Surgical Insights into the Functional Anatomy of the Neo-aortic Root

Abstract

Although the normal aortic and pulmonary roots have some features in common, they also exhibit important anatomical and functional differences. These differences increase when compared with surgically created neo-aortic roots in individuals with congenitally malformed hearts, with the substrates initially having been either a pulmonary root originating from the right or left ventricle, or a truncal root with variable ventricular origin. With the increasing application and success of these surgeries, our attention has now turned toward understanding late outcomes. Subsequent dilation, usually producing neo-aortic valvar regurgitation, is the most common cause of reoperation in any procedure involving a neo-aortic root. In this review, we describe and compare the detailed anatomy of the normal aortic, pulmonary, and common truncal roots, comparing them with the changes observed in the neo-aortic root following various common surgeries providing biventricular repair for congenital cardiac malformations. We focus on the dilated and dysfunctional neo-aortic root, assessing how the anatomical information relates to contemporary surgical techniques. Furthermore, we aim to review risk factors as they relate to the adverse anatomical features which become evident when the new root is dysfunctional, hoping to provide surgical insight toward optimizing its function and durability.

Keywords

aortic root aortic valve arterial switch operation bicuspid aortic valve pulmonary root pulmonary valve transposition of the great arteries Ross procedure

Introduction

The normal aortic root is a complex, dynamic three-dimensional structure which efficiently mediates antegrade systolic flow from the systemic left ventricle to the thoracic aorta, providing at the same time a barrier toward retrograde flow in diastole.¹ An understanding of the interplay between its anatomy and resulting hemodynamics was framed as early as 1487 by Leonardo.² Accelerated by the recent availability of advanced imaging technology and surgical techniques, the intricacies of its form and function have now further been elucidated.^1,3,4 So as to provide a therapeutic option for individuals with various congenitally malformed hearts, it is now possible surgically to create a neo-aortic root. This is commonly created from the native pulmonary root, most often with the root taking its normal origin from the right ventricle, but less commonly when arising from the left ventricle in the setting of transposition. Sometimes this neo-aortic root is created from a common truncal root. Currently, however, there is limited understanding of the similarities and differences between the normal aortic root and its therapeutic alternatives. For example, while creation of the pulmonary autograft was initially described in 1967,⁵ not until the upsurge in technical optimization in the 1990s was emphasis placed on its significant differences with the normal aortic root.⁶ Similar evolutions in understanding relate to the changes induced in the morphology of the root by the arterial switch procedure, and in the conversion of the common arterial root into a neo-aortic root.^7,8

Procedures resulting in a neo-aortic root have now become important options in the surgical armamentarium for children and young adults with various forms of congenital heart disease. With the increasing use of these procedures, our attention has now turned toward understanding late outcomes.^9–11 In this regard, root dilation and valvar regurgitation are the common causes for reoperation. In this review, we aim to describe and compare the detailed anatomy of the normal aortic, pulmonary, and truncal roots. This understanding will be compared with the anatomy of the neo-aortic root following various common surgeries for congenital heart disease resulting in biventricular repair, focusing on the dilated and dysfunctional root, and assessing how this relates to contemporary surgical techniques. Furthermore, we aim to review risk factors as they relate to the observed adverse anatomical features, thus providing surgical insight toward optimizing the function and durability of the neo-aortic root.

Functional Anatomy of the Arterial Roots

The arterial root, whether be it aortic, pulmonary, or truncal, is composed of leaflets, sinuses, and interleaflet triangles. The leaflets attach in semilunar fashion. At the zenith of attachment near the sinutubular junction, the adjacent leaflets meet together to form the commissures. At the nadir of their attachments, a plane can be created to form a virtual basal ring. Considered in three dimensions, the leaflets produce a crown-like configuration.^1,12 The normal three-leaflet design produces a valve which can open and close with little change in the length of the leaflets, and with minimized leaflet stress.¹³

There is no anatomical “annulus.” One entity often described in this fashion is the virtual basal ring. As stated above, this is an imaginary plane created by joining the lowest points of attachment of the leaflets. This dynamic plane is eccentric-shaped in diastole, becoming more circular as its area increases in systole.^3,12,14 These conformational changes may create an average area-derived diameter difference of up to 2 mm between diastole and systole in the normal aortic root. This difference becomes relevant when considering the sizing of valvar replacements, with more dramatic changes seen in the settings of the various forms of bileaflet aortic valves.¹⁵ Within the aortic root, the minor axis of its virtual basal ring spans between the muscular interventricular septum and aortic-mitral fibrous curtain.¹⁶ Diastolic eccentricity of this plane commonly is lost when the valve becomes incompetent in the setting of dilation.¹⁴ In the pulmonary root, limited studies have suggested increased eccentricity and dynamic change of its virtual basal plane throughout the cardiac cycle when compared to the better established changes in the aortic root.^17,18 We are unaware of any explorations of these changes as seen in the truncal root, which is almost always found in the presence of an outlet ventricular septal defect. In this setting, the integrity of the basal ring support of the truncal root is often compromised.

The sinutubular junction forms the distal boundary of the root, separating the arterial root from its subsequent arterial component, whether pulmonary trunk, ascending aorta, or common arterial trunk. Recent morphometric studies have highlighted that the “waist” visualized at the distal boundary of the arterial root is distal and distinct to the plane connecting the leaflet commissures, and therefore subtly different to the plane of the sinutubular junction (Figures 1C and 2).^19,20 This narrowest dimension, nonetheless, is often the plane measured and reported as that of the sinutubular junction.

The Anatomical Myocardial-Arterial Junction

A complete and circular anatomical junction can be discerned within the proximal third of the pulmonary root, forming the boundary between the infundibular myocardium and the fibroelastic arterial tissues (Figure 2). This junction is incomplete in the aortic and truncal roots (Figures 3 and 4). Its extent and location depend on the orientation of the valvar sinuses relative to the ventricles, the number of sinuses, and in the truncal root the relationship to the outlet ventricular septal defect. In the aortic root, the junction is present at the base of the leaflets supported by the coronary aortic sinuses, and the intervening interleaflet triangle.^21,22 The remaining circumference of the base of the aortic root is supported by the fibrous tissues making up the central fibrous body and the aortic-mitral fibrous curtain.¹² These differences in the extent of the myocardial-arterial junction become important when assessing the ability of using the normal pulmonary root as an autograft replacement for the diseased aortic root.

