What a fragmented stack actually costs
A structural heart program needs aortic planning, mitral planning and left atrial appendage planning. Bought separately, each arrives with a quote covering license and support. The quote is the smallest part of what the institution takes on.
Each application carries its own implementation, its own integration into the archive and the reporting environment, and its own security and privacy review. Each carries a training curve, paid in clinician and technologist time at the point where that time is scarcest, and paid again at every staff change. Each carries an upgrade cycle that has to be scheduled around clinical operations, and a renewal negotiation that arrives on its own calendar. Each is a vendor relationship that somebody has to manage.
Then there is the cost that does not appear on any line. Two established planning packages measuring the same 100 patients
produced significantly different mean annulus areas, 482 mm² against 464 mm², along with differing perimeter and coronary height measurements. An institution running separate tools for separate procedures is running separate measurement standards, with no single source of truth for a number that determines device selection. That is a governance exposure, and it is invisible until someone asks which tool produced a given figure.
None of these are hypothetical charges. They are simply distributed across enough budgets that no single one looks large enough to question.
The demand case does not depend on a single policy decision
Two developments in the last year changed the size of the eligible population, and they are worth separating because one of them travels across every market.
The policy one is jurisdictional. CMS
finalized a revised TAVR national coverage determination on 10 September 2026, extending Medicare coverage to asymptomatic severe aortic stenosis under coverage with evidence development, removing the CED requirement for symptomatic patients, and replacing facility procedural volume minimums with infrastructure and quality expectations. The
2025 ESC/EACTS valvular heart disease guidelines moved in the same direction independently, lowering the age for a Class I, Level A TAVI recommendation to 70.
The economic one is not jurisdictional. A
2026 cost-effectiveness analysis modeled early TAVI against clinical surveillance for asymptomatic severe aortic stenosis across nine European healthcare systems and found early intervention to be the economically dominant strategy in all nine, with cost-effectiveness probabilities ranging from 97.3 percent in the UK to 99.9 percent in Belgium. Modeled lifetime cost savings per patient ranged from £1,788 in the UK to CHF 15,802 in Switzerland, with quality-adjusted life year gains of 0.18 to 0.23.
That finding should be presented with its limits, because a finance audience will find them anyway. It is a lifetime Markov model extrapolating from 3.8 years of trial follow-up, and it includes manufacturer-affiliated authors. The accompanying editorial in the same issue
makes the counterargument directly: long-term outcomes are inferred rather than observed, a consistent mortality benefit has not been demonstrated, and the absolute gain is on the order of two to three months of quality-adjusted life expectancy per patient.
Stated with those caveats, the claim still holds. The direction of travel on eligible volume is not contingent on any one ruling.
Where the procedural economics actually sit
Lab time on a structural procedure is among the most expensive blocks an institution schedules, and planning quality moves it measurably.
PREDICT-LAA randomized 200 patients to standard planning or CT simulation-based planning for appendage closure. Its primary composite endpoint was not met, at p = 0.08, and that belongs in any honest summary. The secondary endpoints are where the operational effect appears: procedure time fell from 55.2 to 45.1 minutes, fluoroscopy time from 17.6 to 12.5 minutes, contrast volume from 80 to 59 mL, and total devices used fell significantly. A 2026 meta-analysis across six studies and 978 patients found
procedure time shorter by 13.29 minutes and device resizing reduced at a relative risk of 0.42.
Fewer device changes and shorter room time are the two variables in a structural program that translate directly into cost, and both respond to planning.
On the analysis side, assisted annulus and landmark detection across the cvi42 Interventional Planning Suite covering aortic, mitral and LAAC removes 80 to 100 clicks from a typical case, by Circle's own measurement. Published work on semi-automated measurement in a comparable setting
found reading time roughly halved with identical prosthesis sizing in 96 percent of cases for an experienced reader.
Circle publishes an ROI model that lets an institution run these variables against its own case volumes rather than against a vendor's assumptions. The outputs are explicitly modeled, not guaranteed.
Platform headroom is the part competitors cannot match
Here is the argument that separates a platform purchase from a tool purchase, and it has nothing to do with any individual patient.
Once cvi42 is installed, integrated into the archive and the reporting environment, authenticated against the institution's directory, and trained on, the marginal cost of adding a capability to it is a fraction of the cost of standing up a new vendor to provide that capability. No second integration, no second security review, no second training curve, no second renewal negotiation.
Coronary plaque analysis is a useful illustration of what that headroom is worth. cvi42 | Plaque received FDA clearance in October 2025, and coronary plaque quantification has had a reimbursement pathway under CPT 75577 since January 2026. An institution already running cvi42 for structural planning can add that capability on the platform it already operates.
To be precise about what this is and is not: coronary plaque assessment and structural pre-planning are separate, protocol-specific acquisitions ordered for separate clinical indications. Plaque quantification applies to a coronary CTA independently ordered for suspected or known coronary artery disease. It is not an addition to a structural planning study, and nothing about a TAVR program creates a plaque billing opportunity. The argument here is about the economics of the platform, not about any individual scan.
The same logic applies to cardiac MR, cardiac CT and vascular CT. The platform is the asset. What runs on it is the return.
The utilization that keeps going
Structural patients remain in the imaging service after the procedure. The
2025 multi-society consensus on CT for prosthetic heart valve assessment is clear that cardiac CT is not used for routine screening of valve degeneration, so this should not be modeled as scheduled surveillance volume. It is indication-driven: suspected leaflet thrombosis, suspected valve failure, coronary access assessment, and planning for a redo procedure.
Those indications accumulate with the implanted population. A platform that reads the index planning study, the diagnostic studies before it, and the follow-up study when a clinical question arises, holds utilization across the whole relationship rather than at one moment in it. A planning-only tool stops earning when the valve is deployed.
The comparison worth making
Compare three quotes against one quote and the three win. Compare three integrations, three training curves, three security reviews, three renewal cycles, three measurement standards and one capability ceiling against their single-platform equivalents, and the answer changes.
The second comparison is the one that holds up in the review.
Walk through the total cost comparison for your program →
