To the editor,
We read with interest the review by Strepkos et al. on angiography-derived Fractional Flow Reserve (FFR). While the article provides a broad overview, we wish to address several methodological distinctions and omissions to provide a more comprehensive perspective.
First, the categorization of Murray law–based quantitative flow ratio (μFR), the correct terminology, as QFR is a trademark of Medis, alongside QFR requires correction. As μFR relies on Murray’s law rather than the Gould-Kirkeeide equation computational fluid dynamics similations utilized in QFR, it represents a distinct method to estimate FFR from angiography. Furthermore, regarding physiological assumptions, both μFR and QFR (Quantitative Flow Ratio) account for blood pressure changes at hyperemia. This consideration is logically essential for the accurate calculation of the angiography-derived Index of Microcirculatory Resistance (angio-IMR), a capability that should be clearly articulated.
Regarding specific software platforms, important nuances were overlooked. For instance, FFRangio (CathWorks) necessitates 3 angiographic projections, a workflow drawback compared to fewer views, but offers the distinct advantage of reconstructing the entire coronary tree. Indeed, the acquisition of 3 views per vessel is not routine in the cardiologist’s workflow; therefore, a learning curve is required. Conversely, the utility of CaFFR is limited by its requirement for a consumable pressure transducer (FlashPressure). This adds cost, operational complexities, and procedural time, making the cited 4-minute processing time highly optimistic. Moreover, based on experience as an angio-FFR core lab that validated all 5 methods in Ninomiya et al., it is CaFFR, not QFR, that typically demands extensive manual interaction during analysis (e.g., contour correction and manual calculation of the frame count), and the actual time to analyze a case is likely longer.
The review also omits several emerging technologies: notably MedHub, which uses machine learning to directly estimate FFR, presenting a novel alternative methodology, and several other computational fluid dynamics-derived angio-FFR systems, such as AccuFFR, which also provides an angio-IMR solution. Additionally, we caution against claims regarding derived physiological indices; to our knowledge, current iterations of Medis and Rainmed software cannot estimate Coronary Flow Reserve (CFR) and, consequently, Angio-Microcirculatory Resistance Reserve (Angio-MRR). On the contrary, with angio-FFR capable of deriving the pullback of the entire vessel, the pullback pressure gradient (PPG), a metric reflecting the diffuseness versus focality of the coronary disease phenotype could be derived from angio-FFR with reasonable accuracy.
Finally, the evidence base is evolving rapidly. The PIONEER IV trial, which demonstrated the noninferiority of QFR-guided PCI lesion selection and stent optimization, should be included as a pivotal outcome study for QFR. Similarly, the FLAVOUR II trial and the broader concept of QFR-guided stent optimization were not discussed, despite their clinical relevance. The review also limits its report on μFR; for example, FLAVOUR II was 80% conducted with μFR, and the recently presented FAVOR IV-QVAS trial at AHA 2025 was omitted.
We believe these clarifications are vital for readers to accurately assess the current state of wire-free physiology.
Sincerely,
John TsungYing Tsai and Patrick W. Serruys
CRediT authorship contribution statement
Tsung-Ying Tsai: Writing– original draft, Methodology, Conceptualization. Patrick W. Serruys: Writing– review & editing.
Declaration of Competing Interest
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
