Article Text

Original article
Spontaneous coronary artery dissection and its association with heritable connective tissue disorders
  1. Stanislav Henkin1,
  2. Sara M Negrotto1,
  3. Marysia S Tweet1,2,
  4. Salman Kirmani3,4,
  5. David R Deyle3,
  6. Rajiv Gulati1,2,
  7. Timothy M Olson2,5,
  8. Sharonne N Hayes1,2
  1. 1Department of Internal Medicine, Mayo Clinic, Rochester, Minnesota, USA
  2. 2Division of Cardiovascular Diseases, Mayo Clinic, Rochester, Minnesota, USA
  3. 3Department of Medical Genetics, Mayo Clinic, Rochester, Minnesota, USA
  4. 4Division of Women and Child Health, Department of Paediatrics and Child Health, The Aga Khan University, Karachi, Pakistan
  5. 5Division of Pediatric Cardiology, Department of Pediatric and Adolescent Medicine, Mayo Clinic, Rochester, Minnesota, USA
  1. Correspondence to Dr Sharonne N Hayes, Mayo Clinic, 200 First Street SW, Rochester MN 55905, USA; Hayes.Sharonne{at}


Objective Spontaneous coronary artery dissection (SCAD) is an under-recognised but important cause of myocardial infarction and sudden cardiac death. We sought to determine the role of medical and molecular genetic screening for connective tissue disorders in patients with SCAD.

Methods We performed a single-centre retrospective descriptive analysis of patients with spontaneous coronary artery disease who had undergone medical genetics evaluation 1984–2014 (n=116). The presence or absence of traits suggestive of heritable connective tissue disease was extracted. Genetic testing for connective tissue disorders and/or aortopathies, if performed, is also reported.

Results Of the 116 patients (mean age 44.2 years, 94.8% women and 41.4% with non-coronary fibromuscular dysplasia (FMD)), 59 patients underwent genetic testing, of whom 3 (5.1%) received a diagnosis of connective tissue disorder: a 50-year-old man with Marfan syndrome; a 43-year-old woman with vascular Ehlers–Danlos syndrome and FMD; and a 45-year-old woman with vascular Ehlers–Danlos syndrome. An additional 12 patients (20.3%) had variants of unknown significance, none of which was thought to be a definite disease-causing mutation based on in silico analyses.

Conclusions Only a minority of patients with SCAD who undergo genetic evaluation have a likely pathogenic mutation identified on gene panel testing. Even fewer exhibit clinical features of connective tissue disorder. These findings underscore the need for further studies to elucidate the molecular mechanisms of SCAD.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Video abstract


Nonatherosclerotic spontaneous coronary artery dissection (SCAD) is an uncommon cause of acute coronary syndrome (ACS) and sudden cardiac death primarily affecting young, otherwise healthy women with low incidence of known cardiovascular risk factors. Diagnosis is typically made at the time of urgent coronary angiography when patients have presented with symptoms and findings of ACS.1–3 Opinion regarding optimal treatment strategies varies; therefore, patients may undergo a spectrum of treatments, including conservative management, fibrinolysis, percutaneous coronary intervention or coronary artery bypass grafting (CABG). Regardless of the initial treatment strategy, in-hospital mortality is low;3 ,4 however, the 10 year risk of recurrence, myocardial infarction, heart failure or death is as high as 47%.1 ,5

The underlying mechanism of ACS due to SCAD is not fully understood but is thought to be the result of myocardial ischaemia caused by nonatherosclerotic luminal stenosis. An obstructing intimal flap or a false lumen within the media filled with intramural haematoma creates an occlusion.6 ,7 Potential predisposing factors include pregnancy (peripartum and postpartum), systemic vasculopathies (fibromuscular dysplasia (FMD)) episodes of extreme exercise or emotional stress, systemic inflammatory conditions (systemic lupus erythematous, rheumatoid arthritis, giant cell arteritis, polyarteritis nodosa, Churg–Strauss syndrome, granulomatosis with polyangiitis) and connective tissue disorders.1 ,8 FMD has been correlated with SCAD, with documented prevalence of 50–75% in patients with SCAD.9 SCAD has also been associated with connective tissues disorders (CTD), especially Marfan syndrome or vascular Ehlers–Danlos syndrome.10–15 CTD and FMD are thought to alter the underlying architecture of the vessel, weakening the wall and predisposing to dissection, although a unifying structural vessel wall abnormality has not been identified.

