Advances in the Management of Congenital Malformations of the Aortic Valve

Abstract

Congenital aortic valve disease is a life-long condition that can require multiple interventions. It is one of the most common causes of congenital heart defect, with bicuspid aortic valve present in at least 1−2% of the general population. Surgical management of congenital aortic valve disease consists of either valve repair or replacement. While aortic valve replacement using the Ross procedure can be considered the gold standard management in the pediatric population, advancements in aortic valve repair techniques have proved its usefulness as an initial management approach as it prevents prosthesis-related complications and patient-prostheses mismatch while the patient grows. Overall, all techniques have their benefits and limitations in terms of growth potential, durability of repair, freedom from reoperation and anticoagulation, infection risk, and mortality. Each patient will require an individualized judiciously selected management plan to minimize the number of interventions over their lifetime. The aim of this review is to discuss the merits and drawbacks of the major techniques currently used in both aortic valve repair and replacement.

1. Introduction

The appropriate management of patients with aortic valve disease is perhaps one of the most debated topics. Unlike the mitral valve, management of aortic valve disease is a more delicate process as poor decisions can lead to chronic strain on the left ventricle, development of aortic insufficiency, high re-intervention rates, morbidity, and mortality. For many patients, the first arguable decision is between balloon valvuloplasty and surgical repair. Balloon valvuloplasty has evolved over many decades and is a reliable and straightforward procedure for any patient who needs immediate relief of a stenosed aortic valve. On the other hand, emerging aortic valve repair techniques have also shown satisfactory results for the management of aortic valve disease. However, repairs in severely dysplastic unicuspid and bicuspid valves are still evolving, and though promising results are seen in some studies, its utility long-term, especially in the pediatric population, is still unknown. Aortic valve replacement techniques have also improved throughout the years and have been the method of choice for some irreparable dysplastic valves. As more technically challenging repair and reconstruction techniques are developed throughout the years, it is imperative to understand if repair or replacement is a better option for certain patient populations.

This chapter will review current literature and attempt to address the following gaps: 1) Is balloon valvuloplasty or surgical valvulotomy more superior? 2) What repair techniques are available for management of aortic valve disease? 3) What are the available options for aortic valve replacement? 4) Is aortic valve repair superior to replacement, or is it just another way to delay aortic valve replacement?

2. Balloon valvuloplasty vs. surgical valvulotomy

The debate of whether balloon valvuloplasty or surgical valvulotomy is the superior initial management for aortic stenosis remains controversial. In balloon valvuloplasty, it is imperative to balance reducing the aortic valve gradient while limiting the amount of aortic regurgitation produced [1]. The challenge interventionalists face is having to make the decision of leaving a patient with residual aortic stenosis (AS) or acute aortic regurgitation (AR) when an ideal outcome is not possible. Both outcomes can pose as risk factors for poor long-term results and further reinterventions [2]. On the other hand, surgical valvulotomy for aortic stenosis is an emerging approach with continuous technique improvements to decrease mortality. However, it is not offered at many institutions as it can be technically challenging, especially for neonates. Hence, it is important to analyze the outcomes for both techniques in terms of the age of presentation – critical neonatal aortic stenosis vs. noncritical aortic stenosis.

2.1 Critical aortic stenosis of neonates

Patients under one month of age with aortic stenosis are classified as having critical neonatal aortic valve stenosis [1]. These patients are usually symptomatic, ductal dependent for survival, and have other associated cardiac congenital anomalies [3]. Most patients will have a smaller valve annulus with either a bicuspid or unicuspid aortic valve, although many are too dysplastic to differentiate [1, 3].

Balloon valvuloplasty has been the preferred method for the management of critical neonatal aortic valve stenosis for many decades. When balloon valvuloplasty is performed on patients with bicuspid or unicuspid valves, the tendency to cause a tear in the fused leaflet is high, causing prolapse of the leaflet, regurgitation, and a need for further intervention [4]. Hence, improvements in surgical valvulotomy techniques have raised the question of which is the superior management option for this patient population.

Donald et al. reviewed literature comparing the outcomes of both approaches in neonates and concluded that mortality is higher for balloon valvuloplasty (56%) compared to surgical valvulotomy (19%). They also concluded that undergoing either procedure during the neonatal period is a risk factor itself for poor outcomes [5]. A similar conclusion was reached by Siddiqui and colleagues who reported that for the group who underwent surgical valvulotomy, freedom from reintervention at 10 years for neonates and infants was 53.9% and 75%, respectively. Freedom from reintervention at 10 years for neonates who underwent balloon valvuloplasty was 17% compared to 50% in infants. They further reported that besides balloon valvuloplasty and age < 1 month, other factors associated with reintervention include unicuspid valve morphology, presence of endocardial fibroelastosis and presence of an atrial septal defect [6]. Zain et al. compared both approaches by performing a retrospective analysis on 25 neonates who underwent both balloon valvuloplasty and surgical valvulotomy. The majority of patients had a bicuspid aortic valve morphology, and one patient had unicuspid aortic valve morphology. Like the previous studies, they also concluded that patients who underwent balloon valvuloplasty had a higher reintervention rate. When comparing other long-term results like development of moderate to severe aortic regurgitation, balloon dilation was still the inferior option [7].

There are other studies that focused on the outcomes of balloon valvuloplasty mentioning neonatal intervention as a risk factor for mortality and high reintervention rates [1, 8, 9, 10, 11]. However, many of these studies included patients before the year 1998, when the Norwood procedure was introduced. When excluding patients after the year 1998, mortality from balloon valvuloplasty decreased significantly. Another limitation that is not always mentioned in these studies, especially in retrospective studies, is that neonates who underwent balloon valvuloplasty tend to be more complex or ill at the time of intervention, hence resulting in a biased comparison [12].

