Continuing Education Activity
Aortic valve repair is performed because severe aortic valvular disease uncorrected mortality rates are dismal. Half of the patients presenting with aortic stenosis may be dead within two years. Surgical and non-surgical catheter-based approaches are utilized for aortic valve repair. Main repair indications for both stenosis and insufficiency include severe symptomatic disease and severe asymptomatic disease if the patient's ejection fraction is less than 50% or if the patient has a concomitant indication for cardiac surgery. This activity reviews the evaluation and treatment of aortic valve disease and highlights the role of the interprofessional team in evaluating and treating this condition.
- Identify the anatomical structures relevant to bioprosthetic aortic valve replacement.
- Describe the standard operative technique for performing a safe, durable bioprosthetic aortic valve replacement.
- Summarize potential complications of bioprosthetic aortic valve replacement.
- Outline some interprofessional team strategies for improving care coordination and communication to advance identify indications for bioprosthetic aortic valve replacement and improve outcomes.
Considering the constant stress on the four valves of the human heart, beating an average 80 times per minute for over 70 years for over 3 billion beats, on the whole, human heart valves perform amazingly well. Less than 2% of the population is estimated to suffer valvular disease.
The history of the treatment of valvular disease is marked by daring innovation and multidisciplinary collaboration; from the days of closed digital commissurotomies (sticking a surgeon’s finger through a heart valve) to homemade balls in cages to the pathophysiologic connection between treating strep throat with antibiotics and preventing rheumatic heart disease decades into the future. These historical episodes have something to teach about progress in healthcare. However, this article will focus on surgical aortic valve replacement (SAVR) for stenosis, insufficiency, or endocarditis using bioprosthetic valves.
The interested reader is invited to peruse closely related content, especially surgical aortic valve replacement, surgical aortic valve repair, stentless pulmonary autograft/homograft aortic valve replacement (Ross procedure), minimally invasive aortic valve surgery, and transcatheter aortic valve replacement (TAVR), as well as more general treatments of aortic valve disease and prosthetic valves.
Anatomy and Physiology
When considering the replacement of an aortic valve, it is essential to consider the form and function of the native valve. The aortic valve works in a no man's land, separating two remarkably different zones of pressure over a height of approximately 12 to 18 mm. The ventricular side contracts in systole while the aortic side expands; then, in diastole, the ventricular side relaxes while the aortic side contracts somewhat. The aortic valve must completely separate and insulate these two pressure zones from each other, then be able to rapidly reverse that separation and equalize the pressure gradient, oscillating between these two states fluidly. The aortic valve must also protect and promote blood flow to the coronary ostia.
The anatomic nomenclature surrounding the aortic root region can become quite subtle. On a tour of the region, following the flow of blood from the left ventricle, the first zone encountered is the ventriculo-aortic junction (VAJ), the transition zone from ventricle myocardium to aortic tissue. In diagrams depicting the aortic root in cross-section, where the three-valve leaflets appear as inverted rounded-tipped triangles, the VAJ is marked by a line connecting the inverted apexes of the valve leaflets. Continuing the tour, one observes the lines of those triangular valves rising and inserting onto the walls of the aortic tube from apex to base further distal. Where these valve attachments are inserting is referred to as the basal ring or surgical aortic annulus. The annulus is not a true annulus, as it is slightly ovoid and not uniform. This is because the size, rise, angulation, and attachment of each valve leaflet are slightly different, creating variability in the gap between the VAJ and the basal ring/annulus. The gap leads to areas of "ventricle in the sinus," giving the aortic root extra distensibility and ability to expand 18 to 20% in systole.
Continuing the tour, moving through the aortic valve leaflets, one observes the tops of their insertion points can also be seen as a ring referred to as the sinotubular junction (STJ), marking the end of the aortic root and the beginning of the ascending aorta proper. From the VAJ to the STJ, the lumen bulges slightly, before beginning to taper at the ascending aorta. Now that the tour has gone through the valves, by looking behind at the road traveled, one can observe that under this bulging, each of the valve leaflets accommodates a little hollow space or little eddy pool called a sinus. In the right sinus, there is the origin of the right coronary artery; in the left sinus, there is the origin of the left coronary artery; the third sinus (the posterior one) contains no coronary artery. The coronary origins are anomalous in 1% of the population, with more common variants including separate origins of the LAD and circumflex or a circumflex taking the right coronary as its origin.