Some CTDs such as Marfan syndrome are associated with mutations in a single gene (FBN1), whereas others such as hypermobility type Ehlers–Danlos syndrome are thought to be multifactorial. Recognised genes and associated diseases are shown in table 1. These genes appear to play a role in the development of CTD, which may predispose to SCAD. However, the role these genes play in predisposition to non-syndromic SCAD is uncertain, and no studies have documented the correlation between mutations in these genes and SCAD. The aim of the present study was to evaluate the prevalence of phenotypic features of CTDs in patients with SCAD, the utility of genetic screening and the association of genetic variants with SCAD.

Table 1

Genes associated with connective tissue disorders


Study population

The current study included patients with nonatherosclerotic SCAD participating in the Mayo Clinic SCAD Registry, described in Tweet et al,1 who had undergone medical genetic consultation for CTD at Mayo Clinic between 1984 and 2014 (n=116). Patients in the Registry included local patients affected by SCAD, patients referred by another provider (both within and outside of Mayo Clinic) and self-referred patients. Nonatherosclerotic SCAD (intimal dissection and intramural haematoma) was confirmed by two or more independent interventional cardiologists who reviewed clinical presentation and angiographic findings. Only patients with normal-appearing noninvolved segments of the coronary artery and other coronary arteries without evidence of coronary artery disease were included. Patients with atherosclerotic dissection or iatrogenic coronary artery trauma were excluded. Follow-up comprised scheduled review in the Mayo Clinic SCAD Clinic, medical genetics consultation and follow-up telephone calls as part of the Mayo Clinic SCAD Registry. The study was approved by the Mayo Foundation Institutional Review Board. Only patients who provided authorisation for use of their medical records for research were included in the study as mandated by Minnesota state law.

Imaging protocol

All patients underwent imaging of the neck, chest, abdomen and pelvis with CT angiography (CTA) as part of the SCAD CTA protocol with either a 128-slice single-source CT scanner (Definition AS; Siemens Healthcare, Forchheim, Germany) or a 128-slice dual-CT scanner operated in single-source mode (Definition Flash; Siemens Healthcare), as previously described by Liang et al16 CTA has been shown to more accurately diagnose moderate and severe FMD than more subtle disease; however, we had dedicated vascular and neuroradiologists interpret the studies, possibly increasing the sensitivity of the study. FMD was defined according to the American Heart Association classification as either focal concentric or tubular stenosis (focal FMD); or alternating constriction and dilatation of vessel (multifocal FMD).17 Associated features (aneurysms and dissections) were also noted.

Medical genetics evaluation

The purpose of the medical genetics evaluation was to thoroughly evaluate for physical features of heritable CTD; and to provide counselling and recommendations on genetic testing. Recorded features suggestive of heritable CTD included facial dysmorphisms, myopia, skin features, chest wall deformity, arachnodactyly, joint hypermobility, dolichocephaly, downslanting palpebral fissures, enophthalmos, malar hypoplasia, retrognathia, skin striae, soft or translucent skin, stretchy skin, easy bruising and skin fragility. In addition to qualitative evaluation, when available, the Beighton score was used to classify joint hypermobility (see online supplementary table S1),18 with score ≥5 considered significant.

Although all patients with SCAD seen in the Medical Genetics clinic were advised to undergo genetic testing, only 59 elected to do so. The rest of the patients (n=57) either declined (n=48), had pending insurance approval (n=7) or their insurance company denied their request for genetic analysis (n=2). Genetic testing was performed using a variety of vascular, connective tissue and single gene panels. The tested genes are listed in table 1.

Statistical analysis

Baseline characteristics of patients with SCAD including demographics, medical comorbidities, ACS presentation and extent of FMD were assessed with descriptive statistics. Continuous variables were summarised as mean±SD. Categorical variables were summarised by frequency and percentage. Analysis was performed using JMP V.10.0 (SAS Institute, Cary, North Carolina, USA). Significance was tested at two-sided p value of 0.05.


Baseline characteristics

One hundred sixteen patients with a history of SCAD underwent medical genetic evaluation for CTD, which included a visit with a geneticist and a genetic counsellor. Baseline characteristics of these patients are described in table 2. Mean age at first SCAD event was 44.2±8.6 years. Most (94.8%) were women with low rates of common cardiovascular atherosclerotic risk factors. The most common presentation was non-ST elevation myocardial infarction (NSTEMI; 62.2%) although more than a third of patients (34.2%) presented with ST-elevation myocardial infraction (STEMI). The mean initial ejection fraction at the time of SCAD was 52.5±11.3%.

