Aortic valve replacement (AVR) may occasionally be required in infants and children. Common indications for AVR in children include the following:
Hypoplasia of the aortic valve annulus in the neonate
Progressive stenosis of the aortic valve in infants and children
Multilevel left ventricular outflow tract obstruction in association with aortic valve stenosis not amenable to aortic valve repair that requires enlargement of the outflow tract
Aortic insufficiency as a complication of percutaneous balloon aortic valvuloplasty
Rheumatic aortic valve disease
Aortic valve endocarditis
Although no aortic valve substitute is ideal, desirable characteristics include the following:
Excellent flow hemodynamics with reduction of left ventricular afterload and/or preload to normal values
Lifetime durability without need for reoperation due to structural deterioration
Absence of need for anticoagulation
Absence of risk for late embolic complications
Capability of growth to avoid patient prosthesis mismatch as the child gets older
Resistance to infection and endocarditis
Ease of implantation
Appropriate size availability for all patients
Several aortic valve repair techniques have been used in children, including pericardial leaflet extension, commissural reconstruction, annuloplasty, sinus of Valsalva reduction, sinotubular junction remodeling, and even complete leaflet replacement using autologous pericardium. Aortic valve repair in the child allows for continuing growth and eliminates the need for anticoagulation. However, long-term results have been less than satisfactory, and residual lesions (eg, regurgitation, stenosis) are common.
Despite evidence of stabilization of ventricular dimensions and improvement of functional classification in children following aortic valve repair, the residual lesions continue to progress, with increased regurgitant fraction and/or peak gradients across the left ventricular outflow tract; eventually, patients require reoperation and possible valve replacement. The role of valve repair in children as a curative or a temporizing measure remains incompletely defined. Nevertheless, aortic valve repair delays ultimate replacement until alternative valve replacement options can be offered to patients after completion of somatic growth, pregnancy and increased compliance with anticoagulation regimen.
Mechanical valve prostheses are not ideal valve substitutes in children. Although the incidence of structural valve deterioration is negligible, these prostheses have significant limitations at the time of implant due to the lack of appropriately sized prostheses for small children and neonates. In addition, the absence of potential for growth can result in patient-prosthesis size mismatch as the child grows and may require re-replacement. Moreover, mechanical valves require lifetime anticoagulation with associated activity limitations, difficulties with future pregnancy, and a lifetime risk of thromboembolic and bleeding complications due to potential poor compliance with anticoagulation protocol.
Homografts and bioprosthetic valves are also problematic in children. Although these biologic valves do not require anticoagulation, they do not allow growth, and their durability in the pediatric population is very limited due to the high risk of accelerated structural valve degeneration and early calcification. In addition, the availability of appropriate sized homografts and bioprostheses can be a problem.
The Ross procedure using pulmonary autograft provides excellent hemodynamics flow characteristics, is capable of growth, and does not require anticoagulation. Despite several shortcomings of the Ross procedure, it has emerged as a popular choice for AVR in infants and children. This review focuses on the role of the Ross procedure in the treatment of aortic valve disease in children and young adults.
The procedure of replacing the aortic valve with the patient's own pulmonary valve and then using a pulmonary allograft to replace the pulmonary valve is commonly referred to as the Ross procedure. Lower, Stofer, and Shumway investigated the concept in 1960 using autotransplantation of the pulmonic valve into the descending thoracic aorta of dogs.[1] Pillsbury and Shumway described autotransplantation into the aortic annulus in 1966.[2] However, the first clinical application occurred 1 year later, as reported by Donald Ross in 1967 (see the image below).[3] Since then, the operation has steadily gained acceptance, and the indications for the procedure have expanded.
Thorough knowledge of cardiac anatomy is required to perform the Ross procedure. In particular, an understanding the left coronary artery, the first septal branch of the left anterior descending artery, and their relationship to the aortic root and the right ventricular outflow tract is important (see the image below). During harvest of the pulmonary autograft, an appreciation of the subpulmonary conal musculature (ie, the thin muscular tube beneath the pulmonary valve) facilitates the dissection. Knowledge of the configuration of the left ventricular outflow tract and the relationship to the conduction system is important when enlargement of the left ventricular outflow tract is required (Ross-Konno procedure).
The Ross operation is a technically demanding procedure, and the surgeon's experience with this operation and similar procedures affects the decision-making process. Patient factors that affect this process include the patient's age, lifestyle, and coexisting cardiac and noncardiac disease.
