In the past 15 years valve-preserving aortic replacement has evolved into an increasingly accepted alternative to composite replacement of aorta and valve. Preservation of the native valve has the obvious advantage of obviating the need for anticoagulation, and there is increasing evidence that it minimizes the overall incidence of valve-related complications.
Two basically different principles of valve-preserving aortic replacement are currently used, and minor modifications have been proposed for both. Reimplantation of the native valve within a vascular graft has been shown to normalize aortic root dimensions and restore valve function . Remodeling of the root was originally designed by Sarsam and Yacoub , and it has been demonstrated to restore root geometry and improve aortic valve competence. We have employed this technique for more than 10 years and have been very satisfied with the clinical results .
Compared to the reimplantation procedure, in our hands remodelling has had the advantages of being less complex and time consuming. Nevertheless, it has to be kept in mind that any reconstructive operation on the aortic root may alter aortic valve geometry and function. Thus, it is probably less important which of the two techniques are used, but very relevant how valve-preserving aortic replacement is performed in order to reach the goal, i.e. elimination of aortic pathology and normal (or near normal) function of the aortic valve.
Proximal aortic replacement with remodeling of the aortic root will most frequently be used in ascending aortic or root aneurysm, irrespective of valve function, to eliminate the risk of aortic complications. Standard aortic size criteria will be applicable. Alternatively, root remodeling may be applied in aortic regurgitation as predominant pathology and primary indication for surgery if root and particularly sinus dimensions are clearly abnormal (Video 1). In this situation root remodeling helps to restore normal geometric relationships between root and aortic cusps and stabilizes the root, similar to the role of a prosthetic ring in mitral repair. The valve repair procedure will also have to address any cusp prolapse, which in our experience is frequently present [4, 5]. It is also important to realize that any reduction of root dimensions will automatically alter the complex relationship between cusp and root configuration and may lead to cusp prolapse. Regardless of the choice of operative strategy it is of paramount importance to produce cusp configurations that are as close to normal as possible, and we have developed a particular strategy to achieve this goal.
Remodeling was originally designed for tricuspid aortic valves with a symmetric root. We have subsequently adapted the principle also for the anatomy of bicuspid aortic valves. Remodeling with replacement of only one or two sinuses has been used by some surgeons in the setting of acute dissection or asymmetric dilatation of one or two sinuses.
Valve preserving surgery does not appear reasonable or appropriate in patients who have significant calcification of the aortic cusps or multiple fenestrations that can be expected to diminish long-term cusp stability. Root remodeling is not justified for ascending aortic aneurysm which spares the root and only starts at the level of the sino-tubular junction. The role of valve sparing surgery in the elderly patient may be controversial. A 75 year-old individual with a dilated aortic root can well be treated with a composite root replacement as the easier and quicker operation. On the other hand, preservation of the aortic valve has been shown to be associated with a low incidence of endocarditis, and this ultimately may also be an argument for preservation of the native aortic valve in the elderly patient .
There has been increasing evidence in the last years that remodeling may not be a good operation for every patient with a dilated aortic root. Patients with connective tissue disease commonly have a large so called annulus, better termed aorto-ventricular junction. Root remodeling will not be able to reduce the size of the aorto-ventricular junction, and we limit its use to a diameter of less than 30 mm at aorto-ventricular level, choosing valve reimplantation for individuals with bigger sizes.
Much of the necessary information can be gathered from the preoperative aortogram or transthoracic echocardiogram. A transoesophageal echocardiogram will help to determine root dimensions and cusp configuration more exactly. We always use it to determine the diameter of the aorto-ventricular junction and look at cusp configuration and eccentricity of the regurgitant jet as indicators of cusp prolapse.
While a transverse or limited sternotomy are access options, we still prefer a standard median sternotomy. Aortic and right atrial canulations are used for connection to cardiopulmonary bypass.
After the proximal aortic arch and the brachiocephalic trunk are mobilized, it is most often possible to place a clamp just beneath the trunk. If there is need for complete replacement of the ascending aorta, a separate graft is anastomosed to the arch later in the operation. This can easily be achieved during a brief period of hypothermic circulatory arrest. The aorta is opened by a longitudinal incision, and cardioplegia is given directly into the coronary ostia.