Table 2

Baseline clinical data

All of the patients underwent screening for extracoronary vasculopathy CTA. Forty-eight (41.4%) had evidence of FMD (2.5±1.4 FMD sites). The three most common FMD sites were renal arteries (35 patients (30.2%); 19 (54.3%) with bilateral involvement), external iliac arteries (18 patients (15.5%); 10 (55.6%) with bilateral involvement) and internal carotid arteries (14 patients (12.1%); 11 (78.6%) with bilateral involvement) (table 3). Eight patients (6.9%) were found to have aneurysms; vascular beds included the splenic artery (n=2), internal carotid artery (n=3), vertebral artery (n=1), renal artery (n=1) and coeliac artery (n=1). Six patients (5.2%) were found to have focal extracoronary dissections; vascular beds included vertebral artery (n=2), external iliac artery (n=2), internal carotid artery (n=1) and superior mesenteric artery (n=1).

Table 3

Prevalence of fibromuscular dysplasia (FMD) sites as seen on CT angiography

Physical characteristics

Thirty-eight individuals (32.8%) had a normal physical examination with the remaining (67.2%) demonstrating at least one finding associated with a heritable CTD. The two most common physical exam findings were any joint hypermobility (29.3%) and myopia (26.7%) (table 4). Only six patients (5.2%) had more than three features suggestive of a heritable CTD. Three of these patients (50%) had evidence of FMD on CTA; all had involvement of at least one renal artery.

Table 4

Prevalence of physical findings in evaluation of 116 patients in the medical genetics clinic

Three patients carried the diagnosis of a connective tissue disorder prior to SCAD—a man with Marfan's syndrome, a woman with vascular Ehlers–Danlos, and a woman with undifferentiated connective tissue dysplasia. The individual with known Marfan syndrome had multiple diagnostic features including tall stature; high arched palate; prominent striae over the shoulders, upper chest and back; and pectus carinatum. The individual with previously identified undifferentiated CTD had multiple physical examination findings including myopia; translucent skin; positive wrist sign; pes planus and significant joint hypermobility at the wrists, elbows and hips bilaterally. On the other hand, the patient with vascular Ehlers–Danlos only had evidence of translucent skin without evidence of other physical exam abnormalities.

Genetic evaluation

Fifty-nine patients (50.9%) underwent genetic testing with panel-based sequencing of genes associated with vascular and/or connective tissue disorders. A variety of genetic panels were used, the most common being the 10 gene Marfan Next Generation panel (Ambry Genetics Corporation, Aliso Viejo, California, USA; n=14); 11 gene Marfan syndrome, Loeys–Dietz syndrome, Familial Thoracic Aortic Aneurysms and Dissections and Related Disorders NGS panel (Connective Tissue Gene Tests, Allentown, Pennsylvania, USA; n=11); and 9 gene Marfan syndrome, Loeys–Dietz syndrome, Familial Thoracic Aortic Aneurysms and Dissections and Related Disorders NGS panel (Connective Tissue Gene Tests, Allentown panel; n=10).

There were no significant differences in baseline characteristics between patients who elected to undergo genetic testing compared with those who did not, although there was a trend towards significance for those who presented with STEMI being more likely to undergo genetic testing (p=0.07; table 2).

Only three (5.1%) tested patients were found to have a disease-causing mutation based on in silico analyses. One patient was a 50-year-old man with an FBN1 mutation (T–C change at exon 25) and previously diagnosed Marfan syndrome without FMD who had a dissection of the distal left anterior descending coronary artery and prior Stanford type B descending thoracic aorta dissection and abdominal aorta diameter of 40 mm. A second patient was a 43-year-old woman with a COL3A1 mutation (C–G change at exon 45) compatible with vascular Ehlers–Danlos syndrome who had multiple FMD sites and had recurrent dissection of left anterior descending and ramus arteries. Neither of the two patients had a family history of CTD, vascular dissections or aneurysms. The third patient was a 45-year-old woman with a COL3A1 mutation (C–T change at exon 52) and previously diagnosed vascular Ehlers–Danlos syndrome with infrarenal aorta and bilateral common iliac artery ectasias. She suffered progressive proximal right coronary artery dissection requiring CABG and iatrogenic right external iliac dissection in the setting of coronary catheterisation; multiple family members had the same syndrome. An additional 12 patients (20.3%) had variants of unknown significance, none of which was thought to be a definite disease-causing mutation based on in silico analyses (table 1, see online supplementary table S2). The patient with previously diagnosed undifferentiated CTD elected to forego genetic analysis.


