Correlation of 3D Quantitative Coronary-Angiography Based Vessel FFR With Diastolic Pressure Ratio: A Single-Center Pooled Analysis of the FAST EXTEND and FAST II Studies

Abstract

Background. Vessel fractional flow reserve (vFFR) has a high diagnostic accuracy in assessing functional lesion significance compared with FFR. Nonhyperemic pressure ratios (NHPRs) were noninferior to FFR to guide revascularization of intermediate lesions. Therefore, the diagnostic performance of vFFR compared with NHPR warrants interest. Aim. To evaluate the diagnostic performance of vFFR with a generic diastolic pressure ratio (dPR) as a reference. Methods. The study population was derived from the FAST EXTEND and FAST II studies. Between January 2016 and September 2020, a total of 475 patients were enrolled. Results. Median dPR was 0.92 (interquartile range [IQR], 0.87-0.95), median vFFR was 0.86 (IQR, 0.80-0.90). The sensitivity, specificity, positive and negative predictive values, and diagnostic accuracy of vFFR ≤0.80 for dPR ≤0.89 were 66%, 92%, 79%, 85%, and 84%, respectively. Vessel FFR showed a good agreement with dPR (r=0.68), consistent among specific clinical lesion subsets and a high diagnostic accuracy for dPR ≤0.89 (area under the curve=0.89). Discordance between vFFR and dPR was observed in 78/492 cases (15.6%) and logistic regression analysis did not reveal any clinical, angiographic, or hemodynamic variables associated with vFFR and dPR discordance. Conclusion. Vessel FFR shows a good agreement with dPR and a high diagnostic accuracy for dPR ≤0.89.

Keywords: dPR, NHPR, vFFR, coronary physiology, diastolic pressure ratio, quantitative coronary angiography, vessel fractional flow reserve

Fractional flow reserve (FFR) derived from 3-dimensional quantitative coronary angiography (3D-QCA; CAAS workstation 8.2; Pie Medical) of the vessel (vFFR) has a high diagnostic accuracy in assessing functional lesion significance compared with FFR as the reference.^1,2 Whereas vFFR and other angiography-based FFR software platforms were initially validated against FFR, nonhyperemic pressure ratios (NHPRs) proved to be noninferior to FFR to guide revascularization of intermediate coronary lesions.³ Therefore, the diagnostic performance of vFFR compared with NHPR warrants interest. The aim of this analysis was to evaluate the diagnostic performance of vFFR with a generic diastolic pressure ratio (dPR) as a reference.

Methods

The study population was derived from the retrospective FAST EXTEND and prospective multicenter FAST II studies, in which pressure-waveform data were available for offline dPR calculation.^1,2 Details on inclusion and exclusion criteria and acquisition guidelines for FFR and vFFR measurements have been previously described. In brief, both studies used similar clinical and angiographic entry criteria. Diastolic pressure ratio was defined as the ratio of mean distal coronary artery pressure to mean aortic pressure in the resting state (Pd/Pa) over the entire period of diastole and was calculated from individual pressure waveforms using recently validated software.⁴ Ethical approval was waived by the institutional review board of the Erasmus University Medical Center.

Results

Between January 2016 and September 2020, a total of 475 patients (475 vessels) were enrolled. Median age was 66 years (interquartile range [IQR], 59-73), 67% were male, and most patients (65%) presented with stable angina. A total of 285 interrogated vessels (60%) were left anterior descending arteries, 53 (11%) were left circumflex arteries, 129 (27%) were right coronary arteries, and 8 (2%) were left main stems. Median dPR was 0.92 (IQR, 0.87-0.95) and median vFFR was 0.86 (IQR, 0.80-0.90). The sensitivity, specificity, positive and negative predictive values, and diagnostic accuracy of vFFR ≤0.80 for  dPR ≤0.89 were 66%, 92%, 79%, 85%, and 84%, respectively. The discriminative ability of vFFR ≤0.80 for dPR ≤0.89 as expressed by the area under the curve was 0.89 (95% confidence interval, 0.86-0.92) (Figure 1). The Spearman’s correlation observed in the overall cohort (r=0.68) was consistent among specific clinical and lesion subsets (male gender [67%], r=0.70; bifurcations [16%], r=0.76; tortuosity [16%], r=0.73; calcified lesions [28%], r=0.72; and tandem lesions [9%], r=0.89) as well as in most coronary arteries (left anterior descending [60%], r=0.70; left circumflex artery [11%], r=0.67; right coronary artery [27%], r=0.46; and left main [2%], r=0.81) (Figure 1).