The geometry of these sinuses allows the collection of eddy currents between the bulged wall of the aortic root and the valve leaflets, keeping the valve from getting stuck on the aortic wall and thus slightly open for blood flow into the coronaries and in a ready position to close instantly at the end of systole. The valve leaflets themselves are covered by endocardial endothelial cells and interspersed with valve interstitial cells (VIC) quiescently resting through the three layers: elastin rich ventricularis towards the ventricle, proteoglycan rich spongiosa in the middle, and collagen-rich fibrinosa toward the aorta.
The aortic valve has several surgically relevant anatomic relationships that must be born in mind to avoid injury. The non-coronary and the left coronary cusps of the aortic valve are structurally contiguous with the strongest fibrous tissue of the heart, the central fibrous body, but as such, they are also structurally continuous with the anterior leaflet of the mitral valve. Directly underneath the junction of the right coronary cusp and the non-coronary cusp is the ventricular membranous septum and the AV node. The right and left coronary cusps junction is externally continuous with the pulmonary artery. Quite proximal to the ostium of the right coronary is the branch point of the sinoatrial nodal artery. A network of tiny coronary veins may be present around the ostium of the left coronary.
The pathophysiology of the aortic valve either spring from stenosis or insufficiency (regurgitation) and both are a failure of the valve to insulate the two pressures zones, heart and aorta, from each other. A stenotic valve (less than half the normal 3-5 cm squared surface area) exposes the ventricle to a pressure gradient, leading to heart wall stress. Wall stress sets the heart up for a compensatory cycle of hypertrophy to minimize wall stress at the ultimate expense of impaired coronary blood flow, fibrosis, and more wall stress. An insufficient (or regurgitant) valve promotes an equalization of pressures between the heart and aorta, leading to wall stress and a compensatory cycle of diastolic relaxation, some hypertrophy, and ultimately impaired coronary blood flow and forward failure.
Aortic stenosis is the most common pathology of valves needing correction in developed countries, with most cases due to senile degenerative calcification of normal trileaflet valves or degeneration of congenital bicuspid valves. In low- and middle-income countries, rheumatic heart disease is still common and accounts for most valvular pathology. Rheumatic heart disease most commonly afflicts the mitral valve, but to a lesser extent, causes aortic stenosis. Aortic insufficiency needing correction is less common than stenosis and may be caused by myxomatous disease, connective tissue disorders, aortic dissection, or infective endocarditis.
In broad terms, aortic stenosis is correctable by aortic valve replacement, while isolated aortic insufficiency is correctable by aortic valve replacement or repair.
One of the difficulties with valvular heart disease is determining when patients need an intervention. For example, in aortic stenosis, only half of the patients have symptoms on presentation, yet half of them may die within two years. Thus the indicated timing for intervention is crucial. The American College of Cardiology (ACC) and the American Heart Association (AHA) published consensus guidelines in 2014, updated in 2017, which serve as standard indications for the management of valvular heart disease (VHD), as summarized below.
Valvular heart disease is staged A through D, based on echocardiography parameters: A at-risk features (i.e., bicuspid valve) normal parameters asymptomatic, B moderate parameters asymptomatic, C1 severe parameters still asymptomatic preserved ejection fraction, C2 severe parameters still asymptomatic reduced ejection fraction, and D symptomatic +/- severe parameters. Severe parameters for stenosis include aortic valve area < 1.0 cm squared, mean pressure gradient ≥40 mmHg, aortic jet velocity ≥4.0 m/sec. Severe parameters for insufficiency include regurgitant jet width ≥65% of the left ventricular outflow tract, vena contracta >0.6 cm, holodiastolic aortic flow reversal, effective regurgitant orifice ≥0.3 cm squared, regurgitant fraction ≥ 50%, and regurgitant volume ≥60 mL/beat.
In general, patients with aortic valve stage A receive a screening echocardiogram, patients with stage B disease receive echocardiogram surveillance every 3-5 years, while aortic stage C patients receive echocardiogram surveillance every 0.5 to 1 year with possible intervention, and stage D patients receive an intervention. A Heart Valve Team should assess any patient considered for intervention in a multidisciplinary fashion.