Surgical repair techniques for aortic stenosis can range from a simple blade commissurotomy, to leaflet reconstruction with pericardial patches [13]. Although techniques for the repair of critical neonatal aortic stenosis are evolving, this approach is not adopted at many centers and most still prefer balloon valvuloplasty as the initial palliative method of choice [14]. In patients with tricuspid aortic valve morphology, most repairs consist of a simple blade commissurotomy where the extent of repair is largely within the surgeons’ control [7, 15], unlike balloon valvuloplasty which is a blind technique, and the degree of damage is unknown at the time of intervention [7]. Some have argued that leaflet debridement should also be done for better long-term results during the surgical procedure [6, 14, 16]. For bicuspid and unicuspid aortic valves, surgical repair techniques are more complex as they can range from simple repairs or complete reconstruction of the aortic valve. Repair techniques will be discussed in a later section.

Alexiou et al. analyzed 18 neonates who underwent open valvulotomy for critical isolated aortic stenosis and concluded that operative mortality for surgical repair in this patient population has been decreasing over time as repair techniques improve. They performed simple commissurotomy to the aortic annulus and also excised obstructive nodules on the aortic valvular surfaces if present. Patients with bicuspid aortic valve morphology were not converted to tricuspid morphology. This study yielded excellent results where there was no early mortality and freedom from aortic reintervention was 85% at 5 years. However, the sample size was small and the study excluded patients with complex repairs [15]. Hraska and colleagues agreed with the previous study and stated that surgical valvulotomy in neonates can produce predictable and reliable long-term results for any valve morphology. They analyzed 34 neonates with various valve morphology and achieved a 100% freedom from aortic valve replacement at 20 years for patients with tricuspid valve morphology. They concluded that the underlying morphology and function of the LV are more important compared to the method of repair for determining outcomes. However, the long-term preservation of an acceptable function of the native aortic valve seems to depend on the method and the cusp anatomy. They believe that it is important to achieve tricuspid morphology in a dysplastic trileaflet valve during the repair for a better outcome, but valve reconstruction into a tricuspid morphology from a bicuspid or unicuspid morphology will not yield the same result [16]. Vergnat and colleagues adopted a 2-step approach for 103 neonates with critical neonatal aortic stenosis. It consists of leaflet remodeling and apparatus rehabilitation, and an attempt to achieve a tricuspid arrangement without leaflet reconstruction. They also highlighted the importance of leaflet debridement, which is only possible with surgical repair, to preserve the native valve as balloon dilation resulted in early stenosis [14].

In summary, although preliminary results seem to favor surgical repair of critical neonatal aortic stenosis, there are many other factors to consider. Patients in this population are often very ill and require immediate intervention for survival. This would have contributed to the high mortality rate for balloon valvuloplasty. Additionally, in most studies where the surgical repair was possible, they usually consist of simple surgical techniques like commissurotomy and debridement. Complex repairs are technically demanding and rarely performed in neonates and infants. Therefore, a non-bias method is needed to accurately compare the two approaches. This will be difficult as every patient presents differently and will have specific needs [16]. In terms of replacement options in this population, the Ross−Konno procedure may be the only option because of the small aortic annulus [17]. The Ross procedure will be discussed in a later section.

2.2 Noncritical aortic stenosis

Patients who present after the neonatal period generally have fewer dysplastic valves and adequate aortic annulus size, making surgical repair more feasible [3, 18]. Similar to critical aortic stenosis, surgical repair seems to be preferred in this patient population.

Hill and colleagues performed a meta-analysis to compare both techniques and reported that most literature either determined surgical valvulotomy as the more superior method or found no difference between the two. The meta-analysis consisted of 2368 patients with mean age of 2.9 months. Overall, at 10 years, the survival rate was 87% for balloon valvuloplasty and 90% for surgical valvulotomy. There was a significant difference for freedom from reintervention at 10 years, with balloon valvuloplasty at 46% and surgical valvulotomy at 73%; however, no significant difference for freedom from replacement was found. In a subgroup analysis for infants <1 year of age, the results were similar and only differences for freedom of reintervention were found with balloon valvuloplasty at 40% and surgical valvulotomy at 60% [19].

Brown et al. also concluded that surgical repair is the superior option after performing analysis on 158 patients older than 2 months of age. They reported that the surgical method resulted in greater gradient reduction and significantly less regurgitation. There was also a longer interval for reintervention for the surgical approach [18]. Brown et al. performed a retrospective analysis on 509 patients who underwent balloon valvuloplasty and concluded that patients older than 11 years of age have an increased risk of developing moderate to severe aortic regurgitation after balloon valvuloplasty [20]. Hence, it appears there may be certain age groups where balloon valvuloplasty should be avoided to minimize the possibility of reintervention.

Overall, it appears that surgical repair is superior to balloon valvuloplasty for patients with noncritical aortic stenosis by comparing freedom from reintervention rates. However, without standardization for the definition of successful management, thresholds for reintervention and replacement, or ballooning and repair techniques, it is difficult to compare the two modalities accurately without biases in both critical and noncritical aortic stenosis patients. Furthermore, most studies cover a diverse age group, span over different amounts of follow-up time, or cover different time periods, making the comparison even harder [5, 11, 12]. Since each patient’s anatomy, physiology, and age of presentation are different, the management plan should be tailored for each individual to minimize complications [7, 9]. A successful outcome also depends on the options available and skills of the surgeons and interventionalists [9].