Discordance between vFFR and dPR was observed in 78/492 cases (15.6%), of which 51/78 (65%) were dPR+/vFFR- and 27/78 (35%) were dPR-/vFFR+ (P<.01). Logistic-regression analysis did not reveal any clinical, angiographic, or hemodynamic variables associated with vFFR and dPR discordance.

Discussion

Previous validation studies demonstrated that NHPRs can be used as simpler, faster, and hyperemic–agent-free alternatives to FFR. More recently, 3D–angiography-based software platforms proved to further simplify physiological lesion assessment by abolishing the need for an invasive pressure wire. With growing  evidence demonstrating their adequate diagnostic performance and strong correlation with FFR, along with several large outcome trials on the way, the use of angiography-based FFR holds significant potential in increasing the uptake of physiology-guided percutaneous coronary intervention and subsequently improving patient outcomes.

To our knowledge, the present study is the first to compare the diagnostic performance of vFFR using dPR as a reference. Whereas vFFR showed a significant correlation and a good diagnostic performance when compared with dPR, our study also revealed a relevant number of cases (15.6%) that would receive differential treatment recommendations based on either the vFFR or dPR result. The latter is in line with previous studies reporting discrepancies between iFR and FFR in 15%-20% of cases and appears to be related to clinical and hemodynamic parameters, such as heart rate, blood pressure, left ventricular end-diastolic pressure, and vessel flow pattern affecting hyperemic coronary flow velocity.⁵ When using 3D-QCA-based FFR instead of invasive FFR, the modeled hyperemic flow vs actual observed hyperemic flow should be taken into account as particular subsets of patients are less vulnerable to the induction of hyperemia.

Future studies are needed to better understand and explore the relevance of the discordance between vFFR and dPR along with the clinical value of vFFR in currently ongoing, large-scale outcome trials.

Study limitations. Limitations of the analysis include its retrospective nature and the lack of clinical follow-up, which might be of particular interest in patients with discordant vFFR/dPR values. Moreover, nonavailability of pressure-waveform data for patients treated outside the Erasmus Medical Center precluded dPR computation in all FAST II patients.

Conclusion

The main findings of this study can be summarized as follows: (1) vFFR shows a good agreement with dPR, consistent among specific clinical and lesion subsets; and (2) vFFR shows a high diagnostic accuracy for dPR ≤0.89.

Affiliations and Disclosures

From the ¹Department of Cardiology, Thoraxcenter, Erasmus University Medical Center, Rotterdam, The Netherlands; ²the First Department of Cardiology, Medical University of Warsaw, Warsaw, Poland; and ³Cardialysis BV, Rotterdam, The Netherlands.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. The authors report no conflicts of interest regarding the content herein.

Manuscript accepted January 3, 2022.

Address for correspondence: Joost Daemen, MD, PhD, Department of Cardiology, Room Rg-628, Erasmus University Medical Center, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands. Email: j.daemen@erasmusmc.nl

References

1. Neleman T, Masdjedi K, Van Zandvoort LJC, et al. Extended validation of novel 3d quantitative coronary angiography-based software to calculate vFFR: the FAST EXTEND study. JACC Cardiovasc Imaging. 2021;14(2):504-506. Epub 2020 Sep 30. doi:10.1016/j.jcmg.2020.08.006

2. Masdjedi K, Tanaka N, Van Belle E, et al. Vessel fractional flow reserve (vFFR) for the assessment of stenosis severity: the FAST II study. EuroIntervention. 2022;17(18):1498-1505. doi:10.4244/EIJ-D-21-00471

3. Berry C, McClure JD, Oldroyd KG. Meta-analysis of death and myocardial infarction in the DEFINE-FLAIR and iFR-SWEDEHEART trials. Circulation. 2017;136(24):2389-2391. Epub 2017 Sep 28. doi:10.1161/CIRCULATIONAHA.117.030430 

4. Ligthart J, Masdjedi K, Witberg K, et al. Validation of resting diastolic pressure ratio calculated by a novel algorithm and its correlation with distal coronary artery pressure to aortic pressure, instantaneous wave-free ratio, and fractional flow reserve. Circ Cardiovasc Interv. 2018;11(12):e006911. doi:10.1161/CIRCINTERVENTIONS.118.006911

5. Cook CM, Jeremias A, Petraco R, et al. Fractional flow reserve/instantaneous wave-free ratio discordance in angiographically intermediate coronary stenoses: an analysis using doppler-derived coronary flow measurements. JACC Cardiovasc Interv. 2017;10(24):2514-2524. doi:10.1016/j.jcin.2017.09.021