Class I and Class II indications for intervention for aortic stenosis and aortic insufficiency are summarized below.
Aortic Stenosis Class I
- Stage D (severe parameters + symptomatic)
- Stage C or D (severe parameters regardless of symptoms) if already indicated for another heart surgery
- Stage C or D (severe parameters regardless of symptoms) if ejection fraction <50%
Aortic Stenosis Class II
- Stage C with very severe parameters (jet velocity >5.0 cm/sec)
- Stage C decreased exercise tolerance
- Stage D without severe parameters (low-flow/low-gradient), but low dose dobutamine stress produces high-flow/high-gradient parameters
- Stage D without severe parameters, but anatomic and clinical evidence suggests stenosis is the underlying cause of symptoms
- Stage B if already indicated for another heart surgery
- Stage C if the low-risk patient and jet velocity is increasing >0.3 cm/sec/year
Aortic Insufficiency Class I
- Stage D (severe parameters + symptomatic)
- Stage C or D (severe parameters regardless of symptoms) if already indicated for another heart surgery
- Stage C or D (severe parameters regardless of symptoms) if ejection fraction <50%
Aortic Insufficiency Class II
- Stage C with pronounced left ventricular dilation (>50 mm at end-diastole)
- Stage B if already indicated for another heart surgery
Once a patient meets indications for an intervention, the Heart Valve Team will assess the patient’s operative and interventional risk, using a combination of Society of Thoracic Surgeons (STS) risk calculators, frailty indices, and a patient-specific accounting of failed organs and comorbidities to determine whether SAVR or TAVR is appropriate.
For patients with Stage B or C disease indications above (i.e., asymptomatic patients), at this time, SAVR is the only indicated intervention. For patients with Stage D disease, either SAVR or TAVR may be an option depending on the risk above. For low risk (<4% operative mortality), SAVR is indicated. For intermediate (4-8% operative mortality), either SAVR may be offered as a class I recommendation or TAVR may be offered as a class II recommendation. For high risk (>8% operative mortality), SAVR and TAVR have equivalent class I recommendation status. For prohibitive surgical risk (>50% mortality or ≥3 organ systems already failed), TAVR carries the class I recommendation, and SAVR should not be offered.
Once SAVR is chosen, the team must choose between a valve replacement or repair. Again, in broad terms, stenosis is correctable by replacement, while isolated insufficiency is correctable by replacement or repair. Some groups have also shown success in repair for the bicuspid disease. Whether the repair is an option depends on the quality of the valve and the availability of particular surgical expertise.
Once a replacement is chosen, there are three options for replacement: mechanical valve, stentless pulmonary autograft/homograft (Ross procedure), or bioprosthetic valve. Each option varies in requirements for anticoagulation and long term durability. This variability should be made clear to the patient according to the ACC/AHA guidelines, and the decision process should be a shared one with the patient, whereas formerly, patient age guided the choice. Age still has some role in the decision process, in that mechanical valves are favored for patients under 50 years old, and bioprosthetic valves are favored for patients over 70 years old. The Ross procedure is preferred for young patients with a contraindication to anticoagulation; otherwise, all other patients with a contraindication to anticoagulation are recommended to have a bioprosthetic valve.
Other variables must be considered when selecting a replacement valve. Pre-existing renal disease, hyperparathyroidism, and young age all increase the risk of calcium deposition and bioprosthetic structural deterioration. Lack of access to regular international normalized ratio (INR) assays favors the use of bioprosthetic valves. Co-morbid conditions requiring anticoagulation, high risk for reoperation such as prior chest radiation or porcelain aorta, and a small aortic root size (if a future valve-in-valve procedure is needed) all favor a mechanical prosthesis.
Not to be omitted in a discussion of valvular disease, surgical treatment of infective endocarditis cannot be over-generalized; it must be rooted in the full scenario of the particular patient, i.e., how much should be debrided and when to perform the debridement. The ACC/AHA offers some basic guidelines.
A full culture-driven course of antibiotics should be the first-line treatment until one of the following scenarios pushes the multidisciplinary team toward early surgical source control: deteriorating heart failure, highly virulent or resistant organisms, new heart block, annular abscess, relapsing infection, persistent bacteremia for five to seven days, presence of a prosthetic valve infection, infected defibrillator, pacemaker, or leads, recurrent emboli, persistent vegetations, or mobile vegetation >1.0 cm.
