Introduction
Operations for the aortic root are a great demonstration of how techniques have evolved to tackle specific challenges posed by anatomy and pathology (see Figure 10.1). We are now at an enviable stage where there are multiple options for repair, replacement, and even reinforcement, of the aortic root and we can consider the benefits of long-term survival and re-intervention rates rather than focusing only on operative mortality.
Pathophysiology
The most frequent reason for aortic root surgery is aneurysmal dilatation (Catrovinci et al. 2015). Although other pathologies may affect the aortic root (such as a root abscess in infective endocarditis or involvement of the aortic root in an acute type A aortic dissection), these may be repaired using other techniques such as patch repairs or reconstruction, rather than root replacement. Occasionally an aortic root replacement may be necessary to facilitate an aortic valve replacement in an otherwise entirely calcified (or ‘porcelain’) aorta.
Aneurysmal disease remains the mainstay of aortic root surgery but what constitutes an aneurysm in the ascending aorta?
An aneurysm is defined as a localised dilatation of a vessel that includes all three layers of the wall (ACCF et al. 2010). Normal range values in women lie between 3.50 and 3.74cm, and 3.63–3.91cm in men, with an abnormal aortic root and ascending aorta dilatation considered to be present at a diameter above 40mm (European Society of Cardiology 2014). The danger from aneurysmal disease of the aorta comes from an increase in tension within the vessel wall as the aneurysm gets larger. This increased tension is described by Laplace’s rule (Levick 2010); as aneurysms increase in size, the likelihood of developing an acute aortic syndrome that could lead to rupture and death increases. Acute aortic syndromes include aortic dissection, intramural haematomas and penetrating aortic ulcers (ACCF et al. 2010).
When the ascending aorta or root reaches 55mm, there is a 15% chance of such adverse events occurring (Nardi & Ruvolo 2016). These studies have helped inform international guidelines and determine threshold diameters at which aortic root surgery should be considered and balanced against an estimated elective operative mortality of up to 5%. The aetiology of ascending aortic aneurysms essentially relates to structural weaknesses in the aortic wall. In a meta-analysis of patients who had undergone aortic root surgery, nearly 50% had either a diagnosed underlying connective tissue disease or a bicuspid aortic valve (Moorkhoek et al. 2016). Although good control of hypertension is important for general cardiovascular health, its contribution to ascending aortic dilatation appears to be minimal – with a greater contribution from age-related degeneration of elastic fibres in the media (Goldfinger et al. 2016).
Innate weaknesses within the aortic wall, secondary to connective tissue disease or bicuspid aortopathy, can cause excessively rapid aortic expansion and aneurysm formation with the potential for acute aortic syndromes at reduced diameters of expansion. Operative guidelines reflect this, with interventions recommended at lower thresholds. Of the patients with connective tissue disease who undergo ascending aortic replacement, the most frequent – in the region of 20–30% of the patient population – is Marfan syndrome (Moorkhoek et al. 2016). Marfan syndrome is a genetic condition with autosomal dominance which causes abnormalities in the production of glycoprotein Fibrillin-1 within the elastic lamina of the aortic wall (Martínez-Quintana et al. 2017). The term ‘cystic medial necrosis’ is used to describe this process, which manifests as aneurysmal dilatation in the aorta.
Loeys-Dietz syndrome is another connective tissue disorder causing weakness of arterial walls. In this condition, however, aneurysmal disease can be more widespread in the vascular tree, also appearing at an earlier age with more rapid expansion (Loughborough et al. 2018). Given the aggressive nature of this disease, operative thresholds tend to be at the lowest aortic diameter in the guidelines (see Table 10.1). Patients with bicuspid aortic valves and dilated aortic roots are classed as having bicuspid aortopathy. This is a heterogeneous group of patients who present with aneurysmal disease on average in their fifth decade, approximately 15 years earlier than patients with tricuspid valves and no history of connective tissue disease (Losenno, Goodman & Chu 2012). Either calcified stenotic valves or severe aortic regurgitation may be associated with the dilated aortic root. Cause and effect relationships between the root and valve pathology remain uncertain and, although the condition does appear to have a genetic element, specific inheritance patterns are still being defined.