Balloon valvuloplasty has improved and become safer over the past decade [9]. Having a standardized progression may be helpful for this blind technique in minimizing complications. Porras et al. performed a study to investigate the utility of a Standardized Clinical Assessment and Management Plan (SCAMP) algorithm for the management of congenital aortic stenosis. They used the cut-off of ≤35 mmHg residual aortic stenosis and the degree of aortic regurgitation present after each ballooning to determine the option for further intervention. Following the algorithm, they managed to ensure that all patients achieve a final gradient of ≤35 mmHg without causing greater aortic regurgitation after the ballooning procedure [21]. However, it is worth noting that the study sample size only consisted of 23 patients and 92 controls, and follow-up duration was only 10 years.

There are other studies that agree with Porras et al. that acute residual aortic stenosis gradient and post-dilation aortic regurgitation were factors most strongly related to the decision for long-term aortic valve replacement [2, 10, 12, 20]. Rao and colleagues concluded that the most important factors leading to restenosis are immediate post-valvuloplasty aortic valve stenosis and patients ≤3 years of age at the time of procedure. They also commented that many studies found a linear relationship between follow-up time duration and post-procedural aortic insufficiency, however, no definite causative factors were found [12]. Brown et al. also concluded that a residual aortic stenosis gradient of ≤35 mmHg was associated with greater freedom from aortic valve replacement and stated that having a lower aortic stenosis gradient might be more important than ensuring minimal aortic regurgitation [20]. However, Sullivan et al. disagreed and reported that acute post-procedural aortic regurgitation is associated with a greater risk instead. They discovered that patients with moderate or severe acute aortic regurgitation post-procedure with residual aortic stenosis gradients <30 mmHg had three times greater long-term risk for aortic valve replacement compared to those with mild or less aortic regurgitation and > 30 mmHg residual gradient. However, they did note that the study population consisted of more neonates, and age could be a modifier [2].

Since reinterventions and replacement might be inevitable due to recurrent stenosis or progressive aortic regurgitation [11, 22], balloon valvuloplasty can be considered as an initial palliative option for ill patients [7, 11]. Zain et al. proposed that balloon valvuloplasty should be done for bicuspid valves with equal size leaflets, while surgery should be reserved for thick, nodular dysplastic valves or unicuspid valves with small aortic annulus to prevent the need for multiple interventions [7, 11]. Ballooning in highly dysplastic valves distributes circumferential force unevenly, causing tears in the weakest part of the aortic valve and may therefore disrupt the cusps unevenly leading to poorer outcomes [11, 18]. On the other hand, surgical repair allows for direct inspection of the valve where surgeons can have better control of the extent of commissurotomy and more precise repair [7, 15]. Nonetheless, repeat balloon valvuloplasty should still be considered as the first option in most patients after restenosis as it can still yield excellent results [12].

3. Aortic repair

Many aortic repair techniques have evolved since their first introduction resulting in improved mortality and reintervention rates. However, its place in the management of patients with aortic valve disease is still controversial. Another big debate is whether aortic repair or replacement is superior, despite most agreeing that aortic valve replacement is inevitable in patients with congenital aortic valve disease.

The aortic repair can range from simple techniques such as commissurotomy to complex ones such as aortic valve neocuspidization. To perform a precise aortic valve repair, it is imperative to understand that the aortic valve operates as a unit consisting of the ventriculo-aortic (VAJ) and sino-tubular junction (STJ). These provide a framework for a functional annulus and cusps (Figure 1). Hence, both the annulus and cusps need to be considered at the time of surgery to ensure a proper repair [4, 23, 24, 25, 26]. The ideal goal of any aortic valve repair is to restore the leaflets and functional aortic annulus to their normal geometry while ensuring normal mobility of the valve cusps [24].

Figure 1.

Anatomy of the aortic valve.

A repair-oriented functional classification was developed to improve understanding of the pathophysiology and communication between physicians [27, 28]. With this classification, an algorithm can be established to guide physicians to the correct procedure. Type 1 refers to defects in the functional annulus, and it is divided into several subtypes. Type 1a refers to dilation in the STJ and ascending aorta, type 1b indicates dilation is in the STJ, sinuses of Valsalva, and the VAJ, and lastly type 1c refers to pure dilation of the VAJ. Depending on the subgroup, the regurgitant jet differs and physicians will be able to determine the type of annuloplasty that is appropriate [24, 26, 27, 28, 29, 30]. Type 2 refers to aortic insufficiency caused by cusp prolapse due to excessive motion of the aortic cusps [24, 26, 27, 28]. These are generally managed by resuspension, free margin plication, or triangular resection techniques [27, 28, 29]. Lastly, there is type 3 aortic insufficiency where cusp motion is restricted, possibly due to thickening and fibrosis. This is potentially the most challenging pathology for repair procedures as an additional patch augmentation is frequently needed which can result in structural degradation [24, 26, 27, 28, 31]. Other aortic valve pathologies that are not classified are cusp perforation and aortic stenosis. Cusp perforations are generally repaired directly with a patch [28]. Materials for leaflet and patch repair will be discussed in a later section. Aortic stenosis is commonly managed with commissurotomy [29] or reconstruction, depending on the complexity, if the repair route is chosen.