The Hemodynamic Ventriculo-Arterial Junction

The crown-like configuration of the leaflets forms the underlying infrastructure for the hemodynamic ventriculo-arterial junction present during diastole.²³ The nodules of Arantius are at the center of the zones of coaptation between the leaflets, with the surface area of coaptation increasing laterally to form the lunules (Figure 1). Taken together, the surface area of coaptation comprises approximately two-fifths of the ventricular-facing surface of the leaflets.²⁴ For the aortic root, the midline length of central coaptation is approximately 3 to 4 mm in adults, with an effective height, or height from the plane of the virtual basal ring, to the central tip of the coaptating leaflets reported at between 8 and 9 mm.²⁵ This length increases laterally, within the midportion of each lunule, subsequently decreasing laterally when traced to its commissure.¹² Fenestrations may be present in the coaptating lunules or the noncoapting surface of the leaflet, being reported in between two-fifths and half of normal aortic valves in a recent autopsy-based study. Such fenestrations increase in frequency with age.¹⁹ When occurring in clusters, they may impair normal mechanics.¹⁹ The proximal portion of the leaflet not involved in coaptation can be referred to as its belly. The ventricular-facing surface of each belly then forms the diastolic hemodynamic interface between the ventricle and its arterial trunk.²³ Anatomically, this is reflected in the production of peripheral peaks under each commissure, bordered laterally by the interleaflet triangle, and medially by the lateral bellies of the adjacent leaflets. Centrally, a fourth and shorter peak is formed beneath the area of central coaptation between the three nodules of Arantius. The height of the central peak can be quantified as the difference between the effective height and the coaptation length.¹² Such a qualitative description also holds true for the pulmonary root, although due to the reduced hemodynamic forces, all of these structures are more delicate and flexible.²⁶ For example, the thickness of the leaflets in the normal adult aortic root is approximately 0.6 mm, compared with 0.4 mm in those belonging to the pulmonary root (Figure 4A and D).²⁷ In addition, there is less conspicuous condensation of elastic fibers within the interleaflet triangles of the pulmonary root, likely related to decreased dependency on generating elastic properties during end-diastole in order to aid opening of its arterial valve.^6,27,28 These general descriptions also hold true for the truncal root. Unlike the normal aortic or pulmonary root, however, the truncal root in a minority of individuals possesses four or two sinuses and leaflets, although the majority of truncal roots also have three sinuses and leaflets. In those with four leaflets, we have noted the common occurrence of central crowding of the nodules of Arantius, commonly compromising the central zone of coaptation.

The Normal Aortic Root

The right and noncoronary sinuses are typically larger than the left coronary sinus in the normal trisinuate aortic root.^1,29,30 This results in the interleaflet triangle between the left and noncoronary leaflets having a shorter commissural height above its virtual basal ring.²⁵ This creates an inferior tilt angle directed posterior and leftward of approximately 15° to 20° toward the sinutubular junction in diastole when compared with the plane of the virtual basal ring.¹ The aortic root increases in volume during systole from proximal to distal, with a resulting decrease in this tilt angle, transitioning from a cone during diastole to a cylindrical shape through systole.^31,32 Aortic valvar opening is initiated prior to ejection related to this expansion, with a clover-shaped opening of the valvar leaflets observed in systole secondary to the larger free edges of the leaflets compared with the commissural areas.³² There are four distinct asymmetric modes of dynamic deformation. The first is the circumferential deformations observed at both the basal ring and the plane of the sinutubular junction. The others are longitudinal deformation, shear strain deformation, and torsional deformation.³³ These dynamic anatomical changes combine to optimize efficient systolic stroke volume, while minimizing the stresses placed on the leaflets throughout the cardiac cycle. This is accomplished by optimizing the sharing of the load on the leaflets during diastole and minimizing any transvalvar turbulence.³⁴

As measured by the length of its free margin, the right coronary leaflet is often the broadest leaflet. In contrast, it is commonly the shortest in midline length as measured by its geometric height.^12,25 The dimensions of the leaflets may change, to a limited and variable extent, in parallel to dilation of the aortic root.¹³ Positioned on the base of the left ventricle, the aortic root is typically wedged between the atrioventricular valves such that its noncoronary leaflet aligns with the posteroinferiorly positioned muscular buttress of the atrial septum. The shortest dimension, or minor axis, of its virtual basal ring in diastole is typically parallel to the midline of the right coronary leaflet, with the major axis parallel to that of the left and noncoronary leaflets. Proper central coaptation of the leaflets then depends on their asymmetries,³ with the coaptation center commonly located near the geometric center of the aortic root.¹⁹

There is significant normal variation in the tilt angle, wedging and rotational position of the aortic root relative to the base of the left ventricle.^35,36 This changes the orientation of the individual sinuses relative to the dimensions of the virtual basal plane. Knowledge of this feature is important when teasing out mechanisms for valvar dysfunction.³ There is also a direct relationship between the proportion of myocardial and fibrous support of the aortic root relative to its rotational position.^35,37 This additionally correlates with the amount of ventricular myocardium incorporated into the bases of the coronary sinuses, along with that found at the base of the intercoronary interleaflet triangle.²² Such variation may impact valvar function, with prolapse shown more commonly to involve the valvar leaflets supported by ventricular myocardium.³⁸

Variability in the rotational position also dictates the resulting hemodynamics across the aortic valve, and the tissue biomechanics experienced by the aortic root, the leaflets, and the ascending aorta. The clockwise-positioned aortic root, as viewed by the imager from the ventricular aspect, with the right coronary sinus positioned further anteriorly, has been identified to create less favorable hemodynamics and resulting wall shear stress.³⁹ These features result in a propensity for dilation of the aortic root and ascending aorta in those with a normal trileaflet valve within a clockwise-positioned root.^40,41

It is the origin of the coronary arteries from the right and left facing aortic sinuses that underscores one major difference between the aortic as opposed to the pulmonary and truncal roots. The arteries show variability in both their circumferential and longitudinal position within their respective sinuses. The right coronary artery more commonly originates from the posterior third of its supporting sinus, and is closer to the sinutubular junction, while the left coronary artery often originates centrally.^19,22,42 Normal position of, and flow through, the coronary arteries are established as being associated with an advantageous increase in shear stress and washout near the bases of their supporting sinuses, and with slightly improved opening of the valvar leaflets in systole.⁴³

