Introduction
A heterogeneous population is encountered by those performing systemic arterial valvar surgery in patients with CHD. This heterogeneity is in both age and underlying congenital cardiac disease. The dysfunctional systemic arterial valve may represent a primary malformation or may manifest following surgical operation to repair or palliate a multitude of CHDs. For example, up to 2% of the population are born with a congenitally malformed aortic root, most commonly a functionally bileaflet aortic valve. Reference Basso, Boschello and Perrone1,Reference Nistri, Basso, Marzari, Mormino and Thiene2 Even amongst this common lesion, the anatomy of the root and its valve is markedly variable. Reference Tretter, Spicer and Franklin3 Approximately, 10% of these patients will require surgery or intervention in childhood for aortic valvar dysfunction and/or significant aortic root or ascending aortic dilation, Reference Tripathi, Wang and Jerrell4 with over two-thirds requiring such by the fifth decade of life. Reference Masri, Svensson, Griffin and Desai5 This incidence increases in those with a functionally unileaflet valve. Reference Roberts and Ko6 Various congenital cardiac surgeries result in creating a neo-aortic root and valve, whether utilising a pulmonary or truncal valve. Following the arterial switch operation for transposition of the great arteries, up to 10% will have moderate or greater aortic valvar regurgitation and/or require neo-aortic root surgery by a mean age of 15.2 years, Reference Zhu, Fricke, Buratto, Chowdhuri, Brizard and Konstantinov7,Reference Sengupta, Carreon and Gauvreau8 with approximately 12% requiring surgery into their third decade of life. Reference Zhu, Fricke, Buratto, Chowdhuri, Brizard and Konstantinov7 More concerning, over half of those requiring neo-aortic reoperation will subsequently undergo valvar replacement by 25 years following the initial arterial switch procedure. Reference Zhu, Fricke, Buratto, Chowdhuri, Brizard and Konstantinov7 Similarly, when the Ross operation is performed in an older child, the need for neo-aortic reoperation is approximately 25% by 15 years post-operatively. This risk is less defined in the neonatal Ross population. Reference Konstantinov, Bacha and Barron9 Last, the aortic root and its valve may be at risk in the setting of an outlet ventricular septal defect in isolation or in combination with other malformations, and subaortic membranes, both whether repaired or unrepaired. Reference McMahon, Sendžikaitė and Jegatheeswaran10,Reference Egbe, Crestanello, Miranda and Connolly11
Taken together, those with CHD requiring surgical intervention on their systemic arterial root represent a large, underappreciated population burden. The traditional surgical approach is poorly standardised and relies on basic two-dimensional imaging assessment of the complex, dynamic three-dimensional structure of the aortic or neo-aortic root. The surgeon primarily relies upon their own intraoperative visual inspection of the root and its valve in a limited field of view and non-loaded state. Reference Tretter, Izawa and Spicer12 The general decision to repair or replace the valve is determined and common surgical techniques are then executed based on the surgeon’s interpretation of the overall qualitative assessment. Suboptimal mid- to long-term outcomes following repair have resulted with reported freedom from aortic reintervention ranging from 40 to 80% at 10 years, and median time to reintervention of 4 to 6 years. Reference Sengupta, Gauvreau and Marx13–Reference Rao, Van Arsdell, David, Azakie and Williams16 Mechanical aortic valvar replacement has a high rate of re-intervention in children and younger adults, with risks overshadowed by the need for lifelong anticoagulation. Replacement with a bioprosthetic valve is often avoided due to limited durability seen in this younger cohort. Reference Bauser-Heaton, Barry and Hofferberth17,Reference Hsu, Karnakoti and Abdelhalim18
More recently, promising repair outcomes have been observed in the congenitally malformed aortic valve when guided by quantitative assessment of its root, leaflet geometry, and measures of leaflet coaptation. Reference Ehrlich, Abeln, Froede, Schmitt, Giebels and Schäfers19–Reference De Paulis, Chirichilli and de Kerchove22 Normative data have been established by CT assessment in adults, allowing extrapolation of the proportional relationships towards children. In addition, relationships between these different valvar metrics with the standard planar dimensions within the aortic root and how they relate to maintain physiological leaflet coaptation have been established. Reference Izawa, Mori and Tretter23 Our group has furthered this approach by clarifying accurate imaging assessment of this geometry, Reference Izawa, Mori and Tretter23,Reference Mori, Izawa, Shimoyama and Tretter24 accounting for the variable relationships between the aortic root and its underlying support, Reference Amofa, Mori and Toh25–Reference Toh, Mori and Tretter27 and instructing how these relationships, in combination with this geometric quantitative assessment, may guide surgical personalisation in a broad range of congenital (neo)aortic root malformations. Reference Tretter, Spicer and Franklin3,Reference Tretter, Izawa and Spicer12,Reference Tretter, Burbano-Vera and Najm28 Furthermore, we have illuminated the understanding, and the importance of, the aortic hemodynamic ventriculo-arterial junction, and standardised a three-dimesional approach towards interrogating this junction to further guide surgical personalisation. Reference Burbano-Vera, Alfirevic and Bauer29 In this study, we reviewed the application of this CT-based approach towards surgical planning and execution in this heterogeneous population undergoing surgery for congenital aortic or neo-aortic valvar and root disease and report on the short-term outcomes.