Since bicuspid aortic valves are the most prevalent aortic valve disease, these are also the most commonly repaired morphology [29]. They often present with defects that are combined from the subgroups listed above, with the most common being prolapse of the fused cusp with dilation in the annulus [24, 28, 32]. Cusp configuration of bicuspid aortic valves uses a different classification first proposed by Sievers and Schmidtkes (Table 1). The most common, type 1, refers to a bicuspid valve with a median raphe and asymmetrical aortic sinuses, and type 0 refers to a symmetrical bicuspid valve with no raphe present [24, 27]. Repairs for type 0 valves are usually straightforward and performed in a similar fashion to tricuspid valves, where the nonprolapsed cusp is used as the reference height for plication or resuspension. In type 1 repairs, patch augmentation may be required depending on the amount of native cusp tissue left after shaving or debridement [24]. For patients where annulus dilation is present, a valve-sparing reimplantation technique for annuloplasty might be preferred as it carries less risk for recurrent dilation. Although reimplantation is a more technically challenging procedure, compared to a subcommisural annuloplasty, both techniques are similar in terms of morbidity [24, 33]. Hence, surgeons who have the expertise should consider performing valve-sparing implantation for better long-term outcomes. The overall goal in the repair of bicuspid valves should meet the following two criteria to ensure an optimal aortic root geometry: 1) the basal ring diameter should be reduced to less than 25 mm, 2) effective cusp height should be restored to above 8 mm. This will lead to better long-term valve stability and minimize degradation rate of the tissues [28].

Table 1.

Sievers and Schmidtkes classification for bicuspid aortic valve.

Unicuspid aortic valves were previously classified under the Sievers classification for BAV as Type 2 due to the presence of two raphes; however many now consider it as a separate entity due to its unique anatomy [32, 34]. There are two main types of unicuspid aortic valves – either unicommissural (most common) where an eccentric slit-like orifice is present with one commissural attachment, or a commissural where there is a pinhole orifice with no commissural attachment [34, 35, 36, 37]. Management of unicuspid aortic valve remains a challenge since patients often have dysplastic and severely calcified cusps with no functional native tissue [28]. Therefore, the aortic replacement has been the main method of management for a unicuspid aortic valve for many decades; however, more are turning to repair first and delaying replacement to prevent complications from valve replacement [38].

3.1 Bicuspidization

Schäfers et al. proposed the first unicuspid aortic valve bicuspidization repair technique which consisted of a reconstructive approach for unicommissural unicuspid aortic valve with aortic regurgitation or dilatation [34, 36]. The technique utilized stay sutures on the commissure with normal height, as well as one placed at the same height above the rudimentary anterior commissure to indicate the location of the new commissure. The fused cusp tissue was then incised anteriorly. Glutaraldehyde-treated autologous pericardium was then used to fill the gap between the rudimentary cusp and new commissure, and also for the defect of the fused cusp so that it is in the position of the rudimentary cusp [36]. This ultimately leads to a standard asymmetric BAV configuration with 2 normal commissures [34]. The key parameter in this repair was an effective cusp height of 8 mm in pediatric and 10 mm in adult repairs to ensure a near-normal geometry to guarantee consistency and reliable results. The technique was performed on 20 patients with a mean age of 26. Schafers et al. reported good initial functional results with satisfactory hemodynamics post repair. At 4 years following repair, freedom from moderate or more aortic regurgitation was 77%, freedom from reoperation was 67%, and freedom from aortic valve replacement was 100%. However, the durability of autologous pericardium for cusp extension was not explored in this study. Furthermore, due to the short follow-up period, cusp configuration as a patient grows is also unknown [36].

Aicher et al. investigated hemodynamic effects and overall stability differences between the asymmetric reconstruction technique proposed by Schafers et al. previously, and a similar symmetric bicuspidization cusp reconstruction. The study consisted of 118 patients at a mean age of 27 years with either aortic stenosis, aortic regurgitation, or both. Glutaraldehyde-treated autologous pericardium was used for both groups, and both the effective and geometric heights were kept consistent at 10 mm and > 20 mm, respectively. Overall, there was no significant difference in reoperation rates between the configurations; however, the symmetric configuration yielded better hemodynamics with lower mean and peak gradients, especially during exercise [39]. Furthermore, Aicher et al. and Franciulli et al. combined the repair technique to include root remodeling to address aneurysm of ascending aorta and aortic roots which occur commonly in the unicuspid aortic valve. Dilation of the VAJ had previously been identified as a risk factor for repair failure [40]. Both studies reported improvement in the durability comparted to isolated aortic valve repair. Additionally, Franciulli et al. reported postoperative hemodynamics similar to age-matched patient population with tricuspid aortic valve.

Kolesar et al. extended the symmetrical cusp repair technique to 17 patients with stenotic unicuspid aortic valve and a mean age of 23 years. They also agreed on the importance of ensuring a standardized cusp effective height to minimize the risk of reoperation. Kolesar and colleagues also performed open extra-aortic ring implantation using a Dacron tubular graft in patients with annulus ≥25 mm, given the high risk of aortic annulus dilatation in unicuspid aortic valve. They reported that performing a ring annuloplasty in valve-sparing aortic valve repair significantly increases freedom from valve-related reoperation and freedom from moderate or worse aortic regurgitation. Freedom from valve-related reoperation was 100%, although the follow-up period was too short (about 2 years). Similar to the previous studies, they also highlighted the importance of creating a symmetrical orientation with the new commissure. One difference between the repair techniques proposed by Kolesar and Schafer is that Kolesar used equine pericardium instead of autologous pericardium. Despite this, Kolesar reported no statistically significant difference in freedom of reoperation between the two patch materials [37].

Matsushima et al. [41] proposed that management of unicuspid aortic valve, whether aortic stenosis or regurgitation, should be a three-stage approach - firstly, balloon aortic valvulotomy during infancy or neonatal period, then bicuspidization with cusp augmentation, and lastly aortic valve replacement mainly with a pulmonary autograft later in life if needed. They also used a symmetrical bicuspidization technique and involved 60 patients who are 18 years old or younger in the study. These authors reported an overall survival of 96% at 5 and 10 years. However, 33% of patients required aortic valve reoperation, mainly due to patch degeneration. The authors also agreed with Kolesar et al. that an external suture annuloplasty is necessary during the repair to improve cusp coaptation, reduce cusp stress and prevent suture dehiscence especially if the aortic root is dilated.