Studies have shown harm from interventions on pregnant patients with stenosis or insufficiency, even with stage D disease, unless they have reached the point of severe, intractable heart failure. Interventions on women with stage D disease should be performed before pregnancy. Bioprosthetic valves have shown increased susceptibility to structural deterioration in pregnancy, so these are disfavored for women who may become pregnant. Warfarin carries a risk of teratogenicity in the first trimester, but there are multiple pharmacologic strategies to optimize maternal and fetal health if a woman has a mechanical valve.
As above, patients with high or prohibitively high surgical risk should not be offered a SAVR. Patients taken to surgery for infective endocarditis should be implanted with only a minimum amount of foreign material. Although valve replacement with a bioprosthetic valve is not strictly contraindicated, homograft options more resistant to infection may be preferred. Septic emboli causing a stroke is an exceptional circumstance; if possible, cardiac surgery should be postponed after ischemic stroke for two to three weeks and after hemorrhagic stroke for four weeks.
Some patients may have specific religious contraindications to bovine or porcine bioprosthetic valvular components.
As above, patients with valvular disease should be evaluated at a minimum with a history and physical, chest x-ray, and transthoracic echocardiogram (TTE) with both 2-D and doppler capabilities. Adjunctive modalities such as a transesophageal echocardiogram (TEE), computed tomography angiography, cardiac magnetic resonance imaging, and cardiac catheterization may be employed based on clinical suspicion. Discussion with the patients on their preference for valve takes place preoperatively.
The standard approach to perform aortic valve replacement is through a median sternotomy using cardiopulmonary bypass (CBP). With the advent of TAVR, minimally invasive SAVR approaches such as through a right anterior mini-thoracotomy are also being performed, albeit with hour longer CBP times. CBP arterial cannulation is most commonly through the ascending aorta, and the venous cannulation is through the right atrium; peripheral femoral cannulation has also been used. The left ventricle is typically vented through the right superior pulmonary vein vent. Cardioplegia is commonly given antegrade through the coronary ostia once they are exposed, with care taken to properly identify and perfuse the ostia, especially in cases of bicuspid valves where anomalies may be present, such as a high right coronary origin. Supplemental retrograde cardioplegia through the coronary sinus might be suitable for hypertrophied hearts with aortic stenosis or severe aortic insufficiency. Redosing of cardioplegia must be performed periodically.
After 300 IU/kg heparin has been administered, the ascending aorta cross clamped, CBP established, the patient cooled to 30°C, and cardioplegia is given, the specific steps of aortic valve replacement can begin. The entirety of the work is performed through an aortotomy at the level of the sinotubular junction or 1 cm above; the aortotomy may be transverse or oblique, extending down into the non-coronary sinus. Retraction sutures may be placed at each commissure and the distal ascending aorta to aid exposure, and excision of the native valves can begin. The excision of native valves usually begins with the right cusp and can proceed in either clockwise or counterclockwise direction. Valves are excised sharply with scissors or a blade, debriding the annulus of residual calcium deposits, which are carefully removed from the ventricle, avoiding that any should become trapped in the coronary ostia or embolize to the brain after the aortic cross-clamp has been removed. The mitral valve may now be inspected from the aorta as a matter of thorough surgical practice.
After having selected a valve (see below paragraph), the corresponding commercial valve sizer should be used to choose the valve of the best fit. The sizer at the simplest level is a disk attached to a probe, which is the same size as the desired valve. Since the sinotubular junction is typically where the aorta tapers, this area usually limits the introduction of sizer and valve. In cases where the basal ring/surgical aortic annulus is too narrow, an enlargement procedure may be used, as discussed below. When the appropriate sized valve is chosen, the valve is mounted on the valve introducer, introduced into the field, and attachment is begun. Although some surgeons have explored sutureless valve attachment, most valves are sutured into place using a 2-0 permanent braided suture such as polyester of alternating colors such as teal and white for better clarity identifying which suture goes where in a tight space.