Contraindications include the following:
Pulmonary valve pathology
Known genetic defects in fibrillin, elastin, or collagen in connective tissue disorders (eg, Marfan syndrome, Ehlers-Danlos syndrome)
Significant immune complex disease as a coexisting disease, especially if it is the etiology of the aortic valve disease. (eg, systemic lupus erythematosus, ankylosing spondylitis, Reiter disease)
Advanced 3-vessel coronary artery disease
Significant irreparable mitral valve pathology that requires mechanical valve replacement (considered a relative contraindication by many surgeons)
Significant dilatation of the aortic root in comparison to the pulmonary valve annulus associated with aortic regurgitation (considered a relative contraindication by many surgeons, whereas others may continue to offer the Ross procedure along with aortic annular reduction)
Echocardiography is used preoperatively to assess the aortic valve pathology, levels of left ventricular outflow tract obstruction and associated cardiac abnormalities. The pulmonary valve is assessed for clinically significant regurgitation or any other pathology. Echocardiography is also useful for assessing the sizes of the aorta and pulmonary annulus. A disparity in size of more than 2-3 mm is likely to require augmentation or reduction in the diameter of the aortic annulus.
All procedures are performed though midline sternotomy. Cardiopulmonary bypass is established via standard aortic and bicaval venous cannulation. The left ventricle is decompressed by venting through the right superior pulmonary vein. Mild hypothermia (32-34º) is used with a combination of antegrade and retrograde cold blood cardioplegia. Antegrade cardioplegia is initially administered through the root and then by direct coronary artery cannulation at 20-minute intervals.
The aorta is transected 1.5 cm above the right coronary artery. The aortic valve is inspected and repaired, if possible. If the valve is not repairable, the leaflets are then completely excised and calcium is debrided if present. The main pulmonary artery is partially opened just proximal to the bifurcation, and the valve is inspected to ensure normal anatomy and function.
Once the decision is made to proceed with the Ross procedure, the coronary buttons are prepared. A generous rim of aorta is left around each ostium to allow for suturing to the pulmonary autograft later.
The pulmonary artery is separated from the aorta up to the bifurcation and is completely divided. The autograft is harvested by placing a right-angled clamp through the valve and by bringing the tip through the infundibulum approximately 1 cm below the base of the cusps. The right ventricular outflow tract is then opened circumferentially using scissors. Once the dissection proceeds laterally, the left anterior descending artery and its first septal branch are at risk if meticulous dissection is not performed. Following harvesting of the autograft, retrograde cardioplegia is administered and small venous branches are cauterized or ligated in the bed of the harvested autograft. See the images below.
The autograft and the right ventricular outflow tract are then sized with standard sizers to select an appropriate-sized pulmonary homograft to be prepared. The aortic root annulus is also sized to determine if any discrepancy needs to be addressed. An annular size difference of 2-3 mm is well tolerated. If the aortic annulus is too large, reduction is best achieved with an imbricating suture passed circumferentially at the level of the annulus and tied over a dilator the size of the pulmonary autograft. Alternatively, a series of mattress sutures can be used with care to avoid the region of the conduction system. If the aortic root annulus is too small, then an aortoventriculoplasty combined with the Ross procedure (commonly known as a Ross-Konno procedure) is appropriate. See the images below.
The autograft is sutured to the aortic valve annulus using either a running or interrupted 4-0 polypropylene suture. If no further growth is required, the sutures are tied around a circumferential strip of Teflon felt approximately 3 mm wide, as shown below. The graft should be orientated so that the commissures of the autograft line up with the commissures of the excised aortic valve. A small opening is made in the left coronary sinus of the autograft, and the left coronary artery is anastomosed using a running 6-0 polypropylene suture. The distal aortic anastomosis is then constructed with a continuous 4-0 polypropylene suture.
The aortic root is deaired and insufflated to test the suture lines and to allow proper placement of the right coronary artery once the autograft is distended. The anastomosis is constructed in a similar fashion as the left coronary button. Antegrade cardioplegia can now be administered, and bleeding in the bed of the harvested autograft site can be addressed.
A cryopreserved pulmonary homograft is then appropriately trimmed, and the distal anastomosis is performed using a continuous 4-0 polypropylene suture. The proximal anastomosis is then constructed with continuous 5-0 polypropylene. See the image below.