The authors also noted that bicuspidization using their technique can be done even after a balloon aortic valvuloplasty is performed during neonatal period. Their repair technique can tolerate cusp tear which occurs commonly during balloon valvuloplasty. Therefore, initial balloon valvulotomies are not a contraindication for their bicuspidization cusp repair technique. In balloon dilation for unicuspid aortic valve, tears commonly occur opposite to the normal commissure, which in this case is removed for this bicuspidization technique, hence avoiding the controversial debate on initial management plan [41]. They utilized three different patch materials for cusp augmentation – glutaraldehyde-treated autologous pericardium, decellularized xenogenic tissue, and expanded polytetrafluoroethylene (ePTFE) membrane. For each material, freedom from aortic valve reoperation using autologous pericardium was 72% and 52% at 5 and 10 years respectively. 50% of patients with ePTFE membrane required aortic valve reoperation later in life; however, this sample size was small (2 out of 4 patients). None of the patients with decellularized xenogeneic tissue required aortic valve reoperation. They concluded that each patch material had its own limitations. For example, patch augmentation with glutaraldehyde-fixed autologous pericardium was more prone to patch degeneration, and suture dehiscence was more commonly seen when using the ePTFE membrane. Overall, decellularized xenopericardial patch yielded the best durability, however, a longer follow-up period is required to compare appropriately with autologous pericardium [41].

Other studies have also investigated the utility of different materials in aortic valve repair of various valve morphology. Nezhard et al. compared the repair of various aortic valve malformations with either bovine or non-treated autologous pericardium. They concluded that non-treated autologous pericardium was preferred in easy and less technical repairs, such as perforation patching, and the bovine pericardium was preferred in complex repairs or when autologous pericardium was not available. Overall, the bovine pericardium is comparable with autologous pericardium, however, it still does not compare with native valve tissue and should be avoided if possible [42]. Nordmeyer et al. investigated the durability of decellularized bovine pericardial patch material for aortic valve reconstruction. Despite the reported advantages of the material and excellent short-term results, the authors disagreed with Matsushima and reported concerns regarding long-term durability of the material. Patients with decellularized bovine pericardial patches had thickened leaflets with reduced mobility at 3 years and required further reintervention. Therefore, the search for an optimal patch material is still ongoing [43].

Si et al. addressed the issue of patch degeneration by proposing a bicuspidization unicuspid aortic valve repair with primary leaflet reconstruction and geometric annuloplasty ring repair. They studied patients with aortic regurgitation and attempted to create cusps using primarily the patient’s native leaflet tissue. Using this technique supported with an appropriately sized annuloplasty ring, they created cusps with equal free-edge length, geometric, and effective heights. The authors reported excellent results and concluded that most unicuspid aortic valve variations can be repaired using this standard technique. However, the study was limited due to its small sample size and short follow-up durations. More importantly, the repair technique was limited by the smallest available ring size, which is currently around 19 mm, making the technique inappropriate for repair in infants or children [44].

3.2 Tricuspidization/reconstruction

A recently emerging repair technique for unicuspid, as well as bicuspid aortic valves, is tricuspidization by reconstructing all three leaflets [45]. Reconstruction of aortic valves was first proposed by Duran et al.; these authors reported 51 patients with a mean age of 31.2 years who underwent reconstruction of all three leaflets using a single strip of rectangular glutaraldehyde-treated autologous pericardium [46]. Despite some success, the authors concluded that a standardized procedure is needed to ensure technique reproducibility and that further research is needed to determine the best material for reconstruction [46]. Most recent literature includes various valve morphologies in their studies, with the majority being bicuspid aortic valves.

Ozaki et al. proposed another technique that consisted of the reconstruction of each leaflet independently using glutaraldehyde-treated autologous pericardium [47]. This technique is also known as aortic neocuspidization (AVNeo). Since the technique involves complete resection of all three dysplastic leaflets, it can be applied to any congenital aortic valve malformation, even when root reimplantation is required in patients with annulo-aortic ectasia [47]. The authors performed this technique in over 404 patients with unicuspid, bicuspid, tricuspid, and quadricuspid aortic valves. They believe that an aortic valve should be considered as a collection of different-size cusps, and by measuring the distance between the commissure, one should be able to determine the area of each cusp and perform individual cusp reconstruction. With this technique, reconstruction can more effectively preserve the natural motion of the annulus and the coordination between the structures surrounding the aortic valve. At 4 years of follow-up, the survival rate was 87.7% and freedom from reoperation was 96.2% [48]. In another study, Ozaki and associates focused the technique on bicuspid and unicuspid valve morphology in patients under 60 years and noted commendable hemodynamics in all patients, especially those under 40 years. There were also no signs of calcification, as well as natural motion of all three cusps [49]. Ozaki and colleagues later updated their procedure, and the biggest change was seen in the repair of patients without trileaflet aortic valves. They decided that tricuspidization with three equal-size cusps was preferred for an even movement of the cusps [50].

