Late Follow-up of Surgical MV Repair: Insights for Timing, Innovation

Some of the longest follow-up data yet published suggest that reoperation is rare among patients who underwent surgical repair of a degenerative mitral valve (MV) more than two decades ago. However, while residual mitral regurgitation (MR) was minimal immediately following the initial surgery, rates of recurrent MR increased with time.

Incidence of tricuspid regurgitation and atrial fibrillation also tended to creep higher in surviving patients, all of whom were initially treated by renowned Canadian surgeon Tirone E David, MD (University of Toronto, Canada), and his residents since the 1980s.

Notably, the high rate of reintervention-free survival—more than 60% in these patients—has implications for newer techniques, both transcatheter and less invasive surgical procedures.

“This should be the benchmark,” David stressed in an interview with TCTMD. “This is what traditional surgery can do. If you have a patient that has MV prolapse, and you want to do a percutaneous or surgical repair, this should be the benchmark you use if you're going to fix [it].”

That message applies not only for transcatheter therapies already approved or in development, but also for surgical innovations, he continued. “More and more surgeons are moving to robotic mitral valve repair, and yet, no one has published long-term outcomes. Yes, the patient likes it because there is no incision in the chest, just a bunch of holes, but is it better for the patient? Nobody knows.”

David and colleagues report their findings this week in the Journal of the American College of Cardiology.

A Long-term Series

There are few long-term, randomized controlled trials in valve surgery. Instead, patient series like this one, often from single centers or single surgeons, offer the best available evidence in support of specific techniques or approaches. The majority, though, have only followed patients for a decade at most, making this particular series—34 years in total—relatively unique.

This should be the benchmark. This is what traditional surgery can do. Tirone David

As David and colleagues describe in their paper, MV repair was performed on a total of 1,234 consecutive patients between 1981 and 2010, all of whom were diagnosed with MR due to MV prolapse. Almost all patients received a mitral annuloplasty ring, nearly 70% had chordal replacement, and leaflet resection was used if leaflet height exceeded 20 mm.

Patients were then followed for a median of 13 years, with follow-up on 163 patients passing the 20-year mark.

After two decades, the reoperation-free survival was 60.4%, and the cumulative incidence of cardiac and valve-related deaths was 12%, with just 48 patients (4.6%) needing reintervention on their MV, of whom 32 had developed MR. Just over 1% of patients were diagnosed with endocarditis over the follow-up period, 10.3% developed thromboembolism, and 6.4% had bleeding events.

Addressing the recurrence of MR was a key reason for the study, which comes on the heels of reports linking MR recurrence to worse outcomes, the authors noted. In this series, 12.5% of patients developed moderate or severe MR over the follow-up period and MR was more common in patients who were older, had complete heart block, had undergone MV repair without an annuloplasty ring procedure, or who had myxomatous degeneration. The investigators also found that persistent or new instances of moderate or severe tricuspid regurgitation were not uncommon, occurring in one in five patients, while new-onset atrial fibrillation was diagnosed in one-third of patients. The development of either condition appeared to be unrelated to the presence of recurrent MR.

To TCTMD, David characterized the new-onset A-fib as “mysterious”—or in the words of the authors “puzzling and unexplained,” particularly since they had closed the left atrial appendage during surgery in the “majority” of patients.

“It is what it is,” said David, adding that he hopes readers and researchers will look into the etiology of this phenomenon further.

The tricuspid regurgitation is also “puzzling,” he acknowledged, noting that tricuspid disease is a key focus of the accompanying editorial by David H. Adams, MD, and Anelechi C. Anyanwu, MD (both from Mount Sinai Hospital, New York, New York), who call the high rate of tricuspid regurgitation “concerning.”

A more aggressive approach, “preemptively” treating mild tricuspid regurgitation or a dilated tricuspid annulus at the time of the MV repair, might have reduced later morbidity and mortality in David’s series, the editorialists suggest.

In response, David agreed with the proposal in theory, but pointed out that it remains “speculation.”

“There's no data to date showing that if you repair the tricuspid valve, you prevent this problem in the future,” he said. “We're going to have to wait another 15 years to have that answer.”

For David, it’s the durability of the MV repair itself that is “remarkable” in his series, and he emphasized that this has implications for percutaneous therapies.

“Survival is not perfect, but 60% of people 20 years later are alive and event-free, their valve is still working perfectly well,” he said. “The probability of needing a second operation because their valve failed was only 6% at 25 years,” he added, citing numbers not included in the paper, and was less than 5% at 20 years. “That's very, very low. There's no way that a clip can do that. In other words, if you are a 30-, 40-, 50-year-old, it's ludicrous to have a clip based on this data. An operation would be much better.”

To this point, he stressed that even with all these years of experience he still often performs a second and even third pump run to make sure that there is no residual MR following his MV repairs. “Imperfect repairs must be addressed intra-operatively if we are to attain a higher standard in late outcomes,” the authors note.

Also critical, said David, is the timing of the intervention. Among patients who are in NYHA class I at the time of their procedure, mortality at 20 years was less than 15%. That contrasts starkly with patients in NYHA class IV at the time of their MV repair: 30% died within the first 10 years, and more than half were dead by the time two decades had passed. 

Some of the Best Data to Date

Commenting on the findings for TCTMD, James Gammie, MD (University of Maryland School of Medicine, Baltimore), noted that David is one of the original pioneers of the use of Gore-Tex cords for chordal repair, and as such these results represent the work of a master operator. Given the “astonishing” lack of randomized controlled trials in surgery, said Gammie, this series represents some of the “best data that we have in terms of how good surgical mitral valve repair can be, in the hands of a high-volume, expert mitral valve surgeon.”

Gammie continued: “If a cardiologist sends their patient to Tirone David for a mitral valve repair, and that patient has degenerative disease, he has an almost certain likelihood of getting that patient back with an intact, functioning, [and] high-quality mitral valve repair that is very likely to last for 20 years.”

That said, Gammie pointed out, even with this level of expertise, mitral valve repair is not perfect; there is “a continuous attrition” in terms of MR, the reoperation rate is not zero, and patients do die of cardiac causes. “But this is the best estimate that we have of what is possible,” he said.

And yes, that should be food for thought for people innovating in this space, Gammie agreed. While a lot of attention in recent years has been on the MitraClip (Abbott)—first for degenerative and now for functional MR—David’s repairs did not use the Alfieri technique that the MitraClip emulates and in almost all cases involved an annuloplasty ring.

“For those developing catheter-based approaches, I think that a chordal replacement approach in addition to somehow creating an annuloplasty effect would achieve the goal of approaching the quality of result that can be done with the atrium open,” Gammie said.

Like David, Gammie also stressed the importance of earlier intervention in these patients, pointing out that survival free from intervention was far better among patients who underwent repair before they developed significant symptoms, and before their left ventricular function had degenerated.

“This series really supports the notion of early referral for repair of severe degenerative MR . . . if you want your patient to have a great long-term result,” said Gammie.

Adams and Anyanwu agree. “David et al are to be applauded for following their patients so diligently and giving us the best glimpse to date as to how patients do in the second decade after degenerative mitral valve repair,” they write. “These data should form the benchmark for future guidelines and studies on the role of degenerative mitral valve intervention, particularly in young and asymptomatic patients.”