Measures taken for the aortic root in the clinical setting have revealed relatively equal dimensions between its virtual basal and its sinutubular junction.⁴⁴ These relationships are in keeping with autopsy-based morphometric studies, which show the largest cross-sectional area at the plane of the coaptation center, followed by the tubular plane immediately distal to the sinutubular junction, with the virtual basal ring being smallest.¹⁹ These relative dimensions are markedly different from the normal pulmonary root, where its virtual basal ring plane is often significantly larger than its tubular plane.²⁰ These differences are seemingly unappreciated by clinicians when comparing the dimensions of the roots when considering insertion of a pulmonary autograft. The proportions of the planes as found in the normal aortic root, which is exposed to systemic pressures, result in optimal coaptation length of the leaflets, with less stress on them and the adjacent aortic walls throughout the cardiac cycle.⁴⁴

The Normal Pulmonary Root

Right ventricular infundibular myocardium is incorporated into the bases of all three of the pulmonary valvar sinuses and supports the bases of each interleaflet triangle (Figures 2 and 4D-F). This myocardium, being relatively thin, provides a more delicate attachment of the pulmonary root to the right ventricle, compared with the thicker, albeit incomplete, myocardial support provided for the aortic root (compare Figure 4A-C to D-F).⁶ The presence of the infundibular sleeve lifts the pulmonary root away from the base of the right ventricle, positioning the plane of its virtual basal ring anteriorly and leftward relative to the left coronary sinus of the more deeply wedged and inferiorly positioned aortic root. This also positions the aortic and pulmonary roots at orthogonal angles relative to each other. The wall of the infundibular sleeve is separated by extracavitary fibroadipose tissues from the adjacent walls of the aortic valvar sinuses (Figure 4E and F).^6,45 This intimate relationship between the arterial roots means that, despite its symmetrical myocardial support, the pulmonary root is influenced throughout the cardiac cycle by the higher pressures generated in the aortic root.²⁷ A collagenous connection, commonly referred to as the conus tendon, can be found between the posteriorly positioned interleaflet triangle of the pulmonary root with that between the anterior facing coronary aortic sinuses.⁶ These positional relationships, however, will vary depending upon the variable rotational positions of both the aortic and pulmonary roots. The overall result of the structural and hemodynamic differences between the roots, nonetheless, is that the virtual basal ring plane of the pulmonary root exhibits more significant eccentricity throughout the cardiac cycle when compared to that of the aortic root, with more significant dynamic changes in its dimensions.¹⁸ This corresponds to more significant heterogeneity in the sizes of its leaflets and sinuses compared with the aortic root.^19,20 Although its leaflets are relatively thinner, with less conspicuous condensation of elastic fibers within the interleaflet triangles,^6,27 both arterial roots have similar mechanical characteristics.²⁷

Similar dynamic changes have been described in the shape and expansion of the pulmonary sinuses, although to a smaller degree. The volumes of the sinuses expand only half as much as do those of the aortic sinuses during systole, with less variable change between the expansion at the plane of the commissures compared with the more distal sinutubular junction, or the area now recognized as the tubular “waist.”⁴⁶ These collective changes, in light of the differences noted in the planes of the virtual basal rings, reinforce the other significant differences which exist between the normal pulmonary and aortic roots. In contrast to the aortic root, the basal ring of the pulmonary root is approximately one and a half times the area of its plane at the sinutubular junction, with the coaptation center less commonly in close vicinity to the geometric center.²⁰ While these differences often seem to result in a nonpathological prolapse of the pulmonary leaflets below the plane of the virtual basal ring, such prolapse does not compromise coaptation of the leaflets.

The Pulmonary Root in Transposition

In the setting of discordant ventriculo-arterial connections, the pulmonary root takes its origin from the morphologically left ventricle. When found with concordant atrioventricular connections, and with right-handed ventricular topology, the combination which can be considered to represent “regular” transposition, the pulmonary root originates from the left ventricle, and is usually positioned leftward and posterior relative to the aortic root, which arises from the right ventricle. Almost always in this setting there is mitral-to-pulmonary valvar fibrous continuity, and it is the pulmonary root that is now partly supported by fibrous and myocardial tissues. A ventricular septal defect is associated with approximately half of cases, most commonly perimembranous and opening to the outlet of the right ventricle. Less commonly, the outlet defect has completely muscular borders. The geographical location of the defect results in adjacency to, and often distortion of, the plane of the virtual basal ring of the pulmonary root. There is more significant variability in the rotational position of this pulmonary root relative to the base of the left ventricle in the setting of transposition, when compared with the variability seen in the normal aortic root, as detailed later in this review.^35,47 The pulmonary root, furthermore, with these segmental combinations, is exposed to systemic pressures. The common occurrence of a ventricular septal defect, lack of coronary arteries, and more marked rotational position of the pulmonary root relative to the left ventricle, represent the main structural differences from the normal aortic root. Despite these differences seemingly being minimal when compared with the normal aortic root, the pulmonary root in regular transposition is, on average, dilated to a limited degree within the first weeks of life. This dilation may be more prevalent in the setting of a ventricular septal defect. This suggests that the structural differences have important hemodynamic implications.⁴⁸ The arterial switch operation is designed to transform the pulmonary root to the systemic circulation, converting it to become the new aortic root, and thus restoring the appropriate in-series circulation.