Previously, AVNeo was performed largely in the adult population, where the predominant pathology was acquired calcific aortic stenosis associated with bicuspid valves. In the pediatric population, patients who require further intervention usually are diagnosed with congenital aortic valve stenosis and have previously undergone balloon aortic valvuloplasty resulting in regurgitation. Therefore, since AVNeo requires excision of all leaflets, it is applicable in most pediatric patients as well [45]. The AVNeo procedure was reproduced by Baird and colleagues on the pediatric population with a variety of congenital aortic valve morphologies. Due to the prevalence of bicuspid and unicuspid valve morphology in the pediatric population that required intervention, the authors noted technical differences in the pediatric population when creating the 3 equal leaflets. They also highlighted a need to augment the noncoronary sinus leaflet to accommodate 3 equal leaflets creating similar annular and STJ, given the small aortic valve dimensions [51]. Overall, they found satisfactory post-reconstruction hemodynamics and the possibility for annular growth. They also concluded that the Ozaki procedure is promising for pediatric patients, however, it can be technically challenging in patients with smaller aortic annuli and roots and should only be performed by experienced surgeons [45]. Wiggins and associates also utilized AVNeo, as well as a single leaflet reconstruction technique for tricuspidization in pediatric patients who were unable to undergo either a mechanical replacement or the Ross procedure. They were able to successfully perform the procedure in patients with annular sizes as small as 6.7 mm with resultant favorable post-surgical hemodynamics. The authors also commented on the benefit of increased free margin length in AVNeo, and the advantages of preserving annular hemodynamics to allow growth and further reconstruction surgeries if required [22].

3.2.1 Reconstruction materials

Another controversial topic in the reconstruction of aortic valves is the preferred material for the AVNeo procedure [45]. While autologous pericardium is the most widely used for the procedure due to its availability, convenience, and low cost [4, 52]. Recent studies have shown it to be less favorable due to its poor biomechanical properties, high rates of calcification, and lack of growth potential [4, 45, 51].

The autologous pericardium has been considered the “gold standard” material for aortic valve repair and reconstruction for many decades. Treatment with glutaraldehyde produces autologous pericardium that is more resistant to retraction and degeneration [52]. It can also increase the tensile strength to four times greater than noncalcified native aortic leaflets [50]. However, varying reoperation rates ranging from 15 to 33% have been reported in different studies. Many previous studies focused on the durability of glutaraldehydetreated autologous pericardium in AVNeo performed in the adult population, but its durability in younger patients has not been studied and needs to be further investigated since there is an increased risk of degeneration [22].

The first studies investigating materials for aortic valve reconstruction were by Duran et al., where they compared bovine and autologous pericardium. They determined that autologous pericardium was more favorable as it did not show fibrocalcific deterioration, unlike bovine pericardium. They also concluded that the durability of pericardium will depend on its implanted position and the pretreatment process [53].

Several types of treated bovine pericardium have since been introduced for valve leaflet reconstruction. An example is CardioCel®, where bovine pericardium is treated with glutaraldehyde, as well as by further anti calcification tissue-engineering techniques, prior to reconstruction [45, 51]. Mazzitelli and colleagues utilized CardioCel® for three pediatric AVNeo procedures and reported that there were no reoperations at early follow-up [54]. Despite promising results, long-term follow-up is not available and is needed due to concerns about developing calcification [22, 45, 51]. Another example is the Photofix®, a bovine pericardium that undergoes photo-oxidation fixation. Compared to the glutaraldehyde-treated bovine pericardium, it is more resistant to calcification and inflammation, making it a more durable option. However, Photofix® showed signs of annular separation that are of concern [45, 51]. Lastly, there is also the Matrix®, an untreated equine pericardium, which has shown satisfactory preliminary results. Furthermore, its thin thickness may be advantageous for smaller pediatric patients [45]. Another interesting theory that is yet to be explored is the remodeling characteristics of extracellular matrix scaffolds. It has been proven in some tissues that degradation occurs due to host remodeling responses to naturally occurring biologic extracellular matrix scaffolds. Hence, durable valve leaflets can potentially be produced if extracellular matrix scaffolds are seeded with mesenchymal stem cells to be differentiated into valve leaflets. However, this theory has yet to be proven [4].

3.3 Unicuspid aortic valve: bicuspidization or tricuspidization?

In unicuspid aortic valves, it is debatable whether a symmetrical bicuspidization technique is superior to a tricuspidization method. This may have been related to the variability in commissural height and interrelation of three coaptation lines in tricuspid aortic valves [39].

Kawase et al. performed trileaflet reconstruction on 9 unicommissural unicuspid aortic valve patients. They utilized glutaraldehyde-treated autologous pericardium to independently construct each leaflet so that coordination of the valve leaflets is maximized. These authors reported that tricuspidization is superior to bicuspidization as it produces a longer total length of the free margin of the aortic valve leaflets, resulting in a better full opening of the aortic valve. However, the study was only performed on adults with a mean age of 48.9, and the average follow-up time was 18 months. Hence, a larger and wider study population, and long-term results are needed [38]. Kohei and associates discussed tricuspidization of 2 patients with unicuspid aortic valves and performed aortic valvuloplasty and root construction using autologous pericardium and a prosthetic graft. They agree that bicuspidization is hemodynamically inferior to tricuspidization as it adds stress to the leaflets and restricts opening. However, long-term results are also unavailable to determine the durability of this approach [55]. Therefore, it appears that tricuspidization is an area of repair that yet to be explored for unicuspid aortic valve repair. Both studies discussing tricuspidization techniques were performed on adults with a small sample size, hence durability results in pediatric population and long-term results are lacking.

4. Aortic valve replacement

Before the emergence of surgical repair as an option to delay aortic valve replacement, the common management plan for most patients with isolated aortic stenosis was to perform balloon valvuloplasty until a replacement is needed. Isolated aortic valve replacement is generally associated with low mortality and morbidity especially if performed worsening the myocardial function [56]. However, the decision of which type of aortic valve replacement to choose is complex as it depends on many factors such as patient characteristics, medication compliance, access to care, etc. [57]. In younger patients, it is important to choose a valve replacement that is resistant to degeneration, infection, and has growth potential [56]. Whereas, in the older population, the priority lies in minimizing the need for anticoagulation due to potential thromboembolism complications [57, 58].