Methods
Patients evaluated in our Congenital Valve Procedural Planning Centre who underwent aortic valvar surgery from February 2022 to August 2024 were included in this study. The study was approved by our institutional review board. Prior to entering the operating room, evaluation for surgical planning included two- and three-dimensional transthoracic echocardiography, and cardiac CT if intravenous contrast was not contraindicated. Two- and three-dimensional transesophageal echocardiography were obtained during the pre- and post-operative assessment. Standard transthoracic echocardiographic evaluation was obtained at post-operative follow-up visits, with the timing of follow-up determined by the primary cardiologist. Evaluation and grading of aortic valvar stenosis and regurgitation by echocardiography were performed as per the recommendations of the American College of Cardiology and American Heart Association, Reference Otto, Nishimura and Bonow30 considering an indexed vena contracta in those less than 18 years of age. Reference Colan and Sleeper31 Leaflet substrate was assessed pre-operatively, including the degree of leaflet thickening and presence and degree of leaflet calcifications in considering valvar repair versus replacement approach. Subsequent cardiac CT, when obtained, was also utilised to assess leaflet substrate.
Cardiac CT
A Siemens Dual Force or Siemens Naetom CT scanner was utilised, and the acquisition protocol was determined based on the underlying aortic root disease (Supplementary Table 1). Analysis of the acquired three- or four-dimensional CT data with volume-rendered three- and four-dimensional reconstructions to aid in anatomical visualisation was performed using Ziostation2 (Ziosoft USA, inc., CA). Dimensions of the aortic root, including the virtual basal ring (Figure 1A), widest sinus-to-sinus measurement, and sinutubular junction, were measured both in peak systole (typically 20 or 30% R-R) and the quiescent period of mid-diastole (typically 70 or 80% R-R). Measurements of the leaflet geometry and leaflet coaptation were also measured in the quiescent period of mid-diastole. These measurements included the geometric height, free margin length, commissural height, effective height, and if central leaflet coaptation were present, the coaptation length (Figure 1B–D). The rotational position of the aortic root relative to the base of the left ventricle was quantified in diastole by measuring an angle between the midline of the roof of the atrial buttress and a line connecting the centre of the non-coronary sinus to the intercoronary commissure. The impact of this rotational position was determined by relating the planes of the major and minor dimensions of the virtual basal ring in diastole compared to their relationship to the overlying sinuses and leaflets. Volume-rendered three-dimensional reconstructions were created of the blood-filled lumen of the aortic root and proximal ascending aorta (endocast reconstructions), and virtual dissections of the aortic valve and relevant left ventricular outflow tract and aortic root anatomy. When four-dimensional acquisition was obtained, dynamic visual inspection of the aortic valve in its short axis, and long axis interrogation of each commissure, was performed. The measured aortic dimensions, leaflet geometry, and leaflet coaptation were overlaid on representative reconstructions. In addition, each commissure was inspected throughout the cardiac cycle to assess for pathology impacting the zones of apposition. Reference Burbano-Vera, Alfirevic and Bauer29 The information was reviewed and discussed by the interrogating cardiac imager, cardiac surgeon, and referring cardiologist to guide a personalised surgical blueprint based on the principles in Supplementary Table 2.
