Historical Valvular Database Guides Treatment Innovations

With roots back in the 1960s, the Duke Valve Surgery Database has evolved into a national resource providing metrics for more than 55 studies, guiding significant treatment innovations.

Continuing the early tradition of electronic recordkeeping started by the late Duke Department of Medicine Chairman Eugene A. Stead Jr, MD, the valvular surgery database has been prospectively managed since 1990. Following 14,000 patients who underwent heart valve surgery at Duke, the compendium is one of the largest single-institution databases in the world.

“This database incorporates outcomes from every patient who has received a valve or valve implant since 1963,” says Donald D. Glower, MD, a Duke surgeon who has shepherded the database for decades and continues to advocate for its funding and institutional support.

Duke’s long history of cardiac innovation is highlighted by the resource. The first valve implant at Duke was performed in 1963—1 year after the procedure originated.

“Because there are so few in a country of this size with the extensive historical record of this database, it’s a tremendous resource,” Glower says. “It helps us monitor our patients and their well-being on a long-term basis, of course, but it provides an invaluable historical perspective.”

Valve technology evolves as procedures improve, but the database continues to yield insights into outcomes. “As we monitor new technology changes, this database helps us see clearly that some devices perform better in certain patients than others,” Glower says. “Ongoing clinical trials don’t always demonstrate that.”

The resource has provided clinical sources for 56 studies published in major journals, some of the findings of which are shown in the Table.

Table. Major Findings From the Duke Valve Surgery Database

Patients receiving biological valves do not require postoperative anticoagulants/antiplatelet drugs unless they present with AF. This finding resulted in guideline changes by the AHA and ACC.
Patients receiving dialysis for kidney failure live longer with biological valves rather than mechanical valves. This finding led to guideline changes by the AHA and ACC.
Patients with mitral regurgitation due to coronary disease do better with mitral repair than mitral replacement. This finding proved contrary to popular interpretations of the results from a randomized trial by the NIH.
The third largest series of tricuspid valve replacements reported good results with both minimally invasive tricuspid replacement and repair.
Smaller valves last longer than large valves, according to results from an analysis of the world’s 5 largest series of Carpentier-Edwards porcine valves.
The world’s largest series of patients receiving minithoracotomy mitral valves reported better safety than what is generally achieved with standard sternotomy.
The world’s largest single-institution series of patients receiving minithoracotomy aortic valves reported better safety than what is generally achieved with standard sternotomy.
The world’s largest series of revision mitral valve surgery via minithoracotomy reported a significantly lower mortality rate, less bleeding, and fewer necessary transfusions compared with standard sternotomy.

ACC = American College of Cardiology, AF = atrial fibrillation, AHA = American Heart Association, NIH = National Institutes of Health.

“It’s not an overstatement to say that this resource has changed the practice of valvular surgery,” says Glower. Funded entirely by industry and private donations, Glower’s commitment has maintained the database in recent years.

“We have an extraordinary resource offering decades of valuable research that is unfortunately not living up to its full potential,” Glower says. “We continue to seek new funding sources. This future boils down to future funding.”

Glower believes the Duke Valve Surgery Database and others maintained by leading medical centers should be blended into the database of the Society of Thoracic Surgeons to broaden retrospective resources.