Tissue aortic valves can be either bioprosthetic with bovine or porcine valves, homografts, or allografts. They are generally preferred in older patients due to their low thromboembolism risk [57, 58, 59]. However, tissue valves are prone to a high risk of degeneration and structural failure which translates to high reoperation rates [56, 59, 60, 61, 62], hence their use needs to be carefully evaluated [58]. Additionally, they often come in sizes larger than 19 mm, lack growth potential, and are prone to patient-prosthesis mismatch (PPM) as the recipient grows. One of the few pediatric populations, where tissue replacement can be considered is teenage females who wish to be pregnant in the future due to lack of anticoagulation. However, if accompanying conditions such as atrial fibrillation, previous emboli, or an enlarged left atrium are present, the need for anticoagulation may still be warranted [56, 58, 62, 63]. Homografts may sometimes be available in smaller sizes suitable for younger children, and they are associated with excellent hemodynamics. They are also resistant to infection, which makes them preferred for patients with invasive endocarditis. However, the limitation of rapid valve deterioration leading to recurrent stenosis and regurgitation outweighs these benefits, hence its use is often discouraged [61].

The most attractive aspect of mechanical aortic valves is its superb durability. However, its biggest drawbacks are the need for anticoagulation which can significantly impact quality of life, and the lack of growth potential [58, 64, 65]. In older patients, this can lead to drug interactions and increased thromboembolic risks. In children, there can be possible compliance issues and difficulty with activity restraints, though children seem to have an overall lower risk of bleeding and thromboembolism [58]. Newer generation mechanical valves have been explored to prevent the need for warfarin and an INR goal >2.5, however, eliminating anticoagulation completely is still unattainable in mechanical valves [57]. In children with mechanical valves, the use of antiplatelet drugs instead of warfarin for anticoagulation can be considered [63]. Reoperations are uncommon in patients who have received mechanical valve replacements, and they are mainly due to PPM or subvalvular obstruction, both more prevalent in the younger recipients [58, 61, 64]. Overall, the freedom from reoperation for this subgroup is over 92% at 20 years in both adults and children [64].

The Ross procedure utilizes the patient’s own pulmonary valve to replace the diseased aortic valve and a tissue valve for the pulmonary valve [59]. Though this procedure is more technically demanding than the other replacement methods, it produces better hemodynamics, fewer complications, and allows growth potential in the aortic valve [58, 59, 66]. Since the pulmonary autograft is relatively resistant to calcification or degeneration, trans-valvular gradients mostly remain unchanged from the immediate post-operative period. Additionally, having better post-replacement hemodynamics places less strain on the left ventricle, overall resulting in improved mortality [59].

The biggest disadvantage to this approach is that it results in a double valve disease where both valves will be at risk for reoperation and degeneration [60]. Some argue that most reoperations after a Ross procedure occur on the pulmonary conduit [60, 67], and this results in a higher total reoperation rate if both valves are considered, compared to mechanical replacement [60, 66, 68]. Many alternatives for the pulmonary valve have been investigated but none were superior to a homograft. The only exception was in young infants since homografts have poor durability [66].

The preferred age groups for the Ross procedure are older children and young adults, especially those with normal-sized aortic annulus due to procedure complexity [65, 66]. Infants less than age 1 generally have a high mortality rate of up to 20%, but this decreases as patient age increases and falls to almost 1% for children above 1 year old [58, 60, 69]. It is also important to note that patients who require the Ross procedure as an infant generally have critical stenosis with failed previous repair or balloon valvuloplasty which may have contributed to the high mortality [69]. Besides mortality, pulmonary valve reoperations were also reported to be higher in neonates and infants compared to older children [60, 69].

The most common complication of the Ross procedure is aortic root dilation causing regurgitation [58, 70]. Reoperation rates due to dilation can range from 8 to 30% within 10 years after surgery [66, 70]. Several methods to prevent dilation and improve durability have been explored. The most common strategy is to place external support, using either a Dacron (DuPont, Wilmington, DE) graft or a Gelweave (Vascutek Ltd., Renfrewshire, United Kingdom) sinus of Valsalva graft around the aortic root. Excellent results have been reported for this technique – freedom from dilatation for standard Ross at 3 years was 52%, and is 90% for supported Ross [58, 66, 68, 71, 72]. External stabilization using a strip of Daron at the STJ has also been explored and is considered a standard Ross procedure in some institutions for older children [66, 73]. Another common approach is to wrap the native aortic root around the pulmonary autograft for extra stabilization [70, 71]. However, this approach is limited by the size of the aortic root. In patients with bicuspid aortic valve and aortic insufficiency, their aortic roots are usually already dilated and hence will be unable to provide adequate support. On the other hand, patients with bicuspid aortic valves with aortic stenosis tend to have small aortic roots which are not feasible for this method of stabilization [73].

Despite promising results with different supported Ross procedures, these stabilization techniques should only be used for older children or young adults because it eliminates the growth potential which is a crucial benefit of the Ross procedure as a replacement option [14, 68, 71, 72, 73]. A new supported Ross technique with Konno annular enlargement using a subcoronary technique has been explored to extend the procedure to younger children. However, this procedure becomes extremely complicated, and long-term results are needed to prove its durability [72].

Between tissue and mechanical aortic valve replacements, mechanical replacements seem superior overall. Tissue valves do not require anticoagulation and can be available in smaller sizes; however, their high rate of valve degeneration and reintervention makes it a poor valve replacement option for young patients. On the other hand, mechanical valves have better durability and higher freedom from reoperation rates [64, 69]. Despite better long-term benefits, its impact on quality of life due to anticoagulation needs to be evaluated when considering this replacement option.