Cardiac computed tomography methods using multiplanar reformatting. (a) Short axis of the aortic virtual basal ring, or “annulus” is obtained for measurement of the major and minor axis. (b) In the short axis of the aortic root, an orthogonal long axis plane is created from the midline of a leaflet across the opposite zone of apposition (red line). (c) The created centre bisecting plane is depicted, demonstrating the nadir of attachment and midline of one leaflet, with the double shadow of the zone of apposition between the other two leaflets. (d) The same image is displayed with the aortic virtual basal ring (green line) and sinutubular junction (white line) marked. Measurements of the aortic leaflet geometry and measures of leaflet coaptation are displayed. LCS = left coronary sinus; NCS = non-coronary sinus; RCS = right coronary sinus.
Statistical analysis
Categorical variables are presented as counts and percentages and were compared using Wilcoxon signed-rank test. Continuous variables are presented as mean ± standard deviation for normal distribution and were compared using the paired t-test. A two-sided p-value of <0.05 was considered statistically significant. The statistical analysis was performed using IBM SPSS Statistics (version 29.0, IBM Corp, Armonk, NY).
Results
Seventy-three patients were evaluated and underwent surgery (median age 26.0 years, interquartile range 19 – 44; 65.8% males). Seven patients (9.6%) had previous aortic valvar repair, and five patients (6.8%) had prior balloon aortic valvoplasty. The underlying diagnosis is reported in Table 1. Of note, of the eleven patients with a functionally unileaflet aortic valve, seven (63.6%) were misdiagnosed as a bileaflet valve prior to referral into our valve programme. Nine of these eleven patients had the typical morphology of a functionally unileaflet aortic valve with a variable degree of fusion between both the coronary leaflets and the right and non-coronary leaflets, while two patients had unusual morphology involving fusion between the right and non-coronary leaflets and the left and non-coronary leaflets. Of the 11 patients with left ventricular outflow tract obstruction from subaortic membrane (9 patients) or long-segment tunnel-like obstruction (2 patients) with associated aortic regurgitation, 4 (36%) had moderate or greater aortic regurgitation.
Patent demographics, diagnoses, and surgeries performed
AVSD = atrioventricular septal defect; DORV = double outlet right ventricle; IQR = interquartile range; LVOT = left ventricular outflow tract; PS = pulmonary stenosis; TGA = transposition of the great arteries; VSD = ventricular septal defect.
The primary indications for aortic root surgery are reported in Table 1. Some degree of aortic regurgitation was present in 57 patients (78.1%) with 45 patients (61.6%) having moderate or greater regurgitation. Some degree of valvar and/or subvalvar aortic stenosis was present in 36 patients (49.3%) with a peak gradient of 33.2 ± 31.3 mmHg and mean gradient of 16.9 ± 10.7 mmHg. Twenty-five patients (34.2%) had moderate or greater stenosis (Tables 2 and 3).
Pre- and post-operative transthoracic echocardiographic assessment of aortic valvar function in patients receiving aortic valvar replacement
Pre- and post-operative transthoracic echocardiographic assessment of aortic valvar function in patients receiving aortic valvar repair
Sixty-five patients underwent pre-surgical cardiac CT evaluation. Pre-operative CT assessment was not obtained in the remaining eight patients either related to kidney disease precluding intravenous contrast, or in those with a well-functioning aortic valve, or primary subvalvar or valvar stenosis where three-dimensional echocardiographic assessment was deemed adequate. Supplementary Table 3 reports the mean pre-operative aortic root dimensions. Supplementary Table 4 summarises the pertinent findings by pre-surgical CT assessment and resulting actionable surgical plan. Figures 2–5 demonstrate common findings in the assessed subcohort populations.
Common features of a functionally bileaflet aortic valve with symmetrical commissures and significant aortic regurgitation. (a) The short axis of the aortic valve is displayed with the underlying dilated (3.7 x 2.6 cm) aortic virtual basal ring depicted (green oval line). The normal commissures are 170 degrees apart. The geometric heights of the functional leaflets are equal (blue hashed lines) with significant increased free margin length of the fused leaflets relative to the unfused left coronary leaflet (yellow hashed lines). (b) Tilting the short axis plane reveals significant prolapse of the posterior aspect of the fused leaflet (or non-coronary leaflet) resulting in severe coaptation deficiency. (c) Long axis view demonstrates prolapse of the non-coronary leaflet (NCL) with linear bending through its midportion resulting in severely low effective height (1 mm) relatively to the mildly low effective height (6 mm) of the left coronary leaflet (LCL). (d) In systole, there is lower position, or malalignment, of the NCL attachment (red arrow) at the commissure relative to the LCL attachment (black arrow).
