The TAVR Rule Left One Group of Patients Out
Dear colleagues,
Medicare's new decision on TAVR is easy to read as a single event. I read it as two clocks that start at different times, and the gap between them matters for revenue plans, site strategy and evidence budgets.
On September 10, 2026, CMS issued its final decision memo reconsidering the national coverage determination (NCD) for TAVR, CAG-00430R2. Symptomatic severe aortic stenosis (AS) is covered as reasonable and necessary, without the coverage with evidence development (CED) study requirement that applied under the 2019 NCD. Asymptomatic severe AS is covered only under CED, in a CMS-approved study with an active, contemporaneous comparator. As of October 20, CMS's TAVR CED page listed no approved studies under the new NCD.
Coverage for symptomatic patients is in place under the NCD, but the claims systems have not caught up. As of October 20, CMS had issued no change request (CR) for the decision. The Medicare Claims Processing Manual still expects TAVR claims to carry the two-surgeon modifier, the investigational-service modifier, the clinical research diagnosis code and a trial identifier, even though the NCD allows one operator to perform the procedure. The memo itself states no TAVR-specific effective date.
Recent history suggests patience. In the three most recent heart valve NCDs, claims implementation followed the coverage effective date by 187 to 357 days. For both tricuspid decisions, CMS's Medicare Learning Network (MLN) Matters articles stated that contractors would not search their files for interim claims but would adjust claims brought to their attention.
My advice to hospital programs is to confirm interim claims handling with their Medicare Administrative Contractor (MAC) and their billing compliance staff, and to track every affected claim. My advice to manufacturers is to describe the gap to sites accurately and to leave the billing approach to the provider.
The asymptomatic clock has not started. An FDA label for asymptomatic patients, which FDA approved for one TAVR system on April 30, 2025, does not open Medicare coverage on its own. Medicare volume in those patients begins when CMS approves a CED protocol and the study enrolls. In my view, a 2027 plan that books asymptomatic Medicare revenue without a protocol approval date is ahead of the evidence.
CMS's reasoning tells sponsors where to focus. It judged the symptomatic evidence sufficient, citing randomized trials with 5 to 10 years of follow-up and registry data on more than 450,000 patients. It judged the asymptomatic evidence promising but incomplete, pointing to unknown valve durability and a surveillance arm it considered unlikely to be replicated in real-world practice. I read that as a design brief: comparator choice and long-term reintervention data are likely to decide how quickly asymptomatic coverage can broaden.
The comparator is the question I would settle first. CMS states that surgical aortic valve replacement is "generally not indicated" for asymptomatic patients, which leaves close surveillance as the likely comparator for many protocols. The memo does not say whether CMS treats surveillance as an "established alternative treatment," and that is a question to put to CMS before a protocol is final.
The program rules move the economics from the institution to the operator. Each TAVR operator needs at least 20 transcatheter cardiac valve procedures a year, at least 15 of them TAVR, or 40 and 30 over two years. Mitral, tricuspid and pulmonic procedures count toward the total, and hospital volume minimums and the two-operator requirement are gone. Site targeting built on hospital volume is likely to misclassify some accounts, and operator-level data is the better basis.
One part of the decision deserves more attention than it has received. CMS's third CED question asks whether patient-centered outcome measures can replace volume criteria for TAVR operators. A study designed to test that generates the evidence CMS said it wants on whether outcomes can replace volume criteria. CMS has not said it will revisit the operator thresholds, and it called replacing volume with outcomes "premature."
On quality programs, the American College of Cardiology has written that registries "may meet" the requirement. CMS's own text says that "registry participation alone" does not satisfy its continuous quality improvement requirement. I would plan to the CMS text and document how outcomes are reviewed and what programmatic changes follow.
For investors and boards, the distinction is simple to state and easy to lose in a model. Coverage and claims timing are separate risks. Asymptomatic revenue depends on a protocol approval date that no sponsor can yet point to, because CMS has approved no study.
Over the coming months I will be watching four things. The first is the CR and MLN Matters article for NCD 20.32, which in recent heart valve decisions set the effective date and the claims edits. The second and third are the first studies CMS lists under the new NCD and revisions to the manual sections that still carry the 2019 text. The fourth is the separate aortic regurgitation reconsideration, CAG-00430R3, with a proposed decision CMS expects by March 10, 2027.
I hope this is useful as you plan for 2027.
With kind regards,
Tina Ivovic
UnifiMed
Policy status as of October 20, 2026. Source: CMS final decision memo for the TAVR NCD reconsideration (CAG-00430R2), September 10, 2026. General information; not legal, coding or billing advice.

