Please give us your honest opinion on the results of this study.
I myself was hoping that FLOWER-MI1 would show that the event rate was lower in the FFR-guided group than in the angio-guided group, but the results showed that there was no difference in the one-year event rate between the two groups.
However, the essence of the study is that in STEMI, wherever there is a high likelihood of multiple vulnerable plaques, deferential FFR is associated with similar event rates as stenting of all stenoses.
In the FFR group, about 40% of lesions were deferent by FFR and treated with medical therapy, but there was no increase in events compared to the group in which almost all stenoses were stented. In other words, I think this study negates the plaque-sealing effect of stenting.
What do you think is the reason for the lack of difference in the primary endpoint?
In the post-hoc analysis of FAME 1, there was a 5.1% event reduction in the FFR-guided group compared with the angio-guided group in ACS patients, mainly NSTEMI, as in SAP.2 However, in FLOWER-MI, no difference was observed at all.
The reason for this is that in FAME 1, ACS was entered after five days of onset, while in FLOWER-MI, non-culprit lesions were treated at an average of 2.6 days. Therefore, it is difficult to determine peri-procedural MI, and it is not possible to accurately determine the initial occurrence of MI. In FAME 1, the frequency of early events is clearly higher in the angio-guided group, while no difference is observed in FLOWER-MI. This point is considered to have a significant impact on the results.
Another point worth noting is that there were 10 deaths at one year in the angio-guided group and nine deaths in the FFR-guided group. In the angio-guided group, seven deaths were sudden or cardiac deaths, while in the FFR-guided group, only two deaths were sudden or cardiac deaths. The remaining seven were non-cardiac deaths (three from malignancy, one from post-traumatic hemorrhage, one from acute pancreatitis, one from renal failure, and one from cerebral infarction). This result will be useful for your daily clinical practice.
Will these results have any impact on your daily clinical practice?
For the reasons mentioned above, FLOWER-MI should not negate the validity of FFR-guided PCI, especially in ACS, and should not negate the validity of FFR-guided PCI in routine clinical practice.
Do you think that the results will vary depending on the modality and measurement of FFR, iFR, FFRCT, etc.?
The combined data of DEFINE FLAIR and iFR Swedeheart reported that iFR-based defer is useful in defer lesions of ACS except STEMI.3 However, in the hyperacute phase of STEMI, the coronary circulation is different from that at rest, and it is necessary to be careful when evaluating the resting index, including iFR.
In this regard, if the resting index is used in this study, it is likely that more lesions will be considered for PCI. The same would be true of Angio-guided FFR or virtual FFR such as FFRCT.
What are the findings from this study and what would you like to share with us?
As mentioned at the beginning, this study can be seen as a paper that denied the effect of plaque sealing with stents. Please do not misinterpret this paper as a denial of FFR-guided decision making, but rather as a demonstration that FFR-deferential lesions are safe, even in STEMI culprit lesions.
- Puymirat E, et al. N Engl J Med. in press
- Sels JWEM, et al. JACC Cardiovasc Interv. 2011; 4: 1183-1189
- Escaned J, et al. JACC Cardiovasc Interv. 2018; 11: 1437-1449