Common features of the dilated neo-aortic root in transposition of the great arteries following the arterial switch procedure. (a) The neo-aortic root is depicted. (b) Transparency of the neo-aortic root walls demonstrates the commissures of the valvar leaflets and native sinutubular junction (white line) to be inferior to the re-implanted coronary arteries (re-implanted origins outlined in red), with severe effacement of this plane (4.2 cm, Z-score + 10.8 in this 14-year-old patient) and dilation extending to a tubular waste, or neo-sinutubular junction (white hashed line) which likely represents the surgical anastomosis site. The aortic virtual basal ring is dilated (3.4 x 3.4 cm). (c) The short axis of the neo-aortic valve demonstrates asymmetric enlargement of the posterior, or non-adjacent sinus and leaflet. There is a severe, central coaptation definiency. (d) Tilting the valve, it becomes more apparent that there is mild prolapse of this posterior leaflet, which is supported by the lower measured effective height (4 mm compared to 8 mm for the anterior leaflets).
Common features in aortic regurgitation in the setting of an outlet ventricular septal defect. Short axis of the trileaflet aortic valve in a 19-year-old patient with small perimembranous outlet ventricular septal defect (outlined in orange) demonstrates the underlying eccentrically dilated aortic virtual basal ring outlined in green (3.0 x 2.8 cm) related to prolapse of both the right coronary leaflet and its sinus wall into the defect. The right coronary leaflet (RCL) is deformed and elongated demonstrated by both its geometric height (blue hashed line) and free margin length (yellow hashed line) with unequal effective heights between the three leaflets. This results in a significant coaptation deficiency primarily between the coronary leaflets. LCL = left coronary leaflet; NCL = non-coronary leaflet.
Heterogeneity in (neo-)aortic leaflet morphology. (a) The more typical form of the functionally unileaflet valve is demonstrated with fusion between the coronary leaflets and the right (RCL) and non-coronary leaflets (NCL) (purple arrows) with normal commissure between the NCL and left coronary leaflet (LCL). (b) Two patients were encountered with atypical forms of a functionally unileaflet valve with the normal commissure between the coronary leaflets (yellow arrow). A quadrileaflet truncal valve is demonstrated in systole (c) and diastole (d). In diastole, there is central leaflet crowding resulting in central coaptation deficiency. Variable number of leaflets and leaflet fusion can be seen in the truncal valve. (e) In this example, while there are four leaflets and sinuses, the anterior leaflets are fused together so the valve functions as a trileaflet valve within the quadrisinuate root. The zone of fusion and raphe is evident between the anterior leaflets (purple arrow). (d) This morphology was not appreciated prior to pre-surgical interrogation; however, interrogation of the neo-aortic (truncal) root demonstrated a corresponding hypoplastic interleaflet triangle (purple hashed lines) and corresponding diminished commissural height, aided in confirming the described morphology. LCA = left coronary artery; right coronary artery; VSD = ventricular septal defect.
Forty-eight patients (65.8%) underwent some form of aortic valvar repair. Of these, twenty-two patients underwent a valve-sparing aortic root replacement with leaflet repair, three of which were patients with history of Ross operation. Of these forty-eight patients, seventeen (35.4%) underwent annuloplasty (most commonly external annuloplasty) with leaflet repair, two of which were functionally bileaflet aortic valves who underwent tricuspidisation; two paediatric patients (4.2%) whose body size precluded an adult-sized annuloplasty underwent leaflet repair without annuloplasty; and seven patients (14.6%) underwent subaortic membrane resection with leaflet rehabilitation. Twenty-five patients (34.2%) underwent some form of aortic valvar replacement, all of which was determined during the pre-operative imaging assessment, as outlined in Table 1. Overall, all patients underwent the general repair or replacement strategy which was determined during the pre-operative imaging, primarily guided by cardiac CT, when obtained.