To our knowledge, this is the first study to systemically assess heritable CTD in a large cohort of patients with SCAD. Although 15 patients had reportable genetic variants identified, only 3 patients received a diagnosis of CTD as a result of a definite disease-causing mutation. The clinical implication of the remaining 12 variants of unknown significance is not well understood and could be either clinically unimportant or potentially represent an otherwise subclinical CTD conferring susceptibility to SCAD.

Previous studies have suggested that there may be a genetic predisposition to FMD.19–21 It is notable that more than 40% of SCAD survivors in our study were found to have FMD. It is possible that the true prevalence of FMD is even higher in this population, considering that the diagnosis was made with CTA rather than angiography and the fact that FMD could have been present in vascular beds that were not imaged, raising the possibility that mild forms of FMD or those manifest in non-imaged territories could have been missed. Overall, the high prevalence of FMD in this population suggests that there are yet-unknown genes that predispose patients to have both FMD and SCAD. This hypothesis is supported by a recent publication that identified five pairs of relatives with SCAD (mother–daughter; two pairs of sisters; aunt and niece; and first maternal cousins), implicating both dominant and recessive modes of inheritance.22 These observations support further investigations to determine novel genetic markers that may be associated with FMD and/or SCAD. The Mayo Clinic SCAD Registry provides patients with SCAD the opportunity to contribute to a DNA biorepository, enabling future gene-targeted and genomic testing.

Coronary dissections have been reported in Marfan syndrome, both isolated and in association with aortic dissection.10–14 Defective fibrillin protein caused by mutations in the FBN1 gene causes structural and functional perturbation of connective tissues may predispose individuals to SCAD.23 Although myopia, arachnodactyly and joint hypermobility were common in our patient population, no individual met diagnostic criteria based on physical examination, and only one individual had a disease-causing FBN1 mutation. Two additional patients had the same variant of unknown significance, both with minimal physical examination findings.

An additional five individuals had variants in the COL3A1 gene, two of whom received a diagnosis of vascular Ehlers–Danlos syndrome due to a known disease-causing mutation. The mutation causes defective type III procollagen production, predisposing to rupture of solid organs or arteries. Treatment, either with angioplasty or surgery, is often complicated due to friability of the vessels.24 Neither of our patients with confirmed genetic diagnoses had significant physical findings of CTD. The first patient had two SCAD events—the first treated with CABG after an unsuccessful percutaneous intervention while the second treated conservatively with medical therapy. She was found to have FMD in seven vascular territories with additional two splenic artery aneurysms. The other patient carried this diagnosis prior to SCAD and suffered a NSTEMI at age 45 due to spontaneous right coronary artery dissection; an attempt was made to percutaneously intervene but was unsuccessful due to difficulty with the wire crossing the lesion. She was treated medically but had distal progression of the dissection flap requiring CABG with vein graft to distal right coronary artery. The other three patients with COL3A1 variants of unknown significance had unremarkable physical findings. While two of them were found to have FMD (one patient with three affected vascular beds and one aneurysm; and the other with five affected vascular beds and one aneurysm), the third patient had no evidence of FMD or aneurysms. All three had only one known SCAD event. Although these variants are currently of unknown clinical significance, they may confer risk for a milder form of vascular Ehlers–Danlos, which has been previously reported in several families with a normal life span and less arterial complications.25 The diagnosis of vascular Ehlers–Danlos may be important in prognostication and preventative strategies, such as use of celiprolol, which may prevent arterial complications.26 Therefore, even though we identified only a small proportion of patients with SCAD with a heritable CTD, these diagnoses remain an important finding in patients with SCAD as they may guide monitoring, familial screening/counselling and management strategies.

