Arterial aneurysms in children are typically due to a genetic abnormality and represent the most lethal clinical feature observed in the setting of connective tissue disorders. The most prevalent sites of vascular catastrophe in these young patients are the aortic root and the ascending aorta.

Pediatric aortic surgery for patients affected by connective tissue disorders is a rapidly evolving field. As a better understanding of the underlying aortic wall pathology is achieved, diagnosis is improved and treatment objectives, surgical indications, operative strategies, and expected outcomes change.

Although over the past 20 years treatment goals have shifted from life-saving measures to preemptive surgery in the hopes of achieving long-term event-free survival, risk stratification and surgical decision-making are still imperfect. Surgical indications are mostly based on clinical diagnosis, aneurysm size, and family history. The absence of biomarkers predictive of malignant clinical course, along with the lack of a large body of data on natural history and true long-term treatment outcomes impact our ability to predict the aortic size at which vascular complications may occur in each patient. Aortic root pathology and aneurysms are described in detail in Chapter 39. We will hereby expand on the topic by focusing on patients in the pediatric age group.

Thoracic aortic aneurysm in the setting of connective tissue disorder is a syndromic disease with multiple, complex phenotypes. Within each disease, clinical heterogeneity due to mosaicism, incomplete penetrance, and age-dependence of the phenotype is common. These features account for the lack of a complete understanding of the pathophysiology underlying thoracic aneurysm syndromes.

Historically, pathogenetic models of ascending aortic aneurysm were centered on abnormalities of structural protein of the extracellular matrix as a cause of connective tissue weakness leading to aneurysm formation. The near-uniform finding of elastolysis in the media ultimately leading to cystic medial necrosis was the anatomic hallmark. Along these lines of reasoning, Marfan syndrome was thought to be exclusively caused by deficiency of the microfibrillar protein fibrillin-1, Elhers–Danlos vascular type caused by qualitative and/or quantitative alteration of collagen type III, and bicuspid aortic valve with aneurysm caused by altered vascular collagen production.

Recent insights gained while studying Marfan syndrome highlighted the importance of cytokine dysregulation and, in particular, the potential role of TGF-β in the pathogenesis of aortic aneurysms. Deficiency of fibrillin-1 would lead to failed sequestration of TGF-β in the matrix and hence deregulated activation of and signaling by TGF-β in arterial wall tissues. TGF-β signaling, in turn, would result in the activation of proteolytic enzymes such as thrombospondin-1 and matrix metalloproteinases, causing extracellular matrix degeneration and loss of elastic properties ultimately leading to aneurysm formation.

Even though the etiology of most aortic aneurysm syndromes is still largely unknown, convincing evidence of amplified TGF-β signaling in the aortic wall has also been observed in patients with bicuspid aortic valve and aneurysm and Loeys-Dietz syndrome.^1,2 In patients with Loeys-Dietz syndrome, mutations in the genes for either TGF-β receptor 1 or 2 are associated with increased downstream TGF-β signaling within the arterial media; this in turn leads to increased collagen synthesis, loss of elastin content, and elastic fiber disarray. The ultrastructural end-result is an extremely thin aortic wall that has decreased elastic properties and is prone to dilatation and dissection.

Recent studies on aneurysm progression strongly emphasize molecular events taking place in the adventitia during aneurysm formation, challenging the long-held assumption that aneurysms originate from a disease processes harbored in the media.³ This new concept by which the adventitia assumes an active role in aneurysm formation is supported by the old observation of maintenance of wall integrity following endarterectomy and the fact that patients with failed elastogenensis and generalized medial disorganization will typically not manifest aneurysm formation.⁴

Currently, the pathogenesis of ascending aortic aneurysm in pediatric patients is conceived as the final result of a complex and yet incompletely understood interplay between signaling pathways, most likely orchestrated by altered TGF-β signaling throughout the entire aortic wall. Alteration of specific matrix proteins in the arterial wall may be the initiating event as well the end-result of this process.

Marfan Syndrome

Marfan syndrome is an autosomal dominant disorder with complete penetrance but variable expression that affects 2 to 3 in 10,000 individuals.⁵ In approximately 25 percent of patients, the disease is due to a de novo mutation.⁶ Cardinal clinical findings include annuloaortic ectasia, dislocation of the ocular lens, and long bone overgrowth (Table 82-1).