For optimal exposure, the aorta is completely transsected 5 to 10 mm above the commissures. Stay sutures are placed in order to facilitate mobilisation of the root. We routinely measure the size of the aorto-ventricular junction using standard sizers (Video 2). If aorto-ventricular junction is 29 mm or less, we proceed with the original plan, i.e. root remodeling. Mobilisation of the aortic root down to the lowest point surface sinuses an important step of the operation. This may be done with scissors or cautery, depending on personal preference (Video 3).
We have a pragmatic approach in choosing graft size and take a graft approximately 1 to 2 mm smaller than the diameter of the aorto-ventricular junction. While some asymmetry has been described also for the tricuspid aortic root, we primarily try to create three symmetric tongues. Short incisions are made on one end of the graft, and by simply compressing the graft in different directions we ascertain symmetry (Video 4).
The height of the native commissures is eye-balled, and the respective incision in the graft is kept approximately 20 % shorter than the estimated height of the commissure. It is always easier to simply extend the length of this incision that has to accommodate the commissure rather than to work around an incision that is simply too long. The three tongues are now cut in a configuration that resembles a sinus curve (Video 5). Wide tongues will let the remodelled root bulge more, but may also contribute to secondary dilatation of the root in the long-term course.
The graft is then sutured to the aortic root, carefully following the insertion line of the aortic cusps (Video 6). We always start in the center of the sinus and work our way up to the commissures. By doing so, it is easy to accommodate the exact height of the commissure in the graft. We always start with the left sinus, go on to the right sinus and suture then noncoronary sinus last (Video 7).
Assessment of the aortic valve, i.e. configuration of the cusps is the most important part of the procedure once the graft has been sutured in place. The reduction of sinus dimensions will lead to some degree of cusp prolapse. In addition, prolapse may be pre-existent or aggravated by the operation. Initially we felt that only asymmetric cusp prolapse had to be corrected. We have now taken on the approach of trying to normalize the configuration of all cusps. In order to have an easy and reproducible indicator of prolapse or adequate cusp configuration we have chosen to measure the effective height of each cusp, i.e. the height difference between free margin and aortic insertion .
An effective height of 8 mm measured with a calliper intraoperatively corresponds to a similar height on postoperative echocardiograms and will result in an almost normal configuration of the aortic cusps (Video 8). In order to achieve this height the free margin is shortened by plication sutures using 5/0 or 6/0 prolene.
Once cusp configuration has been adapted, the aortic valve is checked for any residual asymmetric prolapse. This is corrected if found (Video 9).
Reimplantation of the coronaries is done in a standard fashion (Video 10). We always keep the openings in the graft relatively small, that is approximately the size of the coronary ostia. Finally, the aortic graft is trimmed to adequate length and anastomosed to the ascending aorta. The heart is deaired, coronary circulation resumed. While the degree of LV-distension and pulsatility of arterial blood pressure tracing on full bypass provide some thoracic aortic valve function, it is of paramount importance to check every repaired aortic valve by transoesophageal echocardiography. In this instance there is almost normal configuration of the aortic valve with trivial aortic regurgitation (Video 11).
With experience root remodeling can be used as part of an aortic valve procedure in the setting of root dilatation and almost any non-calcified aortic valve. It is important, however, to inspect the cusps carefully. Fenestrations in 2 or 3 cusps may make this valve not an ideal substitute for repair with good long-term results. Stabilisation of more than just 1 or 2 fenestrations with pericardial patches may require a long myocardial ischemic time and distort cusp geometry.
As always in surgery, exposure makes all the difference between an easy or a difficult operation. For this purpose, stay sutures are important. Their direction and their tension for individual steps of the operation are varied in order to provide optimal exposure.
It is of utmost importance to remember that any change in sinus dimensions will automatically alter the configuration of the aortic cusps. In addition, some degree of cusp prolapse is often pre-existing once the patient comes with more than grade II aortic regurgitation . For these two reasons never trust that root remodeling alone will result in a competent aortic valve with a normal configuration. Always check the configuration of the aortic valve after completion of remodeling. In case of doubt transsect the Dacron graft 1 cm above the height of the commissures in order to have better exposure. Subsequent graft to graft sutures are easy to do.
Based on current evidence root remodeling is not the ideal operation for patients with dilated aorto-ventricular junction or Marfan’s syndrome. These situations are much better addressed by valve reimplantation within a vascular graft.
Publication Date: 27-Sep-2006
Last Modified: 27-Sep-2006