The patient is then placed in steep Trendelenburg position. While the aortic and left ventricular vents are aspirated, the cross-clamp can be removed. The remainder of the anterior portion of the homograft anastomosis can be completed with the heart beating.
The patient is then weaned from cardiopulmonary bypass; protamine is administered, and the patient is decannulated. Transesophageal echocardiography is used to assess the function of the autograft and the homograft once the procedure is complete, as shown below.
The autograft implantation technique described is the miniroot reimplantation technique. It is the preferred implantation strategy used at the authors' institution. Other implantation techniques are similar to those described for homografts, such as the subcoronary and the cylinder inclusion techniques, and are preferred by some surgeons. In the Ross registry database, 81% of autografts were implanted using the root technique, whereas 11% used the subcoronary technique, and 6% used the inclusion technique.[4]
Standard postoperative cardiac management is administered. Patients can generally be weaned from ventilatory support in the early postoperative period, the exception is a neonate who was critically ill before surgery.
Patients are examined 4 weeks postoperatively to address any surgical issues. They should also continue to undergo biannual echocardiography to assess function of the right- and left-sided semilunar valves. After undergoing surgical repair of aortic valve disease, patients are given antibiotics to prevent endocarditis before they receive any procedures that may cause bacteremia. For more information, see Antibiotic Prophylactic Regimens for Endocarditis.
The Ross procedure is a safe operation and can be performed with a low mortality.[5, 6, 7, 8, 9] Data from The International Registry for the Ross Procedure, which includes 6088 patients, reveals an early mortality of 3.3%.[4] Perioperative complications are uncommon and include arrhythmias in 3% of patients, bleeding in 2% of patients, stroke in 1% of patients, and sepsis in 1% of patients.
Many groups have reported even more favorable results in children and young adults undergoing the Ross procedure, with an operative mortality rate approaching 0% in different series from experienced centers despite the complexity of the procedure.
In a study that included approximately 420 children and nearly 275 young adults who underwent the Ross procedure, Zebele et al determined that the patients’ 30-day and 1-year survival rates were 99.3% and 98.7%, respectively. Another study, by Luciani et al, of 305 children who underwent the Ross operation, found the 10- and 15-year survival rates to be 93% and 89%, respectively, with no reoperation needed during those same periods for 76% and 67% of patients, respectively.[10, 11]
In a retrospective study, Schneider et al investigated the long-term outcomes in 154 patients who underwent the Ross (n=105) and Ross-Konno (n=49) procedure at their institution from 1994 to 2016, with a median age of 12 years. Survival rates were 86% in the total cohort and 91% in the isolated Ross subgroup at 15 and 20 years. The cumulative incidence of all-cause reoperation was 35.2% at 15 years and 45.3% at 20 years; the cumulative incidence of autograft reoperation was 20.1% at 15 years and 31.1% at 20 years.[12]
Using a national database, Sharabiani et al investigated outcomes in a total of 1,501 patients who underwent aortic valve replacement (AVR) in the United Kingdom between 2000 and 2012 (47.8% underwent a Ross procedure, 37.8% a mechanical AVR, 10.9% a bioprosthesis AVR, and 3.5% a homograft AVR). When compared with mechanical AVR, the Ross procedure had a 12.7% higher event-free (death or any reintervention) probability at 10 years (P = 0.05) in children. In a comparison of the Ross procedure, mechanical AVR, and bioprosthesis AVR in a matched population of young adults, only the patients who underwent the Ross procedure approached the survival of the general population.[13]
Several recent retrospective studies that evaluated the mid- and long-term outcomes of patients who underwent the Ross procedure concluded it provides good clinical and hemodynamic results, with low reoperation rates.[14, 15, 16, 17] Actuarial long-term survival for patients undergoing the Ross procedure in the Ross registry and multiple surgical series is 80-90% at 10 years and 70-80% at 20 years. Those favorable results reflect the selection bias as patients undergoing the Ross procedure are usually young with minimal comorbidities such as coronary artery disease.
Early autograft failure or dysfunction that requires replacement of the autograft at the time of operative insertion or reoperation within 6 months of the original Ross procedure is very rare at experienced centers, with an incidence rate of less than 1% in the Ross registry. This failure or dysfunction is usually due to technical problems, such as leaflet distortion and leaflet injury during the harvest or the implantation of the autograft.
