• Cardiac Surgery

Dilatation of the aortic root leading to aortic dissection and rupture is the leading cause of premature death in patients with Marfan syndrome, which on average may shorten life expectancy by a third.1 ,2 Remarkably, over the last 40 years, life expectancy in Marfan syndrome has been restored to near normal through earlier recognition of affected patients, antihypertensive medical therapy and prophylactic aortic replacement. Prophylactic aortic root surgery in many surgical series from expert centres has carried a 1% risk or less of operative mortality.3 ,4 Furthermore, in the current era, valve preservation is usually possible, so most patients avoid valve prostheses and anticoagulation, although at the price of a median sternotomy and a short period of cardiopulmonary bypass, which are well tolerated in this young population.5 ,6

With such a high benchmark of surgical success, what is to be achieved by revisiting an approach to aneurysmal disease that has been abandoned by most cardiothoracic surgeons and aortic specialists?

Treasure et al7 are correct to point out that while conventional total root replacement, known as the Bentall procedure, is an established, safe and reproducible procedure, it leaves the patient with a valve prosthesis and a nearly 1%/year risk of valve-related morbidity, as well as the restrictions imposed by lifetime anticoagulation when a mechanical valve is used. Valve sparing aortic root replacement has a similar operative safety, but a longer surgical learning curve and as yet an unmeasured lifetime risk of valve failure, which would necessitate late valve replacement by conventional reoperative surgery or perhaps by percutaneous valve technologies. PEARS might be a simpler, safer approach, but much remains to be determined.

The less invasive nature of the PEARS procedure is undeniable. But does it get the job done? To the point of preventing rupture and dissection, I suspect the answer is yes. Though the history of aortic ‘wrapping’ is checkered and reports of its results have not been systematic, wrapping was often performed in desperate situations for atherosclerotic aneurysms, a very different scenario from the young Marfan patient with a connective tissue disorder. The grafts were not customised and their physical properties distinct from PEARS. It is encouraging that there were no dissections or ruptures in Treasure's series, but his patients had smaller aortas with low annual risks of dissection, so the length of follow-up may not be sufficient to support a conclusion of equal efficacy. To the point of preserving the valve, earlier intervention on smaller roots means there is less valve distortion that needs to be corrected, but conversely earlier intervention means greater risk of intervening on patients who ultimately might not have had an aortic complication.

There are two technical points to be made about the PEARS approach. One lesson of valve sparing surgery in Marfan syndrome has been the need to stabilise the aortic annulus to preserve valve competence. Any external prosthesis would therefore need to completely enclose the entire crown-shaped aortic annulus. The surgical dissection to allow seating of the prostheses at the nadir of the aortic root complex can be extensive, even in the cardioplegia-arrested heart on cardiopulmonary bypass, much less through a small upper sternotomy on the beating heart with thinned bulging sinuses. PEARS surgeons apparently have managed to accomplish this, but reproducibility remains a concern, and the one mortality acknowledged in the report was likely the result of poor surgical exposure of a vital structure. Another technical concern is the incisions in the graft to accommodate the coronary arteries, which if too small may narrow the coronary but if too large may lead to coronary button aneurysms. Getting that just right will have its own learning curve.

Another group at the University of Florida8 has pioneered a similar approach known as the ‘Florida sleeve’. It entails use of cardiopulmonary bypass, arrest of the heart and division of the aorta above the root. A conventional vascular prosthesis is placed around the root without resection of any tissue; keyholes are made to accommodate egress of the coronary arteries. Excellent results have been achieved, in support of the concept presented here, though only a minority have had Marfan syndrome.

If PEARS proves safe, reproducible and durable, it will still not replace all conventional root surgeries. Acute aortic dissection will continue to be the initial presentation of a significant number of Marfan syndrome patients, and replacement rather than reinforcement of the aortic wall will be necessary. Similarly, interventions on the aortic valve leaflets will be needed in some patients because of thinning, prolapse, tears and massive dilatation of the annulus. In the future, PEARS may find a niche in treatment of small-to-medium-size root aneurysms with competent, undistorted aortic valves, but only after it stands the test of time, backed with confirmatory experience from other centres and longer follow-up.