Inflammatory processes that affect blood vessels, such as Takayasu and giant cell arteritis, can also cause degeneration of this elastic layer and subsequent aneurysmal change. Such patients may present as part of an acute inflammatory illness which can complicate surgical management with the need for immunosuppressive regimes (ACCF et al. 2010).
Table 10.1: Guidelines for operative intervention on the aortic root
(adapted from ECS 2014 and AHA 2010)
European Cardiac Society (ECS) 2014 | American Heart Association (AHA) Task Force (ACCF 2010) | |
Balloon aortic valvuloplasty (BAV) | Aortic root >5.5cm or Aortic root >5cm if Growth rate >3mm/yr | Root diameter >50mm or Growth rate >5mm/year |
Connective tissue disease | >50mm or >45mm and growth rate >3mm/yr | Marfan >50mm or Growth rate >5mm/year Loeys-Dietz >4.2cm (TOE) >4.6cm (CT) |
Non-connective tissue disease | >55mm | >55mm or growth rate >5mm/year |
Signs and symptoms of aortic root disease
• Presentation of aortic root disease will depend on the underlying pathology. Presentations may relate to the aorta and/or the aortic valve or patients may be completely asymptomatic.
• Symptoms secondary to the aorta may be chest pain. This is a very concerning feature as it may herald acute dilatation or even onset of an acute aortic syndrome.
• Aortic valve disease may exist concurrently with aortic root aneurysms. This may be a stenotic or regurgitate lesion. Patients may present with an incidental finding of a murmur; or, at more advanced stages, there may be evidence of heart failure with breathlessness and exercise intolerance. Patients with aortic stenosis may present with classic symptoms of angina and syncope.
• It is important to ask patients about any family history related to aortic disease, or sudden death, to establish whether there may be a genetic component that might contribute to complications at an earlier stage in their disease process.
Investigations
The three main modalities used to image the aortic root are echocardiography (ECHO), magnetic resonance imaging (MRI) and computed tomography (CT). Each one has particular benefits and may be used to glean different information. ECHO and MRI have the advantage of not using ionising radiation and can potentially provide functional information about valvular and ventricular function. ECHO is far quicker and can be done in a clinic, but operator variability may be greater. CT scans are faster and more readily available than MRI scans and give good reproducible resolution of the aortic wall, but the use of ionising radiation is a disadvantage for those requiring multiple scans throughout their lives. Aortic diameters are measured from the external diameter in both CT and MRI. However, in ECHO, the internal diameter is taken. In all cases the measurement must be taken perpendicular to the direction of flow (ACCF 2010).
Many patients will be under surveillance for many years prior to aneurysmal disease reaching a stage where intervention is necessary. In such cases regular monitoring with interval scans is used. The timeframe of these scans is dictated by both the size and rate of expansion of the aortic root as well as the rate of deterioration of any aortic valve disease. In such cases it’s best to employ a consistent method of scanning to ensure that any changes in aortic diameter are not related to varying imaging techniques. The AHA guidelines provide a model surveillance schedule with which to monitor patients with known aortic dilatation (ACCF et al. 2010).
Medical management
In patients whose aorta shows enlargement but is below operative thresholds, management is aimed at modifying risk factors to prevent further expansion. Controlling hypertension is important to reduce the risk of death from all cardiovascular disease. However, beta-blockers have been associated with a reduced rate of expansion in root aneurysm size, particularly in Marfan patients, making them a preferential choice in management of blood pressure in these patients (Chun, Elefteriades & Mukherjee 2013).
There has also been interest in the use of angiotensin 2 receptor blockers, such as losartan and valsartan, to reduce aneurysm expansion rates, particularly in patients with Marfan and Loeys-Dietz syndromes. Angiotensin 2 blockers appear to interfere with, and slow down, intracellular signalling pathways that contribute to degradation of the elastic fibres in the aortic media (Dietz 2010; Takeda et al. 2016). So far, studies have failed to show treatment translating into any difference in times to surgical intervention or death. However, research is ongoing in this area (Gao et al. 2016).