Pulmonary autograft is a durable replacement for the aortic valve. Initial data collected by Ross demonstrated an autograft failure and replacement rate of 2.5% per patient year and an actuarial event-free survival rate of 48% at 19 years.[3]
Several improvements in technique, including the use of the root replacement technique and procedures to limit potential autograft dilation, have resulted in improved outcomes for the Ross procedure.
In the most recent Ross registry, the rate of freedom from explantation of failing autografts at 10 years and 25 years is 89% and 82%, respectively. Valve durability is clearly superior to that of homografts or bioprosthetic valves because the latter two undergo calcific degeneration and require replacement, especially in younger patients with longer expected survival and accelerated early degeneration of biologic valve substitutes.
Moreover, the freedom from valve-related complications (eg, endocarditis, valve pannus, thrombosis) is very low following the Ross procedure, and valve explantation for these reasons is relatively uncommon compared with mechanical valves.
An increasing concern is dilatation of the neoaortic root following Ross procedure that leads to progression of aortic regurgitation, especially in the setting of geometric mismatch of aortic and pulmonary roots and bicuspid regurgitant aortic valve. These specific subsets of patients have been identified by several groups to have a higher risk of autograft dilatation and recurrence of regurgitation.
A study from Philadelphia examined serial echocardiograms following the Ross procedure in children and young adults.[18] Evidence of progressive neoaortic root dilatation out of proportion to somatic growth was reported, along with progressive aortic insufficiency, especially in patients with prior ventricular septal defect repair or prior aortic valve replacement. Those results highlighted the need for continuous monitoring of those patients for the potential future development of autograft complications.
Dilatation of the sinuses of Valsalva results in root aneurysm, even without significant aortic regurgitation, whereas dilatation of the sinotubular junction causes recurrence of regurgitation, especially in patients who underwent the root implantation technique of the autograft. Dilatation of the sinotubular junction is probably the most common cause of failure of the pulmonary autograft. Although valve replacement may be required, several groups have reported valve sparing replacement of the aortic root and ascending aorta to eliminate the aneurysmal dilated wall and restore valve competency.
Several studies reported better performance of the autograft in younger children compared with older children and adults, with a higher rate of freedom from autograft dilatation and reoperations.[18] These series may suggest the ability of the autograft of a young child to better adapt to systemic pressures. Nonetheless, the Ross procedure is undergoing reconsideration by many surgeons especially in the treatment of congenital and bicuspid aortic valve disease.
Finally, several groups have reported modifications of the valve implantation techniques, such as adjusting the diameter of the aortic annulus and/or the sinotubular junction of the aorta using Teflon strips, implanting the autografts into a Dacron graft, or wrapping the autograft with glutarylaldehyde-treated pericardium to prevent dilatation of the autograft. Further follow-up is needed to assess the use of these modifications.
The pulmonary homograft used for reconstruction of the right ventricular outflow tract is subject to calcific degeneration, which, in addition to its failure to grow with the child, likely requires reoperation and conduit replacement. Although the pulmonary homograft placed during the Ross procedure has greater longevity than the one used for reconstruction of the right ventricular outflow tract for repair of congenital heart disease (presumably due to the orthotopic position, normal pulmonary arteries, and pulmonary vascular resistance), recent data indicate that replacement is still necessary. Factors associated with homograft dysfunction include the use of aortic homograft, small homograft size, recipient age of less than 10 years, homograft storage time, blood-type disparity, and immune-mediated reaction.
The reported incidence of homograft dysfunction ranges from 6-20% at 10 years after the Ross procedure. The freedom from pulmonary homograft replacement at the Ross registry at 10 years and 25 years is 91% and 84%, respectively.
When re-replacement of the subpulmonary homograft is required, the operative results are associated with minimal mortality, therefore providing further support of the use of the Ross procedure for valve replacement in children and young adults.
Most importantly, experience with percutaneous pulmonary valve replacement has emerged in the past few years and has become a valid choice in many patients following the Ross procedure. Current exclusion criteria include unfavorable right ventricular outflow tract morphology (eg, narrowest right ventricular outflow tract diameter of >22 mm or conduits < 16 mm at time of surgical implantation), patient age less than 5 years, or weight less than 20 kg.