Clinical course is heavily dependent on age of presentation. The infantile form of Marfan syndrome has an early onset of aneurysm formation and is distinguished by an aggressive clinical course.^6–9 Yet the aggressiveness of the clinical course is usually dominated by mitral valve disease, which stands as the leading cause of morbidity and mortality in this particular age group.¹⁰ In fact, aortic catastrophes are exceptionally rare in Marfan patients younger than 12 years of age, almost irrespective of the degree of aortic dilatation; in this age group, aortic rupture/dissection occurs mainly in the case of giant aneurysms, often following progressive aortic insufficiency.^6,7,9,11,12 Rupture of an enlarged aortic root is instead the leading cause of death in adolescents.^5,13,14 In these patients, aortic regurgitation rarely begins at an aortic diameter <6 cm, but many experience dissection at significantly smaller root diameters.¹⁵

In children, determinants of aortic complications are aneurysm absolute size and growth rate, and family history of premature dissection.⁵ In young patients, the average size of the aortic root at the time of dissection is 5.1 cm and when aortic diameter exceeds 6 cm, there is a 4 to 8-fold increase in the cumulative risk of rupture.^16–18 Also, if the aortic root expands at a rate greater than 15 mm/year, aortic dissection is more common.¹⁴ Finally, family history of aortic complications at relatively small diameters suggests the presence of a particularly frail aortic wall and a greater propensity for premature aneurysm rupture.

Loeys–Dietz Syndromes

Loeys–Dietz syndrome is an autosomal dominant connective tissue disorder caused by mutations in the genes encoding for transforming growth factor β receptors 1 and 2 (TGF-β). It is characterized by premature ascending aortic aneurysm and dissection at small aortic diameters and at younger age compared with any other connective tissue disorder (Table 82-1 and Fig. 82-1).¹⁹

Clinical expression is variable with features that, on the mild end of the spectrum, are similar to those of Marfan syndrome.^20,21 On the opposite end of the clinical spectrum, is a complex phenotype in which aortic dissection or rupture commonly becomes apparent in early childhood.¹⁹ Two main syndromic pictures have been described. Patients with Loeys–Dietz syndrome type I are characterized by the clinical triad of hypertelorism (90 percent), bifid uvula and/or cleft palate (90 percent) (Fig. 82-2) and generalized arterial tortuosity with aneurysm formation (84 percent). None of these clinical features are present in patients with Marfan syndrome and are key for clinical diagnosis. Ninety-eight percent of these patients develop aortic root aneurysm early in life.² Patients with Loeys–Dietz syndrome type II lack these craniofacial features (though bifid uvula may be present) and have at least two manifestations of the Ehlers–Danlos syndrome (easy bruising, characteristic facial appearance, wide and atrophic scars, joint laxity, translucent/velvety skin, visceral rupture) with no type III collagen biosynthesis deficit. Patients with Loeys–Dietz syndrome type I endure a more severe cardiovascular course than those with Loeys–Dietz type II. Importantly, more severe craniofacial abnormalities are related to worse cardiovascular outcomes among type I patients.²

The most important feature of Loeys–Dietz syndrome is that aneurysms commonly dissect at sizes smaller than those seen in patients with Marfan syndrome. In children with Loeys–Dietz syndrome, aortic dissection often occurs at diameters well under 5.0 cm.^2,22 It is imperative therefore to normalize aneurysm size to age and body size (z-score) to guide risk stratification for aortic complications and appropriately plan for surgery.

Vascular Ehlers–Danlos Syndrome

Vascular Ehlers–Danlos syndrome is a rare autosomal dominant disorder, characterized by extreme fragility of the connective tissue leading to arterial and hollow organ rupture as well as anastomotic complications following surgical intervention (Table 82-1). Dissection and rupture of large arteries are the hallmark of this disease. The estimated prevalence of all Ehlers–Danlos syndromes (six subtypes) ranges between 1:10,000 and 1:25,000, and the vascular type represents about 5 to 10 percent of all cases.²³ Mutations in the COL3A1 gene coding for type III procollagen (mainly located in skin, vessel walls, and hollow organs) cause the phenotype, with no genotype–phenotype correlation.