Anatomical considerations in aortic root surgery
We have considered why we might need to perform aortic root surgery; now we will address the different techniques and when they are suitable. The aortic root extends from the annulus of the aortic valve up to the sinotubular junction and includes the aortic valve leaflets, the coronary ostia and the sinuses of Valsalva. The patency of the aortic valve within the aortic root is governed by the interaction of the aortic annulus and valve cusp morphology. If the aortic cusps remain unaffected by disease, it may be possible to perform an aortic root repair rather than replacement by re-establishing annulus geometry.
Annulus
When we think of the aortic valve annulus, we refer to the hinge points of the aortic valve leaflets themselves. However, when observing the leaflet insertion points, we can see that in fact they are not oriented in a circular annular arrangement but suspended in a crown-like position within a cylinder, with the sinotubular junction at the top and the nadirs of the cusps at the base (Anderson 2000). If either end of the cylindrical aortic root dilates, leaflet coaptation is compromised, ultimately leading to valvular insufficiency, which is usually seen on echocardiography as a central jet.
Cusps
Although annular changes can be solely responsible for aortic valve failure, abnormal geometry of the cusps can also be responsible. The ratio of the free margin of the aortic valve leaflet to the length of its cusp insertion determines how mobile a cusp is and its propensity to prolapse. A proportionally greater length of cusp insertion to a small free margin can be seen in bicuspid valve disease. This leads to a restrictive pattern of leaflet motion, with the cusps in effect being held closer to the sinuses (El Khoury & de Kerchove 2013). Conversely, in connective tissue disease, excessive leaflet tissue may cause the overly mobile cusp to prolapse into the ventricle. These types of pathology are more likely to create an eccentric pattern of aortic regurgitation (in contrast to annular disease), as the asymmetry of the valve leaflets will direct a regurgitant jet in a particular direction.
Coronary ostia
As part of the planning for any aortic root procedure, it is essential to identify the position of the coronary ostia and any disease affecting them (such as calcification or extensive aneurysm). Coronary arteries that are diseased may warrant concurrent bypass grafts. Coronary ostia that are heavily calcified or even fixed by aneurysmal disease, far from their normal anatomical position, may need additional Dacron extensions attached to the coronary ostia to reach the repaired root. These are known as ‘Cabrol’ grafts (Cabrol et al. 1981).
Identification of the heights of the coronary ostia on CT or ECHO is also important, particularly in patients with bicuspid aortic valves. In these patients, the height and position of coronary ostia may vary from the usual pattern, becoming more distal on the aorta or closer to the commissures. Foreknowledge of this may prevent the surgeon inadvertently making aortic incisions very close to ostia that may compromise coronary button formation at a later stage.
Replacement or repair?
Whichever operative strategy is used, the aim should be an enduring and satisfactory long-term outcome with low associated morbidity and mortality. As with many other aspects of cardiac valvular disease, correction of aortic root pathology can include either aortic root replacement or repair. Both strategies involve replacing dilated coronary sinuses with Dacron grafts and reimplanting coronary ostia as ‘buttons’. Where the techniques differ is in management of the aortic valve and basal annulus.
Valve-sparing root repairs may use techniques to repair or refashion native valve cusps, in addition to supporting and reinforcing the annulus complex. Replacement, on the other hand, involves complete excision of the aortic valve, along with the coronary sinuses, followed by replacement with either a bio prosthetic or mechanical prosthesis embedded in a Dacron tube graft (see Figure 10.2) (Lansac et al. 2013).
Although the average age of patients presenting for aortic root surgery is the sixth decade, there is also a large proportion of younger adults with connective tissue disease or congenitally bicuspid aortic valves. The advantage of an aortic root repair over replacement is particularly relevant in these cases, as preservation of their own valve may eliminate the need for antithrombotic medication and its side effects. This is especially important in women who are planning families and hoping to avoid potentially teratogenic coumarins. With these younger patients, the long-term durability of repairs and likelihood of re-intervention is extremely important. Any repair strategies must be able to stand up to the high standards offered by root replacement with a mechanical valve conduit.
Importantly, any patient undergoing a valve-sparing aortic root repair must be consented for a replacement; and valve options should be discussed in the event of failure to produce a satisfactory repair. Ultimately the best operative strategy will depend on the underlying pathology of the aortic root, combined with what is the most acceptable long-term outcome for the patient.