The mean follow-up period following surgery was 4.2 ± 6.1 months. At the most recent follow-up assessment, only one patient (1.4%) had moderate or greater aortic valvar stenosis, with 57 patients (78.1%) having no stenosis. This single patient was a three-year-old with severe tunnel-like subaortic stenosis whose pre-operative imaging evaluation did not include cardiac CT. The peak gradient at follow-up was 16.9 mmHg ± 10.7 mmHg with a mean gradient of 9.5 ± 6.4 mmHg. Only three patients (4.1%) had moderate or greater regurgitation, with 51 patients (69.9%) having no regurgitation (Tables 2 and 3). Those with moderate post-operative regurgitation included: a 20-year-old with repaired Tetralogy of Fallot, right coronary leaflet vegetation with severe aortic regurgitation, severe developmental disabilities, and seizure disorder precluding mechanical aortic valve replacement for seventh time median sternotomy with difficulty in deep root dissection necessitating internal aortic annuloplasty; a 22-year-old with repaired common arterial trunk and severe truncal valvar regurgitation with functionally trileaflet and quadrisinuate truncal root with diffuse mild-to-moderate leaflet thickening and retraction; and a 42-year-old with the history of Ross procedure presenting with severe neo-aortic root dilation and moderate valvar regurgitation with diffuse mild-to-moderate leaflet thickening and retraction. In the first patient, the integrity of the internal annuloplasty seemed to be compromised at follow-up, while in the second two patients, the degree of leaflet free margin thickening despite attempts at leaflet thinning appeared to compromise leaflet coaptation. One patient who underwent repair of a functionally unileaflet aortic valve had dehiscence of the valvar repair within the second post-operative day due to an unexpected severe rise in blood pressure requiring returning to the operating room where a Ross procedure was undertaken with excellent result. A second patient who underwent tricuspidisation of a functionally bileaflet aortic valve with very asymmetrical commissures had dehiscence of the valvar repair within the first post-operative day and had re-repair performed with mild residual aortic regurgitation. Of note, both patients were younger children, and due to concerns about allowing proper growth, an aortic annuloplasty was not performed.
Discussion
A semi-quantitative and objective, CT-based approach towards pre-surgical evaluation in the common, heterogeneous paediatric and adult population encounter with congenital aortic or neo-aortic root disease may provide an improved, standardised framework for pre-surgical planning and execution (Supplementary Table 2, Supplementary Figure 1). Reference Tretter, Burbano-Vera and Najm28 Application of this proposed framework has led to excellent short-term results, with two-thirds of patients undergoing aortic valvar preservation, and only four patients (5.5%) having moderate or greater residual aortic valvar stenosis or regurgitation. Of note, the limited alternative replacement options were suboptimal in these four relatively complex cases, necessitating attempts towards valvar repair despite the anticipated difficulties. This three-dimensional, geometric approach towards evaluation, appropriate patient selection, and repair has been pioneered by Schafers and colleagues, Reference Ehrlich, Abeln, Froede, Schmitt, Giebels and Schäfers19–Reference Matsushima, Heß, Lämmerzahl, Karliova, Abdul-Khaliq and Schäfers21 and El Khoury, de Kerchove and colleagues. Reference Jahanyar, de Kerchove and El Khoury32–Reference Jahanyar, Aphram and Munoz34 We have adopted this approach of evaluating the geometry of the aortic root and its leaflets, with established geometrical relationships of the normal aortic root providing the benchmark for surgical reconstruction. Reference Tretter, Izawa and Spicer12,Reference Izawa, Mori and Tretter23,Reference Tretter, Burbano-Vera and Najm28 However, their reported evaluation has been primarily based on pre-operative echocardiographic and intraoperative assessment, both limited in their evaluation. Cardiac CT has the advantage of high, consistent spatial resolution, evaluating the aortic root and its valve in its native hemodynamic state, and providing improved visualisation of the dynamic three-dimensional root compared to its ventricular support. Reference Tretter, Izawa and Spicer12,Reference Toh, Mori and Tretter27,Reference Tretter, Burbano-Vera and Najm28 In addition, this evaluation can be done days to months before the operation, providing non-pressured, thoughtful discussions both between the evaluating imager and surgeon, as well as appropriate education and informed shared decision-making with the patient and family.