Common Arterial Truncal Root

The solitary ventriculo-arterial junction, which is the phenotypic feature of the common arterial trunk, will be required exclusively to support the systemic circulation following surgical repair. It is deserving of specific attention. The native truncal root, however, shows marked variability. As we have already described, although usually being trisinuate, it can be quadrisinuate or bisinuate, with the root usually supporting an equivalent number of leaflets. The presence or absence of fusion between leaflets, nonetheless, will likely determine the function of the valve within its root. As with the bileaflet aortic valve, the extent of leaflet fusion will inversely correspond to the height of the involved interleaflet triangle.¹²

As previously mentioned, an interventricular communication is typically present in those with common arterial trunk. It is the ventricular component of an arterial-ventricular defect, and usually is situated between the ventricular outlets.⁴⁹ Most commonly, its posteroinferior border is muscular, but less commonly it can extend to become perimembranous. Rarely it can be confluent with the ventricular component of an atrioventricular septal defect. The truncal root has variable relationship to the ventricles, whether originating primarily from the right or left ventricle, or equally committed to both ventricles. The combination of these two features will dictate the support of the virtual basal ring of the truncal root, and the geometry of the pathway to be created as a baffle from the left ventricle to truncal root during its repair. The pulmonary arteries can originate proximal to, at the level of, or distal to the sinutubular junction of the truncal root, with further variation according to whether they share a confluence from the root. This relationship may disturb the integrity of the sinutubular junction. The coronary arteries, unlike the situation for the aortic root, have no uniform origin. In the truncal root with four sinuses, the arteries can arise from opposite sinuses, but again, there is no uniformity in this arrangement. Similarly, when the truncal root is trisinuate, the individual arteries show marked variability in their origins. Solitary coronary arteries are far from rare.⁵⁰

Functional Anatomy of the Neo-Aortic Root

Following the Arterial Switch Operation

There is a rapid increase in the dimensions of the neo-aortic root over the first year following the arterial switch procedure. This is followed by a more linear average increase of approximately 0.6 mm per year through adolescence and early adulthood.⁵¹ Up to one-tenth of patients are known to develop moderate or greater neo-aortic valvar regurgitation and may require reparative surgery on the root or its contained valve.^9,52–54 Certain risk factors for progression of both dilation of the root and neo-aortic valvar regurgitation have been identified, such as history of repair of ventricular septal defect, and older age at the time of the arterial switch operation. Speculation remains, however, with regard to potential anatomical and hemodynamic aberrations which may further portend risk. Aberrations proposed include a diminished and altered distribution of collagen, and a less-firmly embedded root, especially in the setting of an outlet ventricular septal defect.

Several groups have evaluated different surgical techniques used to reimplant the coronary arteries, assessing their relationships to the occurrence of neo-aortic root dilation. The results have been discrepant.^55,56 Our own surgical experience suggests a potential relationship of high reimplantation of the coronary arteries within the proximal ascending aorta. We infer that this approach, by disturbing the native sinutubular junction, especially when adjacent to a commissure, may adversely alter the postoperative hemodynamics. Based on this inference, we believe that understanding the cause for any deficiency of coaptation of the leaflets, and identifying the means of rectifying this problem surgically, depends on distinguishing the native sinutubular junction, which is the plane connecting the commissures, from what appears externally to represent the sinutubular junction at the site of anastomosis between the native pulmonary trunk and the intrapericardial ascending aorta (Figure 5). In our experience, the site of anastomosis is often mistaken for the native sinutubular junction. This mistake may confound the numerous investigations assessing coronary reimplantation techniques relating to the long-term integrity of the neo-aortic root.

Evaluation of postsurgical hemodynamics in small cohorts using 4D flow cardiac magnetic resonance after the arterial switch procedure demonstrates increased helical flow patterns in the neo-aortic root and proximal ascending aorta in those with mild root dilation, which itself may perpetuate ongoing dilation.⁵⁷ As referred to earlier, there are several surgical parameters whose effects are not yet well understood. In silico methods, namely computational fluid dynamics and fluid structure interaction (FSI) simulations are becoming increasingly common to evaluate such surgical variables. One such idealized computational fluid dynamic study suggested that during reimplantation of the coronary arteries post arterial switch procedure, surgeons should feel relatively free to vary the coronary artery outlet angle as variations in the range 60° to 120° led to insignificant changes in pressure and velocity distributions.⁵⁸ A recent computational fluid dynamic study to evaluate the cause and effect of neo-aortic root dilation after the arterial switch operation in patient-specific geometries revealed that severe dilation shows disrupted patterns of the wall shear stress distribution due to the formation of flow recirculation zones in the neo-aortic root.⁵⁹ Such zones are believed to be caused in patients with smaller aortic arch angles,⁶⁰ with studies reporting that acute angulation of the aortic arch predisposes a patient to ascending aortic dilatation and neo-aortic regurgitation late after the arterial switch operation.⁶¹

It is incorrect surgically to consider the pulmonary root in the setting of discordant ventriculo-arterial connections as simply replacing its normal aortic counterpart. An assessment by cardiac magnetic resonance in a cohort of adolescents and young adults who underwent the arterial switch operation in the neonatal period revealed marked variability in the rotational position of the neo-aortic root relative to the base of the left ventricle. The rotational angle between the center of the nonadjacent leaflet relative to the midline of the roof of the inferoseptal recess within the left ventricle ranged from −52° to +78°.⁴⁷ This variation is much more dramatic than that exhibited in the normal aortic root, which ranges between −32° and +45°.³⁵ Individuals showing the extremes of this increased variability proved to have increased risk for both neo-aortic root dilation and neo-aortic valvar regurgitation.⁴⁷ These adverse relationships are further supported by similar clinical studies in the normal aortic root,^40,41 and by fluid-structure interaction models.³⁹

Following the Ross Procedure or Insertion of a Pulmonary Autograft

The feasibility of harvesting the normal pulmonary root to be used as an autograft relates to its support by the free-standing muscular infundibular sleeve. As already discussed, this feature then means that infundibular myocardium is incorporated in annular fashion within the bases of all three pulmonary valvar sinuses, and at the bases of all three interleaflet triangles.^6,26 During removal of the root, it follows that the risk of damaging the first septal perforating coronary artery is mitigated if the surgeon is able to identify the inferior boundary of the infundibular sleeve.²⁶ Similar care must be taken in regards to the left main and proximal left anterior descending coronary arteries, both which invariably have an intimate relationship to the posterior aspect of the pulmonary root.⁶² Once the autograft has been harvested, and the supporting subpulmonary infundibulum devascularized, it also follows that its myocardial base will become necrotic (Figure 6C and D). This devitalization process will not spare the myocardial crescents incorporated into each sinus. This sets up the potential for its dilation at the level of the virtual basal ring.⁶ Recognizing this possibility, contemporary techniques recommend trimming the infundibular sleeve to a length of no more than 2 to 3 mm. Implantation of such a shortened sleeve within the left ventricular outflow tract aims to allow the site of the virtual basal ring of the removed native aortic valve to provide the necessary support for the autograft.⁶³