Aortic root replacement
The modified Bentall procedure has been described as the ‘gold standard’ for aortic root pathology. This is secondary to the high standards it has set in short- and long-term outcomes, with which valve- sparing operations must now compete. The modified Bentall procedure has proven itself reproducible, compared with the technical challenges posed by repair strategies, with some series showing 30-day mortality rates of less than 2% in elective cases (Girardi 2008). Recent metaanalysis has shown freedom from reintervention in Bentall procedures with mechanical valves of 99%, and rates of prosthetic valve endocarditis that are less than 1% (Pantaleo et al. 2017). The initial operation described by Hugh Bentall and Anthony de Bono was an adaptation of a planned ascending aorta repair, when they encountered aortic tissue at the sinotubular junction that was too fragile for a direct graft anastomosis (Bentall & De Bono 1968).
The Bentall procedure itself involved opening the dilated ascending aorta, resecting the valve and implanting the new valve and conduit within the existing aortic tissue. Two holes were then made in the conduit wall at the level of the coronary ostia and the aorta around the coronary ostia was sutured directly to the conduit. The aneurysmal aorta was closed around the outside of the prosthesis, to control haemorrhage from the porous graft material; this became known as the ‘inclusion’ technique.
The main complications arising from this technique revolved around haemostasis, with pseudoaneurysm formation occurring within the aortic wrap and around the coronary ostia because anastomosis sites were difficult to visualise. To combat this issue, the technique was ‘modified’. Preservation of the existing aortic aneurysm was abandoned. Instead, the coronary arteries were excised from the aorta with some surrounding aortic tissue as a ‘button’ that could then be directly anastomosed onto the aorta. The aortic tissue was also excised more closely at the root and various techniques have now been developed to maintain a haemostatic seal at the annulus-valve-conduit join (Kouchoukos, Karp & Lell 1977). In addition, advances in the engineering of the graft material, with gel-impregnated Dacron and the addition of haemostatic sealants, have helped eliminate many previous problems with haemostasis.
There are many ways of carrying out aortic root replacement. The following method is routinely used in our department.
Operative technique for aortic root replacement
Patient preparation
Monitoring:
• Central venous access
• Arterial line monitoring for blood pressure
• Transoesophageal echocardiography: check the left ventricular function pre- and postoperatively; review prosthetic valve function to ensure no there are no postoperative paravalvular leaks.
Draping:
• Ensure exposure of chest, groins and both legs to the ankle; if the implantation of coronary button fails, a saphenous vein graft might be necessary.
Initiation of cardiopulmonary bypass
Cannulation:
• Most often cannulation will be central, with direct cannulation to the aorta and venous drainage from the right atrium.
Cardioplegia:
• Antegrade delivery (alone or in combination with retrograde) may be used, depending on surgeon preference and the presence of aortic regurgitation.
Temperature:
• A temperature on cardiopulmonary bypass of 34°C is maintained (please note that this varies in every institution and it depends on the operating surgeon and patient condition).
Vent:
• A right superior pulmonary vein vent is used to decompress the left ventricle.
Aortic cross-clamp:
• The clamp is applied high on the ascending aorta to provide space for the distal anastomosis to be sutured.
First surgical steps
Aortic transection:
• The aorta is transected above the sinotubular junction to be sure of any high coronary artery positions.
Excision of valve and preparation of the root and coronary buttons:
• Stay sutures (5/0 prolene) are placed in the tissue above the coronary buttons to minimise direct handling
• The coronary ostia are dissected away from the aortic root tissue and the aortic tissue is cut, with a 5mm cuff left around the hinge points of the aortic valve
• In non-connective tissue disease, the coronary buttons are formed, leaving a rim around the ostia for ease of suture placement. However, in connective tissue disease, this rim is minimised.
Valve and graft preparation and insertion/position
If a mechanical aortic valve is to be used, there are composite options in which the valve is already embedded within a tube graft (see Figure 10.3). However, this is not the case for tissue valves, and the tissue valve needs to be positioned inside the tube graft (see Figure 10.4).