Both tissue replacement and the Ross procedure share similar immediate post-operative hemodynamic and are both resistant to infections. Furthermore, they are both available in small sizes and do not require anticoagulation. In terms of surgical technicality, tissue replacements are generally easier since it does not involve the pulmonary valve. However, tissue valves are associated with rapid valve deterioration, resulting in higher mortality and reoperation rates. Therefore, though it can be an option if resources are available, it is less preferred in the younger patients [59, 60, 69, 74].

Overall, there is no statistically significant difference in reoperation rates between mechanical replacement and the Ross procedure [65, 69]. However, surgeries involving the mechanical valve are less complicated since it does not involve the pulmonary valve or the aortic root procedures [64]. The Ross procedure may have superiority since it allows for growth potential and does not involve anticoagulation, resulting in more favorable cardiac and valve-related mortality [59, 60, 65, 69]. Nonetheless, it is difficult to compare these two approaches since the Ross procedure is usually performed on younger patients for congenital causes, whereas the mechanical replacement is favored in older patients with connective or rheumatic tissue pathologies, or preoperative aortic regurgitation [75].

Ultimately, all approaches to replacement have their own risk and benefits. To properly evaluate the best option, it is important to first consider the age of the patient, the need for growth potential, and the possible need for reoperations in the future. The second important point is then to assess for possible complications and comorbidities.

4.1 Repair vs. replacement

The topic of whether aortic valve repair or replacement is the preferred approach to the management of aortic valve pathology is again controversial. Both aortic valve repair and replacement have produced similar results and shown their respective benefits and limitations [76]. Many institutions have started to express their preference to repair valves and delay replacement for as long as possible. Burrato and associates noted an increase in durability when the Ross procedure is done as a reoperation after valve repair and attributed this to the presence of postoperative scarring aiding in natural stabilization [77]. Similarly, favorable results were also reported by Popov et al. regarding mechanical valve replacement after previous repair [61].

There are subgroups of patients in whom repair has been shown to be a superior approach to replacement. Etnel et al. performed a meta-analysis and concluded that replacement is a suboptimal option in the pediatric population [60]. Generally, the repair is preferred in younger patients as it allows for growth potential, where most replacement options are unable to accommodate growth and will require multiple reoperations. Repair also avoids anticoagulation complications and activity restrictions [23, 78]. Danial et al. compared repair and the Ross procedure and concluded that both options were similar in terms of mortality and freedom from reintervention. However, aortic valve repair pulled ahead in terms of surgical morbidity and complications. They also noted that in patients who require immediate intervention, repair with a patch is the preferred option since it yields better short-term outcomes [64]. It is important to note that both young age and unicuspid valve morphology were identified as the greatest risk factors for poor outcomes in repairs [31]. Additionally, pediatric patients who present with an aortic valve deformity usually have complex pathology that requires multiple interventions. Hence, these risks and repair complexity should be weighed against the limitations of having a replacement [79]. It is possible that aortic valve repair should be considered for neonates and infants, whereas both aortic valve repair and the Ross procedure can be options for older children [78]. Ultimately, replacement may be unavoidable for patients who present with aortic valve disease at a young age.

Another population where a repair can be more favorable is in adults where replacement is contraindicated or those who are already undergoing other cardiac procedures at the same time [23, 27]. Regardless of the patient population, the surgeon’s comfort level and experience in performing the aortic valve repair can also deter the decision as well [23, 80]. Another thought is to choose the management approach depending on whether the pathology is aortic stenosis, insufficiency, or mixed disease. Valve repair has shown satisfactory low reintervention rates for patients with aortic stenosis. However, the repair had a higher reintervention rate for patients with mixed disease or aortic insufficiency. This could however be institution dependent as repair outcomes depend largely on the surgeon’s experience and expertise [81].

In conclusion, it is difficult to compare both modalities without bias as repair techniques are still evolving and there is still a lack of long-term data available. Additionally, many complex repair techniques, especially for children, are only available at several specialized institution whereas aortic valve replacement techniques have been around for many decades and are known to be safe and successful among most surgeons [23]. Studies that focused solely on either repair or replacement usually consist of different patient populations with varying aortic valve pathology [29].

To increase the widespread practice of aortic valve repair techniques, there needs to be a standardized range of techniques to help guide surgeons through variations in valve deformities. More surgeons will need to be trained and familiarized with these techniques so that they will have the expertise to explore repair instead of relying on the reliable replacement option [23, 81]. An example algorithm was proposed by Danial and colleagues, where repair should be preferred if native cusps with good mobility can be achieved after shaving and/or debridement, there is at least one functional commissure, and a cusp-free margin that is composed of half native tissue. However, if the aortic annulus is less than two standard deviations, both repair and replacement can be done together with the Konno procedure [67, 81]. Another management pathway that was suggested for the pediatric population depended on peak systolic gradient or Doppler mean gradient [82].

5. Conclusion

There are many decisions to make when managing a patient with aortic valve disease, and it is crucial to conscientiously evaluate each decision as they can all have a huge impact on the patient’s life. Regardless of the management approach, it is certain that young age of intervention, especially for neonates and babies under the age of 1, is the main risk factor for poor outcomes. For the pediatric population, an attempt to preserve the native valves for as long as possible is important to minimize reintervention rates. A possible basic algorithm for patients with aortic stenosis could be to either perform balloon valvuloplasty or simple surgical valvulotomy as an initial step and to delay replacement for as long as possible. The choice between balloon or surgery will be dependent on the age, condition severity, and symptoms at presentation. If replacement is then needed at a later age, a supported or non-supported Ross procedure may be the most appropriate. Unfortunately, the management of patients with aortic regurgitation or mixed disease is a lot more complex. Many promising repair techniques have been developed for highly dysplastic bicuspid or unicuspid valves, however, long-term and age-matched results for durability and morbidity are still lacking.

Conflict of interest

The authors declare no conflict of interest.