Once translocated, the pulmonary root is immediately exposed to systemic pressures. It is hardly surprising that the more delicate pulmonary root, with decreased elastic properties compared with the aortic root, and resulting less compliant response toward physiological systemic pressures,^27,28,64 may quickly undergo adverse changes when abruptly and persistently exposed to higher pressures. In fact, up to three-fifths of the extent of dilation observed at a follow-up of one year has been reported to be present prior to discharge following the Ross procedure.⁶⁵ Management immediately subsequent to the operation, therefore, should include tight control of blood pressure, thus permitting a slow compensatory adaptation as the pulmonary root becomes the neo-aortic root. This process itself involves remodeling of the sinusal walls and the valvar leaflets. The leaflets increase their thickness over time, eventually becoming comparable with those of the normal aortic valve. This is the consequence of apposition of fibrous tissue on their ventricular surfaces.⁶⁶ Aiming best to facilitate such adaptive change, and appreciating the asymmetric sizes and dynamic expansion of the normal aortic and pulmonary sinuses, some have suggested positioning the right- and anterior-facing sinuses of the pulmonary autograft in the initial location of the right and left coronary aortic sinuses.⁴⁶

Excessive fluttering of the leaflets during systole has been suggested to lead to valvar deterioration, increasing the repetitive load experienced by a valve over time. Simulations of the procedure have demonstrated the potential benefits of commissural plication when creating neo-sinuses so as to produce more favorable leaflet kinematics.⁶⁷ Others have suggested mitigating dilation by encasing the autograft in a Dacron graft, or a personalized externalized neo-aortic root support.⁶³ This approach, while effective in those who have adult-sized autografts, may not be advisable in those requiring ongoing growth. Whether this approach limits the natural dynamic nature of the arterial root, with resulting impairment in valvar function, also remains to be understood. An alternative approach maintains autologous aortic tissue to use as external support to the autograft, in combination with external support of the virtual basal ring, commonly referred to as an external annuloplasty combined with an interposition graft.⁶³ The sinutubular junction in the normal aortic root is approximately up to one-sixth larger in area compared with that of the virtual basal ring.¹⁹ This contrasts with the pulmonary root, where the area of the virtual basal ring is almost one and a half times as large as its sinutubular junction.²⁰ Modifying and supporting these two planes becomes paramount, aiming better to replicate the relationships accounted for in the normal aortic root (Figure 6A and B).⁶³ Both described approaches, whether graft or native aortic root inclusion, accomplish this feat. Long-term studies are necessary; however, to better understand which approach provides the most durable neo-aortic root with favorable hemodynamics.

In younger children, where autograft reenforcement may be prohibited, reoperation due to failure of the autograft occurs at rates of up to approximately 3% per year.¹⁰ In young to middle age adults, where reenforcement is often possible, rates of reintervention on the autograft are not significantly improved. They range from 1% to 2% per year,⁶⁸ and up to 20% within 25 years following the procedure.¹¹ Outside of the presence of an autograft supporting a bileaflet valve, or evidence of significant regurgitation prior to its insertion,⁶⁹ structural and hemodynamic risk factors for its dilation, and valvar regurgitation, have yet to be identified. Unlike the situation in the setting of the arterial switch procedure, the coronary arteries are routinely translocated into the holes made into the autograft. Most commonly they do not distort the sinutubular junction and distal anastomosis.

Patient-specific simulations may be an effective tool to optimize surgical techniques and determine patient-specific risk of aneurysm formation after a Ross procedure. A case study performing a virtual Ross operation was able to identify areas of increased stress in the pulmonary autograft. Reinforcement in these high-stress areas to support the autograft wall may potentially reduce future reoperations. Surgeons could also benefit greatly from patient-specific simulations of growth and remodeling of the autograft after the Ross procedure.⁷⁰ However, current models in this space are only limited to theoretical estimates.

Following Repair of Common Arterial Trunk

The variable fusion of leaflets seen in the truncal root changes the orientation of the opening area of the valve relative to the plane of the sinutubular junction. Even in the absence of valvar stenosis, this will impact the hemodynamics and resulting tissue biomechanics imposed upon the neo-aortic root and proximal ascending aortic wall, impacting the risk for aortic dilation.⁷¹ It is not surprising, therefore, that dilation of the neo-aortic root following repair of common arterial trunk, and subsequent interventions on the neo-aortic valve, are greater in those deemed to have bileaflet or quadrileaflet valves.⁷²

As is the case for those operating on the aortic valve with two leaflets, strategies for repair should include specific evaluation of the morphology of both sinuses and leaflets, preparing the way to create symmetrical leaflets, irrespective of their number, which coapt snugly and have an unrestricted opening (Figure 7).³ This may include maintaining the native leaflet configuration, suturing a hypoplastic or prolapsing leaflet to an adjacent leaflet, or single leaflet resection with reduction of the virtual basal ring.⁷³ In the older child and adult, this often includes stabilization of both the virtual basal ring and the sinutubular junction. The sinutubular junction is not uncommonly dilated in the older child or adult with a history of repair of their common arterial trunk in infancy. This feature may partly be a ramification of the origin of the pulmonary arteries relative to the truncal root.

Conclusion

There is an increasing number of patients now living with arterial roots supported by the left ventricle that were not initially aortic. Most such neo-aortic roots were initially pulmonary, but some were present in the setting of common arterial trunks. In this review, we aimed to compare and contrast the anatomy and function of the normal aortic, pulmonary, and truncal roots with those that were not native to the morphologically left ventricle. This has permitted us to assess the contemporary outcomes, and established risks related to the neo-aortic root and its valvar dysfunction, along with the need for surgical reintervention. While other nonanatomical variables certainly impact the outcomes of these patients, we have sought to draw inferences pointing to potential risks and, when possible, to suggest surgical modifications which may lead to improved durability of the neo-aortic root (Table 1).

Table 1.

Summary of Surgical Recommendations for the Neo-Aortic Root.