As no valve replacement substitute in children is ideal, studies compared outcomes between the Ross procedure and mechanical valve replacement in children.[19, 20, 17] Evidence of improved short-term and long-term mortality was reported in children who underwent the Ross procedure. This survival advantage, although present in all children, was most evident in younger children who required smaller mechanical aortic valve prostheses. Moreover, in addition to early mortality, survival was stable for children undergoing the Ross procedure with minimal late mortality, whereas a constant attrition rate was noted in children who received mechanical valves (due to sudden death and thromboembolic complications).
Although autograft survival remains an issue, especially in those with underlying rheumatic activity and aortic regurgitation, autograft longevity was superior in patients with underlying congenital etiology.
Freedom from endocarditis was comparable between the two valve substitutes; however, freedom from anticoagulation-related morbidity was obviously superior in children who underwent the Ross procedure.
Patients undergoing the Ross procedure do not require anticoagulation. The patients have minimal restrictions on their lifestyle and do not require cardiac medications to maintain or preserve valve function. Despite development of mildly elevated gradients across the pulmonary homograft, patients undergoing the Ross procedure have near normal exercise endurance, and most are in New York Heath Association (NYHA) Class I.
An echocardiographic study that compared rest and exercise hemodynamics after the Ross procedure showed that hemodynamic characteristics and exercise performance in athletes after Ross procedure were similar to those in age-matched healthy athletes.
Although the Ross procedure for aortic valve replacement (AVR) in the pediatric population is a more demanding procedure than straightforward valve replacement, it offers distinct advantages, including excellent hemodynamic flow characteristics, potential for growth, excellent patient survival, and minimal incidence of late embolic complications. Reoperations on the reconstructed right ventricular outflow tract are infrequent and are associated with low operative risk. A future role for percutaneous pulmonary valve replacement is noted. Surveillance for autograft dilatation is necessary because it may result in aneurysm formation and/or development of recurrent aortic regurgitation.
Reoperation on the autograft may be required; however, valve-sparing root replacement can be performed, preserving the autograft valve. The advantages of the Ross procedure, despite its limitations, make it the current preferred AVR choice in children. Improvements in aortic valve repair technique and developments in valve-substitute technology may offer a better alternative choice for children with aortic valve disease in the future.
There may be more controversy regarding the Ross versus mechanical valve replacement in the young adult population. The operative mortality for the Ross procedure is approximately 2-3%, whereas for isolated AVR in a young adult the risk may be 1% or less.[21]
In the second decade of follow-up after the Ross operation, the risk of reoperation on the autograft itself may approach 15-20% with approximately 5% requiring reoperation on the pulmonary homograft. The reoperations on the autograft are complex, such as a valve sparing aortic root replacement as mentioned above. Although reported mortality rates for reoperation on the autograft have been low, they are associated with major morbidity including postoperative ECMO and future reoperation. Furthermore, just less than 80% of these reoperations on the autograft required an aortic valve or aortic root replacement.[6]
In a propensity score-matched study assessing late survival in young adult patients after a Ross procedure versus those after mechanical AVR with optimal self-management anticoagulation therapy, Mokhles et al found that the linearized all-cause mortality rate was 0.53% per patient-year in the Ross procedure group compared with 0.30% per patient-year in the mechanical valve group (matched hazard ratio, 1.86; 95% confidence interval, 0.58-5.91; P = 0.32), with late survival comparable to that of the general German population.[22] Of note, valve related mortality (n=4) was seen only in the Ross group. Eight patients in the Ross group required valve replacement with none in the requiring reoperation in the mechanical valve group. The linearized bleeding rate was 0.15% per patient-year in the Ross procedure group compared with 0.36% per patient-year in the mechanical valve group (P = 0.15).
During follow-up, five Ross patients and one patient with a mechanical valve experienced a thromboembolic event. The linearized thromboembolism rate was 0.38% per patient-year in the Ross procedure group compared with 0.06% per patient-year in the mechanical valve group (P = 0.10). The linearized endocarditis rate was 0.15% per patient -year in the Ross procedure group compared with 0.00% per patient-year in the mechanical valve group (P = 0.16). The results of this study apply to only the first postoperative decade prior to the increasing reoperation hazard of the Ross procedure in the second decade taking effect. They hypothesize that late mortality after AVR is driven mainly by patient characteristics and that prosthesis selection plays only a minor role, if any.
Despite having an excellent track record, Zebele and colleagues note a decline in the use of the Ross procedure in children and young adults in the United Kingdom National Database.[10]