The decision on repairability of the aortic valve initially hinges on the quality of the leaflet substrate with which the surgeon has to work with. This commonly can be determined by standard transthoracic echocardiographic assessment of leaflet thickness, rigidity, and presence of calcifications. We have adopted a low threshold to obtain CT for further assessment, with improved interrogation of leaflet thickness and calcifications, along with the potential benefits provided by detailed aortic root dimensions, quantification of the leaflet geometry, and measures of coaptation. Reference Tretter, Izawa and Spicer12,Reference Tretter, Burbano-Vera and Najm28 For example, significant leaflet retraction, quantitatively assessed by its geometric height, and comparison to the dimensions of the virtual basal ring parallel to the midline of the impacted leaflet may provide more objective understanding towards the need for leaflet augmentation. The anticipated need for more extensive leaflet augmentation with patch material may weigh into decision-making to pursue a replacement strategy. Reference dUdekem, Siddiqui and Seaman15 Any replacement approach will benefit from an accurate assessment of the virtual basal ring plane, including considering graft or annuloplasty sizing. In the context of the Ross procedure, comparison to the virtual basal ring of the pulmonary root can also be achieved with proper timing of the intravenous contrast. Considering the Ozaki procedure, dilation of the virtual basal ring and sinuses may limit the durability of this replacement approach. Reference Frankel, Robinson and Roselli35 Additional interrogation of the coronary arteries will benefit any option requiring their reimplantation (Supplementary Figure 2).
The most common aortic valvar abnormality encountered is the functionally bileaflet aortic valve. In this malformation, the angular relationship of the normal commissures and height of the fused commissure relative to the normal commissural heights are the primary determinants towards guiding any repair strategy. Reference Tretter, Spicer and Franklin3,Reference Jahanyar, El Khoury and de Kerchove36 When the commissures are symmetrical (160–180 degrees), maintaining the bileaflet configuration has demonstrated the most durable results. Reference Ehrlich, Abeln, Froede, Schmitt, Giebels and Schäfers19,Reference Jahanyar, El Khoury and de Kerchove36 Central plication to reduce the free margin length of the fused leaflets may be necessary when significantly larger than the non-fused leaflet (Figure 2). In addition to thinning the zone of fusion and its raphe, more extensive leaflet rehabilitation and commissural support strategies may be required. Reference Xiang, Chen and Chemtob37 It is common for fibrotic tissue from a subaortic membrane to extend onto the ventricular surface of one or more of the leaflets, and for extension of any fibrotic tissue from an underlying ventricular septal defect patch to extend onto the ventricular surface of any adjacent leaflets. Reference Najm, Dakik, Barodi, Costello, Ahmad and Tretter38 Less commonly, when the commissures are very asymmetrical (less than 139 degrees) and the fused commissure is well developed (>50% the height of the normal commissures), we have adopted a personalised tricuspidisation approach. This involves using the free margin lengths and differences in commissural heights to pre-fabricate a fixed-pericardial butterfly patch. The zone of fusion is divided and the butterfly patch is used to augment the adjacent leaflet free margins and resuspend the neo-commissure equal to the height of the normal commissures. Reference Najm, Barodi, Costello, Ahmad, Dakik and Tretter39 However, current data suggest that the introduction of extensive patch material, pericardial, or synthetic, is a risk factor for decreasing repair durability, Reference dUdekem, Siddiqui and Seaman15,Reference Ehrlich, Abeln, Froede, Schmitt, Giebels and Schäfers19,Reference Froede, Abeln, Ehrlich, Feldner and Schäfers20 and some support maintaining a bileaflet configuration. Reference Jahanyar, de Kerchove and El Khoury32 Whether or not these patient-specific measurements towards leaflet augmentation improves the durability of any tricuspidisation approach requires longer-term follow-up. Furthermore, with our growing experience in these less common valves with very asymmetric commissures, we would lean towards the Ross procedure in the pre-adolescent and younger teenager, where complete stabilisation of the virtual basal ring cannot be performed due to growth considerations.