General surgical considerations:
Variation in the rotational orientation of the individual sinuses relative to the diastolic dimensions of the virtual basal ring plane may provide information on the mechanism of valvar incompetency and guide strategies to reduce the virtual basal ring dimensions. Rotational orientation may additionally impact the relative fibrous versus muscular support of the root. Significant discrepancies in the free margin length tend to associate with leaflet prolapse and valvar incompetency and may benefit from central leaflet plication of the elongated free margin leaflet. The final comparison of the free margin length has to be determined after any annuloplasty or valvar-sparing root replacement. Aiming for an approximately 1:1 relationship between the neo-aortic root virtual basal ring and sinutubular junction plane best replicates the well-functioning aortic root. When addressing an outlet ventricular septal defect, care should be taken to avoid attaching the patch immediately adjacent to, or along the attachment line of the semilunar valvar leaflets. However, such avoidance may not be possible with deficiency of the posterior caudal limb of the septomarginal trabeculation and/or deficiency of the outlet septum. Short- and long-term postoperative management should include optimal blood pressure management to minimize stress on the neo-aortic root and its valve. This is especially important early following leaflet suture repair and early and late after the Ross procedures to allow the pulmonary autograft to slowly acclimate to systemic pressures.
Considerations for the Ross procedure:
Minimizing the devitalized subpulmonary infundibulum (<2-3 mm length) with implantation into the left ventricular outflow tract and external stabilization of the virtual basal ring plane during the Ross procedure may optimize the long-term neo-aortic root function. Positioning the right- and anterior-facing sinuses of the pulmonary autograft in the initial location of the right and left coronary aortic sinuses may best appreciate the coinciding asymmetries of the normal arterial roots. External reenforcement of the autograft may be beneficial, preventing root dilation and achieving the desired 1:1 normalized dimensions between the virtual basal ring and sinutubular junction planes in replicating the normal aortic root.
Considerations for the arterial switch procedure:
Reimplantation of the coronary arteries within the sinuses (what is referred to as the “closed technique,” where the distal suture line of the neo-aorta is completed before the buttons implanted) and preserving the native sinutubular junction and valvar commissures may preserve integrity of the native sinutubular junction. Minimizing the proximal main pulmonary artery maintained with the neo-aortic root, and maximizing the translocated length of the ascending aorta may best balance placing the surgical anastomosis as close to the native sinutubular junction as possible to help prevent significant native sinutubular junction effacement, while ensuring less acute aortic arch angles.
Considerations for repair of common arterial trunk:
Attempts should be taken to restore/create relative symmetry of the truncal root and its leaflets. Attempts should be made to restore the involved sinuses and integrity of the sinutubular junction during separation of the pulmonary arteries. Careful evaluation of the closure suture line is necessary to maximize preservation of the sinutubular junction. In the presence of leaflet fusion, dysplasia or hypoplasia with significant valvar incompetency, a surgical plan should consider the ability to improve the symmetry of the valve. This may involve creating leaflet fusion or leaflet resection and/or replacement with virtual basal ring reduction.

Footnotes

Acknowledgment

The authors are thankful to the European Synchrotron Radiation Facility for the open access use of the complete scan at 19.89 µm of the heart of the body donor S-20-29(Version 1), which was used for interrogation to support and illustrate the anatomy displayed in .⁷⁴

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Justin Tretter is a consultant for Cara Medical, Ltd.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Georgios Belitsis

Robert H. Anderson

Justin T. Tretter

References

Tretter

Spicer

Mori

Chikkabyrappa

Redington

Anderson

. The significance of the interleaflet triangles in determining the morphology of congenitally abnormal aortic valves: implications for noninvasive imaging and surgical management. J Am Soc Echocardiogr. 2016;29(12):1131–1143.

Robicsek

. Leonardo da Vinci and the sinuses of Valsalva. Ann Thorac Surg. 1991;52(2):328–335.

Tretter

Burbano-Vera

Najm

. Multi-modality imaging evaluation and pre-surgical planning for aortic valve-sparing operations in patients with aortic root aneurysm. Ann Cardiothorac Surg. 2023;12(4):295–317.

Tretter

Izawa

Spicer

, et al. Understanding the aortic root using computed tomographic assessment: a potential pathway to improved customized surgical repair. Circ Cardiovasc Imaging. 2021;14(11):e013134.

Ross

. Replacement of aortic and mitral valves with a pulmonary autograft. Lancet. 1967;2(7523):956–958.

Hokken

Bartelings

Bogers

Gittenberger-de Groot

. Morphology of the pulmonary and aortic roots with regard to the pulmonary autograft procedure. J Thorac Cardiovasc Surg. 1997;113(3):453–461.

Dodge-Khatami

Mavroudis

Jacobs

. Past, present, and future of the arterial switch operation: historical review. Cardiol Young. 2012;22(6):724–731.

Mavroudis

Jonas

Bove

. Personal glimpses into the evolution of truncus arteriosus repair. World J Pediatr Congenit Heart Surg. 2015;6(2):226–238.

Zhu

MZL

Fricke

Buratto

Chowdhuri

Brizard

Konstantinov

. Outcomes of neo-aortic valve and root surgery late after arterial switch operation. J Thorac Cardiovasc Surg. 2024;167(4):1391–1401.e3.

10.

Moroi

Bacha

Kalfa

. The Ross procedure in children: a systematic review. Ann Cardiothorac Surg. 2021;10(4):420–432.

11.

Notenboom

Melina

Veen

, et al. Long-Term clinical and echocardiographic outcomes following the Ross procedure: a post hoc analysis of a randomized clinical trial. JAMA Cardiol. 2024;9(1):6–14.

12.

Tretter

Spicer

Franklin

RCG

, et al. Expert consensus statement: anatomy, imaging, and Nomenclature of congenital aortic root malformations. Ann Thorac Surg. 2023;116(1):6–16.

13.

Thubrikar

Labrosse

Zehr

Robicsek

Gong

Fowler

. Aortic root dilatation may alter the dimensions of the valve leaflets. Eur J Cardiothorac Surg. 2005;28(6):850–855.

14.

Petersen

Voigtländer

Schofer

, et al. Geometric changes in the aortic valve annulus during the cardiac cycle: impact on aortic valve repair. Eur J Cardiothorac Surg. 2018;54(3):441–445.

15.

Boccalini

Bons

van den Hoven

, et al. Bicuspid aortic valve annulus: assessment of geometry and size changes during the cardiac cycle as measured with a standardized method to define the annular plane. Eur Radiol. 2021;31(11):8116–8129.