There were several physical findings that were common in our population, including joint hypermobility, myopia, translucent skin and arachnodactyly. However, each of these is non-specific and not independently indicative of a defined CTD. Previous studies examining physical features in individuals with FMD found that joint hyperextension, myopia, pes planus and high arched palate are the most common physical findings, but no common phenotype has been described.19 Population studies have shown a range of ∼10–20% of hypermobility in any joint,27 with increased incidence in younger patients, women and athletes.18 ,28 Consistent with prior studies of individuals with FMD, our results also suggest that there may not be a specific vascular phenotype associated with predisposition to SCAD.

We did not identify common genetic mutations in patients with SCAD, underscoring the importance of research to discover genetic variants that confer risk for SCAD. Currently, acute management recommendations encourage conservative management with close surveillance in patients with SCAD who present with preserved coronary blood flow and haemodynamic stability, as percutaneous intervention in SCAD is associated with higher rates of complications compared with those with atherosclerotic coronary artery disease ACS.29 ,30 Chronic treatment strategies are grounded in non-SCAD postmyocardial infarction indications.1 ,22 Identifying SCAD susceptibility genes and defining disease mechanisms may ultimately facilitate an individualised treatment approach.

The strengths of this study include the large cohort of patients with a rare condition, made possible by the Mayo Clinic SCAD Registry. Through social media, online communities and physician referral, Mayo Clinic has been able to recruit more than 550 patients to the registry. Additionally, all of the patients were evaluated by the same three medical genetics physicians, minimising variability in clinical assessment. Although all patients were offered genetic testing, just over one-half completed it. The main limitation of the study is its retrospective nature with the consequent variability of the specific genetic analysis performed. This was primarily due to the increased availability of more comprehensive panels during the study period. However, at the discretion of the medical geneticist, seven individuals underwent only single gene testing: four patients tested for COL3A1 only, two tested for FBN1 only and one tested for MTHFR only. Additionally, as Mayo Clinic is a tertiary referral centre, there is inherent referral bias and thus we cannot estimate true population prevalence.

In conclusion, this is the first study to systematically evaluate patients with SCAD for heritable CTD as predisposing conditions. We observed no specific CTD or extracoronary vascular phenotype in patients with SCAD. The most common physical findings were joint hypermobility, myopia, translucent skin and arachnodactyly, all of which are non-specific. Genetic mutations for CTD were rare in this population of patients with SCAD, and a distinct vascular phenotype was not identified. Although few patients demonstrated genetic mutations, those with heritable CTD displayed minimal abnormalities on physical examination. Although it remains reasonable to assess patients with SCAD for heritable CTD as identification of such may significantly affect clinical care recommendations and testing of family members, further prospective studies should be performed to evaluate the clinical benefit of genetic testing on long-term morbidity and mortality of patients with SCAD, especially those with abnormal CTA findings. However, the small number of patients with CTD, notable prevalence of FMD and finding of family members who share history of SCAD suggest genetic abnormalities not yet identified. Whole-genome sequencing holds the potential to identify the molecular genetic foundation of SCAD and may ultimately enable individualised diagnosis and treatment recommendations.

Key messages

What is already known on this subject?

  • Although the underlying causes of spontaneous coronary artery dissection (SCAD) are largely unknown, a case series of familial SCAD implicating a genetic link has been previously reported. The role of genetic screening in patients with SCAD has not been previously described.

What might this study add?

  • Non-specific physical examination findings, including hypermobility, myopia, translucent skin and arachnodactyly, are common in the SCAD population. However, only 5% of patients had a connective tissue disorder identified as a result of genetic testing.

How might this impact on clinical practice?

  • Although only a minority of patients with SCAD have a known heritable connective tissue disorder, screening patients for a connective tissue disorder is reasonable as positive findings may impact short-term and long-term recommendations.


The authors thank the many patients with SCAD who are participating in our registry, without whom our work would not be possible.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • SH and SMN contributed equally.

  • Contributors The authors contributed to (1) conception and design or analysis and interpretation of data, or both; (2) drafting of the manuscript or revising it critically for important intellectual content; (3) final approval of the submitted manuscript and (4) agree to be accountable for all aspects of the work.

  • Funding Division of Cardiovascular Diseases, Mayo Clinic, Rochester, Minnesota; and SCAD Research, Inc., Scottsdale, Arizona.

  • Competing interests None declared.

  • Patient consent Obtained.

  • Ethics approval Mayo Foundation Institutional Review Board.

  • Provenance and peer review Not commissioned; externally peer reviewed.