Cardinal clinical findings are: facial appearance (emaciated face with prominent zygomata and sunken cheeks, thin nose and undefined lip edges), thin skin with visible veins and easy bruising, rupture of arteries, gravid uterus or intestine.^23,24 Diagnosis is often missed, with approximately 75 percent of patients presenting in adulthood with major vascular complications.²⁵ Median survival is 48 years and the primary cause of death is arterial dissection or rupture (80 percent), often involving the thoracic aorta.^24,25 Although vascular complications are rare in early childhood, patients with vascular Ehlers–Danlos syndrome have a 25 percent risk of experiencing a major vascular complication by the age of 20 years.^23–25 As for virtually all the orphan diseases, diagnosis is often missed until severe complications occur.

Congenitally Bicuspid Aortic Valve

Congenitally bicuspid aortic valve is a common cardiac anomaly that clusters in families and has a prevalence of 0.5 to 2 in 100 individuals among the general population.^26–28

There are three major types of aortic aneurysms associated with a congenitally bicuspid aortic valve. The most common is the fusiform aneurysm of the ascending aorta (Table 82-1), which begins above the sinotubular junction and terminates proximal to the innominate artery. The most rare type involves the aortic root and extends below the sinotubular junction, resembling aneurysms found in patients with Marfan syndrome. An intermediate type may also be observed.

In patients with a congenitally bicuspid aortic valve, aortic dilatation is a well-known phenomenon and its evolution has been recently characterized since the early phase. In these patients, aortic dilatation begins by 1 year of age and steadily progresses throughout childhood. About one-third of the patients with an initially normal aortic size develop an ascending aortic aneurysm by adulthood. Predictors of larger ascending aortic aneurysm and aortic root diameter are age and smaller body surface area, whereas valve gradient and right–left valve fusion have been found to be predictors of more rapid aneurysmal growth in the ascending aorta.²⁹ Although aortic dissection is more likely in patients with a dilated aorta, in some families dissection has been described at diameters under 5.0 cm similarly to other thoracic aortic syndromes.^11,30 Again, risk factors for malignant course are mostly unknown.

Current nonsurgical management of thoracic aortic syndromes includes serial aortic imaging along with echocardiographic studies and use of oral medications aimed at reducing mechanical stress on the aortic wall.

Imaging of the entire aorta and large vessels is performed at diagnosis and yearly thereafter unless the clinical course of the disease warrants a more frequent reassessment computed tomography (CT or MR angiography). Similarly, echocardiograms are obtained at diagnosis and subsequently every 6 months unless otherwise indicated by severity and progression of aortic valve insufficiency. Following surgery, ventricular and aortic valve function are assessed to establish a baseline for subsequent evaluations performed at 1-month follow-up and yearly thereafter.

β-adrenergic blockers (e.g., atenolol) have been a pharmacological cornerstone in the treatment of Marfan syndrome for years, only substituted by calcium channel blockers or angiotensin receptor blockers (ARBs) when β-blocker intolerance occurs (mostly due to asthma, depression, fatigue). β-blocker agents may stabilize aortic root growth, although this has not been a consistent finding.³¹ As murine models of Marfan syndrome showed the key role of TGF-β overexpression in aortic root enlargement, ACE-inhibitors (e.g., perindopril) and ARBs (e.g., losartan), known to have intrinsic anti-TGF-β activity, were tested in small cohorts of Marfan patients.³² Although observed reduction in aortic root enlargement raised hope for medical treatment for this syndrome, the small sample of the patient population investigated and the relatively short study time posed some legitimate concerns on the beneficial effect of anti-Angiotensin II medications, especially when the underlying genetic variability is taken into account. Mechanisms of action, relative potency, and duration of protection with each class of medication remain to be elucidated. The lack of circulating markers to assess short-term efficacy also affects the ability to individualize treatment. While awaiting final results of randomized clinical trials comparing Losartan with β-blockers in children and young adults with Marfan syndrome, β-adrenergic blockade remains as first line of treatment in patients with ascending aortic aneurysms.