16.

Suchá

Tuncay

Prakken

, et al. Does the aortic annulus undergo conformational change throughout the cardiac cycle? A systematic review. Eur Heart J Cardiovasc Imaging. 2015;16(12):1307–1317.

17.

Sun

Hao

Sebastian Kiekenap

, et al. Four-dimensional computed tomography-guided valve sizing for transcatheter pulmonary valve replacement. J Vis Exp. 2022(179). doi:10.3791/63367

18.

Schievano

Capelli

Young

, et al. Four-dimensional computed tomography: a method of assessing right ventricular outflow tract and pulmonary artery deformations throughout the cardiac cycle. Eur Radiol. 2011;21(1):36–45.

19.

Dudkiewicz

Lis

Yakovliev

, et al. Aortic root morphometry revisited—clinical implications for aortic valve interventions. Clin Anat. 2024;37(7):719–729.

20.

Lis

Krawczyk-Ożóg

Hołda

, et al. Pulmonary valve morphometry revisited: clinical implications for valvular and supravalvular interventions. Clin Anat. 2023;36(2):234–241.

21.

Tretter

Spicer

Jacobs

Anderson

. The aortic valve with two leaflets. JTCVS Open. 2022;9:89–90.

22.

Toh

Mori

Tretter

, et al. Living anatomy of the ventricular myocardial crescents supporting the coronary aortic sinuses. Semin Thorac Cardiovasc Surg. 2020;32(2):230–241.

23.

Burbano-Vera

Alfirevic

Bauer

, et al. Perioperative assessment of the hemodynamic ventriculoarterial junction of the aortic root by three-dimensional echocardiography. J Am Soc Echocardiogr. 2024;37(6):626–633.

24.

De Kerchove

Momeni

Aphram

, et al. Free margin length and coaptation surface area in normal tricuspid aortic valve: an anatomical study. Eur J Cardiothorac Surg. 2018;53(5):1040–1048.

25.

Izawa

Mori

Tretter

, et al. Normative aortic valvar measurements in adults using cardiac computed tomography- A potential guide to further sophisticate aortic valve-sparing surgery. Circ J. 2021;85(7):1059–1067.

26.

Stamm

Anderson

. Clinical anatomy of the normal pulmonary root compared with that in isolated pulmonary valvular stenosis. J Am Coll Cardiol. 1998;31(6):1420–1425.

27.

Stradins

Lacis

Ozolanta

, et al. Comparison of biomechanical and structural properties between human aortic and pulmonary valve. Eur J Cardiothorac Surg. 2004;26(3):634–639.

28.

Azadani

Chitsaz

Matthews

, et al. Biomechanical comparison of human pulmonary and aortic roots. Eur J Cardiothorac Surg. 2012;41(5):1111–1116.

29.

Suzuki

Mori

Izawa

, et al. Three-dimensional volumetric measurement of the aortic root compared to standard two-dimensional measurements using cardiac computed tomography. Clin Anat. 2021;34(3):333–341.

30.

Berdajs

Lajos

Turina

. The anatomy of the aortic root. Cardiovasc Surg. 2002;10(4):320–327.

31.

Lansac

Lim

Shomura

, et al. A four-dimensional study of the aortic root dynamics. Eur J Cardiothorac Surg. 2002;22(4):497–503.

32.

Lansac

Lim

Shomura

, et al. Aortic valve opening and closure: the clover dynamics. Ann Cardiothorac Surg. 2019;8(3):351–361.

33.

Dagum

Green

Nistal

, et al. Deformational dynamics of the aortic root: modes and physiologic determinants. Circulation. 1999;100(19 Suppl):II54–II62.

34.

Cheng

Dagum

Miller

. Aortic root dynamics and surgery: from craft to science. Philos Trans R Soc Lond B Biol Sci. 2007;362(1484):1407–1419.

35.

Tretter

Mori

Saremi

, et al. Variations in rotation of the aortic root and membranous septum with implications for transcatheter valve implantation. Heart. 2018;104(12):999–1005.

36.

Mori

Anderson

Takaya

, et al. The association between wedging of the aorta and cardiac structural anatomy as revealed using multidetector-row computed tomography. J Anat. 2017;231(1):110–120.

37.

Amofa

Mori

Toh

, et al. The rotational position of the aortic root related to its underlying ventricular support. Clin Anat. 2019;32(8):1107–1117.

38.

Jahanyar

Tsai

Arabkhani

, et al. Functional and pathomorphological anatomy of the aortic valve and root for aortic valve sparing surgery in tricuspid and bicuspid aortic valves. Ann Cardiothorac Surg. 2023;12(3):179–193.

39.

Sundström

Jonnagiri

Gutmark-Little

, et al. Effects of normal variation in the rotational position of the aortic root on hemodynamics and tissue biomechanics of the thoracic aorta. Cardiovasc Eng Technol. 2020;11(1):47–58.

40.

Powell

Almeneisi

Alsaied

, et al. Rotational position of the aortic root is associated with increased aortic dimensions in Marfan and Loeys-Dietz syndrome. Pediatr Cardiol. 2021;42(5):1157–1161.

41.

Saremi

Cen

Tayari

, et al. A correlative study of aortic valve rotation angle and thoracic aortic sizes using ECG gated CT angiography. Eur J Radiol. 2017;89:60–66.

42.

Pejković

Krajnc

Anderhuber

. Anatomical variations of coronary ostia, aortocoronary angles and angles of division of the left coronary artery of the human heart. J Int Med Res. 2008;36(5):914–922.

43.

Moore

Dasi

. Coronary flow impacts aortic leaflet mechanics and aortic Sinus hemodynamics. Ann Biomed Eng. 2015;43(9):2231–2241.

44.

Marom

Halevi

Haj-Ali

Rosenfeld

Schäfers

Raanani

. Numerical model of the aortic root and valve: optimization of graft size and sinotubular junction to annulus ratio. J Thorac Cardiovasc Surg. 2013;146(5):1227–1231.

45.

Mori

Bradfield

Peacock

Anderson

Shivkumar

. Living anatomy of the pericardial space: a guide for imaging and interventions. JACC Clin Electrophysiol. 2021;7(12):1628–1644.