Continuous, interrupted, or horizontal mattress technique may be used, with or without pledgets, to reinforce weak annuli. The suture is commonly passed through the annulus from the ventricle to the aorta side then through the sewing ring surrounding the valve. If pledgets are used, placing the pledgets below the annulus may allow the valve to sit on top, accommodating a larger valve. Suturing may begin at any cusp and proceed in either direction as long as it is systematic. Once all sutures are placed with equal spacing, crossing none, the valve is then held up straight with the introducer and valve and introducer are “parachuted” down by the surgeon and assistant pulling all of the sutures slowly tight. The introducer is removed, the sutures are tied, and the tails cut; the valve is confirmed to sit well with no gaps, which could become leaks, and the coronary ostia are approved patent. Retraction sutures are cut, the aortotomy is reapproximated and closed with running polypropylene suture in two layers. Trendelenburg positioning, CBP circuit manipulation, venting, and transesophageal echocardiogram (TEE) are all used to de-air the heart and safely remove the aortic cross-clamp. TEE is also used to confirm excellent valve function with no paravalvular leak. Chest tubes and pacing wires are placed, hemostasis assured, and the patient is weaned off CBP, heparin reversed, and decannulated in the usual fashion, and the chest is closed.
When selecting a suitable bioprosthetic valve, the surgeon considers multiple variables. The most obvious variable is size. Prostheses range from 19 mm to 31 mm; however, size does not correspond between manufacturers; a 21 mm in one brand may be a 23 mm in another. Using too small a valve for the patient may leave a residual pressure gradient, disallowing the full benefits of ventricular remodeling. The largest size that can be accommodated by the patient’s annulus is chosen, represented by the effective valve orifice area (EOA). When the EOA indexed to the patient’s body surface area (iEOA) is too small, this situation is called patient prosthesis mismatch (PPM). To avoid PPM, the iEOA for aortic valve replacements should be >0.85 cm/m. Severe PPM is defined as iEOA ≤ 0.65 cm/m. All this may be calculated out preoperatively based on imaging, then fine-tuned with interoperative measurements. More subtle are the differences in hemodynamic performance and durability of valve prostheses. It has already been noted that durability and hemodynamic performance is generally superior in mechanical compared to bioprosthetic valves, but this superiority has decreased with newer bioprosthetic models.
Supraannular bioprosthetic valves tend to have better hemodynamics than infrannular bioprosthetics. Stentless bioprosthetic valves lack a metal stent, so thus lack an intrinsic gradient and are considered to have better hemodynamic performance, but they also require more time to implant properly. Whether porcine or bovine pericardial, all bioprosthetics are subject to calcific deterioration. Early bioprosthetics were treated with high-pressure glutaraldehyde fixation to add stability and decrease xenograft antigenicity, but this was found to expose too much collagen and phospholipid for calcium-binding. Most modern treatments favor low or zero pressure fixation, and many of the models above employ additional proprietary anti-calcium chemical treatments to reduce the calcification problem. Some argue for an additional immunologic and atherosclerotic component to calcific bioprosthetic degeneration.
In a tight annulus which still is at risk of PPM, despite trialing different sizes from different manufacturers with better iEOA, then one of three aortic enlargement procedures may be performed. The Nicks-Nunez procedure enlarges the aorta posteriorly by incising the aorta between the left and non-coronary sinuses vertically up into the aortic root, then closing the incision with a patch. The Rittenhous-Manouguian procedure enlarges the aorta posteriorly by incising the non-coronary sinus down into the annulus and through the anterior leaflet of the mitral valve, followed by patching. The Konno-Rasten procedure enlarges the aorta anteriorly by incising between the right and left sinuses through the left ventricular wall; the aortotomy is closed with a patch, and the ventriculotomy is closed with two layers of the patch.
Postoperatively, valve replacement patients are recovered similar to other cardiac surgery patients. Patients with severe ventricular hypertrophy may have their cardiac compliance temporarily worsened coming off CPB so that they may need increased preload and central venous pressures maintained higher at 15 to 18 mmHg in the immediate postoperative period. Post-op valve patients also need to execute the anticoagulation plan made preoperatively. Although the advantage of a bioprosthetic valve is less need for anticoagulation, they should still be maintained on anticoagulation with an INR target of 2.5 for a minimum of three months while the valve is undergoing endothelialization. Anticoagulation should be extended to a goal of six months for further stroke risk reduction. Indefinite daily aspirin 75 to 100 mg is recommended. Valve surveillance echocardiography is recommended for bioprosthetic valves before discharge, at six to twelve months, at five years, and when clinical suspicion arises.
