Seeing the Big Picture: Outcomes Registries for Cardiac Surgery and Transcatheter Valve Therapy Offers Insight Into Current State of Care

May 1, 2026

Share this page

Cardiovascular medicine has benefited from a remarkable stream of surgical and procedural breakthroughs over the past few decades, establishing safe and effective treatment options for even the most serious heart conditions. As these treatments become available across an increasing number of programs, accurate outcomes data are key to understanding the current state of care.

The Society of Thoracic Surgeons and the American College of Cardiology (ACC) have been at the forefront of the effort to establish a shared, data-driven definition of quality. This partnership has led to voluntary registries that have helped to standardize how outcomes are defined while accounting for patient risk, measure those outcomes at scale, and turn the results into actionable feedback.

Adult Cardiac Surgery Database (ACSD)

The STS Adult Cardiac Surgery Database (ACSD) was launched in 1989, originally to enable consistent measurement of outcomes and support quality improvement across adult cardiac surgery programs. Over time, it has grown into what STS describes as a premier clinical outcomes registry, containing millions of procedure records and broad national participation.

“It took a long time because the ACSD is voluntary,” notes Paul Kurlansky, MD, Associate Professor of Surgery at Columbia and Associate Director for the Center of Innovation and Outcomes Research. “Now, 95-98% of non-governmental hospitals that perform cardiac surgery in the country participate in the ACSD. So it has truly become a functional national database.”

The ACSD houses almost 8.5 million adult cardiac surgery procedure records and currently has more than 3,500 participating physicians. STS also maintains the General Thoracic Surgery Database (GTSD), with nearly 800,000 records of general thoracic surgery procedures, and the Congenital Heart Surgery Database (CHSD), with more than 700,000 records of congenital heart surgery procedures.

Transcatheter Valve Therapy (TVT) Registry

The STS/ACC Transcatheter Valve Therapy (TVT) Registry emerged from a different but equally important need: how do you safely monitor and improve outcomes for rapidly evolving device-based therapies, such as transcatheter aortic valve replacement (TAVR) and transcatheter mitral repair, as they move into widespread use?

Originally approved for patients who were not candidates for surgery, these minimally invasive procedures have rapidly evolved into viable alternatives to surgery for a much broader patient population. And with that rapid expansion comes an urgent need for a structure designed to support device surveillance, real-world outcomes tracking, and quality improvement at a national scale.

The STS and American College of Cardiology (ACC) partnered to create the TVT Registry, working in close collaboration with the FDA, CMS, and the Duke Clinical Research Institute. Established in 2011, the TVT Registry now contains data on over 500,000 patients, giving the clearest picture yet of the safety and efficacy of transcatheter therapies.

It’s worth noting that while the registry's public reporting program is voluntary, participation is generally mandatory for hospitals seeking Centers for Medicare & Medicaid Services (CMS) reimbursement for transcatheter valve procedures.

What Do These Registries Track?

Both registries rely on prospectively collected, audited clinical data using standardized definitions and structured case report forms, which are then risk-adjusted to account for the complexity of high-risk cases. 

This use of risk adjustment models and observed-to-expected approaches allows outcomes to be compared more fairly across sites and over time. (For example, a center that treats higher-risk patients may look worse on paper, even if the care provided is excellent.)

The STS audit program examines the accuracy, consistency, and completeness of submitted data and selects a subset of participating sites for independent audits. This thorough approach to data integrity is important because registry data are used for a range of additional calculations, including benchmarking, ratings, payer programs, or public reporting.

Both registries use a 3-star rating system: three stars represent “better than expected,” two stars represent “as expected,” and one star represents “worse than expected” outcomes. For the surgical registry, there are six categories given star ratings:

The TVT registry tracks criteria like mortality, stroke, bleeding, and kidney injury across the following categories:

How The Data Helps

Registries like the ACSD and TVT Registry drive improvement in several ways. By highlighting high-performing programs, it offers visibility and encourages other programs to study and emulate what works. Stronger-performing programs may have resources and strategies that extend beyond the procedure and can be added to address deficiencies.

“It shows not just the processes of care, but also the depth of the bench, if you will,” says Dr. Kurlansky. “It helps you see what works. I can tell you that at Columbia, we have expert cardiac surgeons and cardiac surgical fellows, but also dedicated, trained pulmonary and anesthesia intensivists. So many skilled, intelligent people watching over these patients and putting input into their care so no one falls through the cracks.”

For one-star programs, identifying weaknesses in their own care and suggesting areas to dedicate resources for improvement can be transformational. “One of the things I do at Columbia is an initiative called HeartSource: we contract with outside institutions in order to help them build or improve their cardiac program,” says Dr. Kurlansky. “And one of the things that we regularly do for surgical programs is go over their STS data.”

Even for high-performing programs, when registry results show an outlier, it can be an opportunity to reassess and look for potential improvements. “A few years ago, we had a bump in mortality. It amounted to a statistical outlier. But Hiroo Takayama, MD, PhD, Vice Chair of Cardiac, Vascular, and Thoracic Surgery, and the surgeons got very upset about it. They really focused on it and drilled down, and they've done a remarkable job. Since Arnar Geirsson, MD, arrived to direct our Robotic Mitral Valve Surgery program, we’ve performed over 200 mitral valve repairs with 0% mortality. Dr. Takayama has performed over 200 aortic cases with a mortality rate of roughly 2%. This is phenomenological. I mean, it's just truly amazing.”

For patients, the star system offers a simple, easy-to-understand decision-making tool by providing objective measurements of program success. “The star ratings try to look at what distinguishes your program,” says Dr. Kurlansky. “It shows the probability range for your outcome, using statistical models and the general experience in the cardiac surgical data, based on your patients. Two stars isn't necessarily bad; in fact, it may be good in certain cases, but three stars is good.”

There is nuance to consider: for example, a program that focuses only on simple cases may receive a three-star rating, whereas the same rating for a program like Columbia that specializes in high-risk cases indicates a much broader level of excellence. “Right now, we're in a situation where actually all four NewYork-Presbyterian Hospital sites doing cardiac surgery have rankings in the three-star range. So this is really outstanding.”

Challenges

Ideally, the registries would provide near-real-time insights, but cleaning, auditing, risk adjustment, and adjudication take time. As a result, registry reporting lags behind real-world activity.

Maintaining up-to-date registry participation requires trained personnel, time, and institutional investment. For smaller programs or those with limited administrative bandwidth, this can be a difficult hill to climb, especially when data definitions change, new devices are added, or requirements expand.

Outcomes such as 30-day events, 1-year survival, functional status, or quality of life can be difficult to capture reliably, especially when patients receive follow-up care elsewhere. And because electronic health record (EHR) data is not yet consistently structured, standardized, or easily exportable, there is a manual element to program submissions that can lead to duplication and errors.

Despite these challenges, these registries have shown that they are worth the investment, delivering data that institutions can use to guide their cardiac surgery and interventional cardiology programs. Comparing outcomes to national risk-adjusted expectations provides a better benchmark than relying on perception or intuition. And the audits and standardized definitions used help make results more credible for internal improvement and external reporting, enabling teams to turn them into actionable feedback.

These registries continue to expand, including more cases and adding new procedures, metrics, and risk models that further refine the data, helping to bring the big picture of cardiac care into even greater focus.

Related

Back to top