46.

Lansac

Lim

Shomura

, et al.

Aortic and pulmonary root: are their dynamics similar?

Eur J Cardiothorac Surg. 2002;21(2):268–275.

47.

Tseng

Tretter

Gao

Ollberding

Lang

. Aortic root rotational position associates with aortic valvar incompetence and aortic dilation after arterial switch operation for transposition of the great arteries. Int J Cardiovasc Imaging. 2023;39(5):1013–1021.

48.

Hourihan

Colan

Wernovsky

, et al. Growth of the aortic anastomosis, annulus, and root after the arterial switch procedure performed in infancy. Circulation. 1993;88(2):615–620.

49.

Spicer

Steffensen

. A rare presentation of common arterial trunk with intact ventricular septum. J Cardiovasc Dev Dis. 2020;7(4):43. doi:10.3390/jcdd7040043

50.

Gupta

Aggarwal

Shaw

, et al. Clarifying the anatomy of common arterial trunk: a clinical study of 70 patients. Eur Heart J Cardiovasc Imaging. 2020;21(8):914–922.

51.

van der Palen

RLF

van der Bom

Dekker

, et al. Progression of aortic root dilatation and aortic valve regurgitation after the arterial switch operation. Heart. 2019;105(22):1732–1740.

52.

van der Palen

RLF

Blom

Kuipers

, et al. Long-term outcome after the arterial switch operation: 43 years of experience. Eur J Cardiothorac Surg. 2021;59(5):968–977.

53.

Vida

Zanotto

, et al. Left-sided reoperations after arterial switch operation: a European multicenter study. Ann Thorac Surg. 2017;104(3):899–906.

54.

Sengupta

Carreon

Gauvreau

, et al. Growth of the neo-aortic root and prognosis of transposition of the great arteries. J Am Coll Cardiol. 2024;83(4):516–527.

55.

Salve

Edington

Vijayaraghavan

, et al. Technique of coronary button transfer has no impact on neoaortic root size in simple transposition. Semin Thorac Cardiovasc Surg. 2023;35(2):377–386.

56.

Cha

Baek

, et al. Growth pattern of the neo-aorta after arterial switch operation during childhood. Korean Circ J. 2021;51(1):83–93.

57.

Geiger

Hirtler

Bürk

, et al. Postoperative pulmonary and aortic 3D haemodynamics in patients after repair of transposition of the great arteries. Eur Radiol. 2014;24(1):200–208.

58.

Verhey

Bara

. Influence on fluid dynamics of coronary artery outlet angle variation in artificial aortic root prosthesis. Biomed Eng Online. 2008;7:9. doi:10.1186/1475-925X-7-9

59.

Park

Lee

Seo

. Hemodynamic analysis in aortic dilatation after arterial switch operation for patients with transposition of great arteries using computational fluid dynamics. J Cardiovasc Transl Res. 2025;18(1):79–90.

60.

van der Palen

RLF

Juffermans

Kroft

LJM

, et al. Wall shear stress in the thoracic aorta at rest and with dobutamine stress after arterial switch operation. Eur J Cardiothorac Surg. 2021;59(4):814–822.

61.

Agnoletti

Celermajer

, et al. Acute angulation of the aortic arch predisposes a patient to ascending aortic dilatation and aortic regurgitation late after the arterial switch operation for transposition of the great arteries. J Thorac Cardiovasc Surg. 2008;135(3):568–572.

62.

Dong

Sun

Tang

Zhang

. Types of anatomic relationship between left main coronary artery and pulmonary sinus of valsalva: implications for balloon pulmonary angioplasty and catheter ablation in the pulmonary root. Int J Cardiol. 2019;288:34–38.

63.

El-Hamamsy

Vricella

. Late pulmonary autograft dilation: can we make a good operation great? The tailored approach. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu. 2024;27:42–46.

64.

Nagy

Fisher

Walker

Watterson

. The in vitro hydrodynamic characteristics of the porcine pulmonary valve and root with regard to the Ross procedure. J Thorac Cardiovasc Surg. 2000;120(2):284–289.

65.

Hokken

Bogers

Taams

, et al. Does the pulmonary autograft in the aortic position in adults increase in diameter? An echocardiographic study. J Thorac Cardiovasc Surg. 1997;113(4):667–674.

66.

Van Hoof

Verbrugghe

Jones

EAV

, et al. Understanding pulmonary autograft remodeling after the Ross procedure: stick to the facts. Front Cardiovasc Med. 2022;9:829120.

67.

Zhu

Wilkerson

Pandya

, et al. Biomechanical engineering analysis of pulmonary valve leaflet hemodynamics and kinematics in the Ross procedure. J Biomech Eng. 2023;145(1):011005. doi:10.1115/1.4055033

68.

Mazine

El-Hamamsy

. The Ross procedure is an excellent operation in non-repairable aortic regurgitation: insights and techniques. Ann Cardiothorac Surg. 2021;10(4):463–475.

69.

Sharifulin

Bogachev-Prokophiev

Zheleznev

, et al. Factors impacting long-term pulmonary autograft durability after the Ross procedure. J Thorac Cardiovasc Surg. 2019;157(1):134–141.e3.

70.

Middendorp

Braeu

Baaijens

FPT

Humphrey

Cyron

Loerakker

. Computational analysis of heart valve growth and remodeling after the Ross procedure. Biomech Model Mechanobiol. 2024;23(6):1889–1907.

71.

Sundström

Tretter

. Impact of variation in commissural angle between fused leaflets in the functionally bicuspid aortic valve on hemodynamics and tissue biomechanics. Bioengineering (Basel). 2023;10(10). doi:10.3390/bioengineering10101219

72.

Sengupta

Gauvreau

Shukla

, et al. Natural history of truncal root dilatation and truncal valve regurgitation in truncus arteriosus. Ann Thorac Surg. 2023;116(1):78–84.

73.

Naimo

Buratto

Konstantinov

. Truncal valve repair in children. J Thorac Cardiovasc Surg. 2021;162(5):1337–1342.

74.

Walsh

Brunet

Berruyer

, et al. Complete scan at 19.89 um of the heart of the body donor S-20-29 (Version 1) [dataset]. Eur Synchrotron Radiat Facil. 2024. 10.15151/ESRF-DC-1773964017