Given the variety of valves on the market, it was recognized early that a common language was needed to compare products and outcomes. The American Association of Thoracic Surgeons (AATS) and the Society of Thoracic Surgeons (STS) maintain an Ad Hoc Liaison Committee for Standardizing Definitions of Prosthetic Heart Valve Morbidity, which updated its latest definition in terms in 1996.
The guidelines recognize hospital mortality before a patient’s discharge as distinct from 30-day mortality, also known as operative mortality. Valve-related mortality is mortality due to one of the below morbid categories not related to progressive heart failure.
The recognized morbid categories are as follows. Structural valve deterioration (SVD) is stipulated as any change intrinsic to the valve leading to stenosis or insufficiency, including calcification, fracture, tear, and suture disruption but excludes prosthetic valve endocarditis (PVE) and thrombotic dysfunction, which comprise their own mutually exclusive categories. Nonstructural dysfunction is not intrinsic to the valve leading to stenosis or insufficiency, including obstruction from improper placement, a leak from improper sizing, and hemolytic anemia (an indicator of the leak). A bleeding event is an event that leads to hospitalization, transfusion, or death, but does not require taking anticoagulation.
An old study reporting ten years follow up of one particular mechanical valve now unavailable gives representative rates of these complications for the aortic valve position.
- Structural valve deterioration 0 events/patient years
- Nonstructural dysfunction 0.2 to 0.8 events/patient years
- Thrombotic events 0.0 to 0.2 events/patient years
- Embolic events 1.4 to 2.5 events/patient years
- Bleeding events 0.8 to 2.5 events/patient years
- Endocarditis events 0.4 to 0.7 events/patient years
- Reoperation 0.3 to 1.8 events/patient years
By comparison, representative statistics for bioprosthetic valves at twelve years in the aortic position include 87% freedom from valve-related mortality, 84% freedom from reoperation, 93% freedom from SVD explantation for patients over 60 years old, and 76% freedom from SVD explantation for patients under 60 years old. A 25-year study of another bioprosthetic valve, albeit in the mitral position, also gives representative outcomes: thromboembolism 0.5%/valve year, bleeding event 0.7%/valve year, endocarditis 0.4%/valve year, and SVD 2.3%/valve year. SVD is the most common cause of reoperation for bioprosthetic valves, especially after seven or eight years. Freedom from SVD at ten years is cited to be between 70% and 90%, while at fifteen years, it is cited to be 50% to 80%.
The clinical significance of surgical aortic valve replacement and bioprosthetic valves should be seen in the context of aortic valvular disease, which uncorrected, is dismal. Severe asymptomatic aortic stenosis carries a five-year survival rate quoted between 38 and 83%, and with the advent of symptoms, a patient faces a sudden cardiac death risk of 2% per month. The natural history for aortic insufficiency is less stark but still concerning, with five- and ten-year survival rates at 75% and 60%, respectively, for asymptomatic patients, while symptomatic patients face a mortality risk of 10% per year.
Juxtaposing this natural history against SAVR mortality rates, it is a life-saving surgery. Thirty-day mortality for isolated SAVR is quoted at 3.4%. Fifteen-year all-cause mortality from a trial in the late 1970s is quoted at 66% for mechanical valves, 79% for bioprosthetic valves. Given improvements in SAVR technology and after-care since the 1970s, modern long-term follow up is anticipated to be even better.
Survivorship is not the only clinically significant benefit. The left ventricle begins to remodel and regress its hypertrophy 18 months through 5 years after surgery, and 70 to 90% of patients with heart failure symptoms find symptomatic regression to New York Heart Association class I levels, remaining stable up through ten years postoperatively.
Enhancing Healthcare Team Outcomes
Complex health care interventions such as cardiac valve replacement demand excellence in nontechnical domains such as communication, human factors, teamwork, safety culture, and optimizing the operative environment. There is an increasing consensus backed by data that attention to these issues is essential to achieving high-quality interventions.