from other series. This statistic certainly underestimates the true prevalence of aneurysm disease in kindred, however, because many family members may harbor unknown aneurysms. A limitation of this study is that it is based on interviews and not on routine radiographic screenings of family members. Another limitation is that only a fraction of all the patients in our database underwent the labor-intensive patient and family interviews. We have a program currently underway for screening family members of probands with thoracic aortic aneurysm by two-dimensional ultrasound scans of the thoracic and abdominal aortas. This study should approach the true incidence of familial inheritance of this disease more closely and is essential for accurately phenotyping affected family members for the purpose of linkage analysis. Simultaneously, we are performing genome-wide screening of a large cohort of patients that may identify specific genetic mutations for this known genetically heterogeneous disorder [22]. Although there is much additional genetic clarification to be done, we believe the present study, with a wealth of clinical information in a large cohort of

Ann Thorac Surg ALBORNOZ ET AL 2006;82:1400 6 FAMILIAL THORACIC AORTIC ANEURYSMS 1405 patients, clearly confirms the genetic nature of thoracic aortic aneurysm and dissection. We look forward to the day when complete identification of errant genes has been accomplished and specific blood tests can be developed for clinical screening purposes. References 1. Coady MA, Davies RR, Roberts M, et al. Familial patterns of thoracic aortic aneurysms. Arch Surg 1999;134:361 7. 2. Biddinger A, Rocklin M, Coselli J, Milewicz DM. Familial thoracic aortic dilatations and dissections: a case control study. J Vasc Surg 1997;25:506 11. 3. Hasham SN, Lewin MR, Tran VT, et al. Nonsyndromic genetic predisposition to aortic dissection: a newly recognized, diagnosable, and preventable occurrence in families. Ann Emerg Med 2004;43:79 82. 4. Cannon Albright LA, Camp NJ, Farnham JM, MacDonald J, Abtin K, Rowe KG. A genealogical assessment of heritable predisposition to aneurysms. J Neurosurg 2003;99:637 43. 5. Vaughan CJ, Casey M, He J, et al. Identification of a chromosome 11q23.2-q24 locus for familial aortic aneurysm disease, a genetically heterogeneous disorder. Circulation 2001; 103:2469 75. 6. Hasham SN, Guo DC, Milewicz DM. Genetic basis of thoracic aortic aneurysms and dissections. Curr Opin Cardiol 2002;17:677 83. 7. Kakko S, Raisanen T, Tamminen M, et al. Candidate locus analysis of familial ascending aortic aneurysms and dissections confirms the linkage to the chromosome 5q13-14 in Finnish families. J Thorac Cardiovasc Surg 2003;126: 106 13. 8. Pannu H, Fadulu VT, Chang J, et al. Mutations in transforming growth factor- receptor type ii cause familial thoracic aortic aneurysms and dissections. Circulation 2005;112:513 20. 9. De Paepe A, Devereux RB, Dietz HC, Hennekam RC, Pyeritz RE. Revised diagnostic criteria for the Marfan syndrome. Am J Med Genet 1996;62:417 26. 10. Rizzo JA, Coady MA, Elefteriades JA. Procedures for estimating growth rates in thoracic aortic aneurysms. J Clin Epidemiol 1998;51:747 54. 11. Francke U, Berg MA, Tynan K, et al. A Gly1127Ser mutation in an EGF-like domain of the fibrillin-1 gene is a risk factor for ascending aortic aneurysm and dissection. Am J Hum Genet 1995;56:1287 96. 12. Biddinger A, Rocklin M, Coselli J, et al. Familial thoracic aortic dilatations and dissections: a case control study. J Vasc Surg 1997;25:506 11. 13. Milewicz DM, Chen H, Park ES, et al. Reduced penetrance and variable expressivity of familial thoracic aortic aneurysms/dissections. Am J Cardiol 1998;82:474 9. 14. Muller BT, Modlich O, Prisack HB, et al. Gene expression profiles in the acutely dissected human aorta. Eur J Vasc Endovasc Surg 2002;24:356 64. 15. Jondeau G, Muti C, Boileau C. Aortic aneurysms excluding Marfan s syndrome. Arch Mal Coeur Vaiss 2003;96:1074 80. 16. Horike K, Kanoh M, Kurushima A, et al. Familial aortic dissection; cases involving a father, mother, and son. Kyobu Geka 2003;56:445 7. 17. Kuivaniemi H. Are there genes for aneurysm in the blueprint of the human genome? Ann Vasc Surg 2004;18:2 3. 18. Hasham SN, Lewin MR, Tran VT, et al. Nonsyndromic genetic predisposition to aortic dissection: a newly recognized, diagnosable, and preventable occurrence in families. Ann Emerg Med 2004;43:79 82. 19. Davies RR, Goldstein LJ, Coady MA, et al. Yearly rupture or dissection rates for thoracic aortic aneurysms: simple prediction based on size. Ann Thorac Surg 2002;73:17 28. 20. Coady MA, Rizzo JA, Goldstein LJ, Elefteriades JA. Natural history, pathogenesis, and etiology of thoracic aortic aneurysms and dissections. Cardiol Clin 1999;17:615 35;vii. 21. Absi TS, Sundt TM 3rd, Tung WS, et al. Altered patterns of gene expression distinguishing ascending aortic aneurysms from abdominal aortic aneurysms: complementary DNA expression profiling in the molecular characterization of aortic disease. J Thorac Cardiovasc Surg 2003;126:344 57; discussion 357. 22. Elefteriades JA. Beating a sudden killer. Sci Am 2005;293:64 71. INVITED COMMENTARY The study by Elefteriades and colleagues at Yale is yet another in the series of seminal contributions his group has made to the study of thoracic aortic disease [1]. This study is particularly important because it highlights the common occurrence of familial patterns of inheritance in thoracic aortic aneurysms (TAA) and dissections occurring in patients without defined genetic syndromes as well as the nature and natural history of these aneurysms. Although this concept is not entirely new to this report, there have been remarkably few previous studies documenting nonsyndromic familial predisposition to thoracic aortic dissection and aortic aneurysm disease. In fact, one of the earliest reports documenting a familial pattern of inheritance for TAA was also published by the Yale group [2] in 1999. The findings of the current study confirm and extend the preliminary conclusions of the earlier study in a much larger cohort of patients. The key findings of both articles are that approximately 20% of non-marfan s (MFS) syndrome patients with TAA have an inherited pattern; the growth rate for familial non-mfs patients is significantly greater than that for sporadic or MFS patients; and these aneurysms occur at a younger age than sporadic aneurysms, suggesting a more aggressive pathophysiology. The authors also report that the most common pattern of inheritance is autosomal dominant with a correlation between ascending aneurysms in the patients (probands) and ascending aneurysms in the kindreds (family members). Similarly, those with descending aneurysms are more likely to have descending thoracic or abdominal aortic aneurysms in family members, suggesting disparate genetic bases for non-mfs familial aneurysms in these two locations. Although the familial occurrence of abdominal aortic aneurysms was first described nearly 30 years ago [3], the appreciation for the importance of familial inheritance for thoracic aneurysms is quite recent. As the authors of the current report discuss, the actual incidence of familial TAA is likely to be higher than the 20% estimate due to variable penetrance, the large number of asymptomatic aneurysms likely to be found in kindreds and the relatively advanced age at presentation. Thus, in contrast to MFS patients, many first- 2006 by The Society of Thoracic Surgeons 0003-4975/06/$32.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2006.06.044