Similar considerations are given to the functionally unileaflet aortic valve, in considering conversion to a bileaflet arrangement. Reference Matsushima, Heß, Lämmerzahl, Karliova, Abdul-Khaliq and Schäfers21,Reference Jahanyar, Aphram and Munoz34 Misclassification as a bileaflet valve is common, identified in approximately two-thirds of patients in the current study (Figure 5A, B). If the fused commissure more closely positioned symmetrical to the normal commissure is greater than half the height of the normal commissure, and the patient is a child or young adult with good leaflet substrate, we would consider repair. However, if the assessed fused commissure is less than half the normal height, and especially in the setting of more thickened and calcified leaflets, and/or in the older adult patient (greater than 30 years of age), we would be less enthusiastic to pursue a repair strategy, with increasing advantage in the Ross procedure. Reference Abeln, Matsushima, Ehrlich, Giebels and Schäfers40 With the same concerns related to leaflet augmentation in both children and adults, Reference Patel, Unai and Moore41,Reference Baird, Cooney, Chávez, Sleeper, Marx and Del Nido42 we have also transitioned towards considering the Ozaki procedure only in select patients. Those considered often possess a primarily stenotic valvar lesion with normal aortic root dimensions, poor leaflet substrate for repair, and approaching mid-adult years.
The truncal valve is highly variable in the number of leaflets and sinuses, and the involvement of leaflet fusion. Accurate morphological description serves as the basis for appropriate surgical planning (Figure 5C–F). Reference Tretter, Spicer and Franklin3 In general, when repair is deemed feasible, we utilise the same principles discussed above for determining an appropriate repair approach, often aiming to maintain relatively equal functional leaflet units. In our younger paediatric population, we would consider single-leaflet pericardial reconstruction if replacing one functional leaflet unit would restore general leaflet symmetry. Reference Najm, Zaki, Ahmad, Pettersson and Karamlou43
Regardless of the congenital substrate, when feasible in the adult or adult-sized child, the dilated virtual basal ring will benefit from normalisation in dimensions and reinforcement. Reference Ehrlich, Abeln, Froede, Schmitt, Giebels and Schäfers19,Reference Froede, Abeln, Ehrlich, Feldner and Schäfers20 During root replacement, this is achieved by the inserted graft. In the absence of the need for root replacement, this can be achieved by an annuloplasty. Our preference has been to perform an external annuloplasty with a Dacron ring when possible, though data are lacking to support one technique over another. Reference Federspiel, Ehrlich, Abeln and Schäfers44 The geometric heights and desired effective height and coaptation length of the leaflets roughly may be used to guide the ideal dimensions of the virtual basal ring using the following equation:
(annulus/2)2 + (effective height − coaptation length)2 = (geometric height − coaptation length)2. Reference Tretter, Burbano-Vera and Najm28
When maintaining a bileaflet configuration in the setting of root replacement, achieving a ratio between the free margin length and aortic graft dimension of approximately 1.5 may provide favourable hemodynamics. Reference Choi, Sharir and Ono45 Similarly, if the sinutubular junction is dilated, this may be stabilised, aiming to achieve approximately equal dimensions to the virtual basal ring plane to restore normal morphometric relationships. Reference Dudkiewicz, Lis and Yakovliev46 Ascending aortic graft replacement, when indicated, will alternatively accomplish this stabilisation if extended proximally to the sinutubular junction plane.
Study limitations
The current study is limited in its retrospective nature with short-term follow-up. The cohort studied is heterogeneous both in age and underlying aortic disease, which prohibited the ability for meaningful quantitative assessment of the application of the described aortic root metrics and measures of leaflet coaptation.
Conclusion
The heterogeneity and complexity of the dysfunctional and/or dilated aortic or neo-aortic root encountered in those presenting for surgical repair or replacement necessitate a methodical, detailed three- and four-dimensional assessment. By applying such an approach, we have aimed to standardise not only the assessment, but also description and surgical execution in this challenging patient population. Excellent short-term results have been achieved, necessitating long-term follow-up to understand the potential benefits towards this personalised approach.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1047951125109475.
Acknowledgements
We are indebted to the outstanding CT technologists and congenital cardiac sonographers at the Cleveland Clinic involved in our efforts to optimise image acquisition protocols specific to patients evaluated in our Congenital Valve Procedural Planning Programme.
Financial support
None.
Competing interests
Justin T. Tretter is a consultant for Cara Medical, Ltd The remaining authors have nothing to disclose.
Ethical standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national guidelines on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008, and has been approved by the institutional review board at the Cleveland Clinic.
Open access
