Fractional flow reserve (FFR) has never been investigated in patients with aortic stenosis (AS). From 2002 to 2010, we identified 106 patients with AS and coronary artery disease with at least one intermediate lesion treated according to FFR guidance. We matched 212 contemporary control patients with AS in which revascularization was decided on angiography only. More patients in the FFR-guided group underwent percutaneous coronary intervention (24% vs 13%; p = 0.019), whereas there was a trend toward less coronary artery bypass grafting (CABG) performed. After FFR, the number of diseased vessels was downgraded within the FFR-guided group (from 1.85 ± 0.97 to 1.48 ± 1; p <0.01) and compared with the angio-guided group (1.48 ± 1 vs 1.8 ± 0.97; p <0.01). Less aortic valve replacement was reported in the FFR-guided group (46% vs 57%; p = 0.056). In patients who underwent CABG, less venous conduits (0.5 ± 0.69 vs 0.73 ± 0.76; p = 0.05) and anastomoses (0.61 ± 0.85 vs 0.94 ± 1; p = 0.032) were necessary in the FFR-guided group. Up to 5 years, we found no difference in major adverse cardiac events (38% vs 39%; p = 0.98), overall death (32% vs 31%; p = 0.68), nonfatal myocardial infarction (2% vs 2%; p = 0.79), and revascularization (8% vs 7%; p = 0.76) between the 2 groups. In conclusion, FFR guidance impacts the management of selected patients with moderate or severe AS and coronary artery disease by resulting into deferral of aortic valve replacement, more patients treated with percutaneous coronary intervention, and in patients treated with CABG, into less venous grafts and anastomoses without increasing adverse event rates up to 5 years.
Fractional flow reserve (FFR) is a widely used invasive index to detect the ischemic potential of coronary stenoses. FFR-guided revascularization strategies have been shown beneficial in patients with intermediate stenoses and multivessel coronary artery disease (CAD). The reliability and clinical usefulness of FFR has not been tested in patients with concomitant aortic stenosis (AS). Nevertheless, these are the very patients difficult to investigate with noninvasive functional testing and presenting with associated CAD in up to 50% of the cases. We aimed to investigate the impact of FFR measured at the time of the diagnostic coronary angiography on percutaneous and surgical revascularization strategies and its related clinical outcome in patients with AS.
Methods
From 2002 to 2010, we retrospectively identified 106 patients with AS and significant CAD in whom at least one intermediate lesion was either revascularized with an FFR value ≤0.80 or deferred with FFR >0.8 (FFR-guided group). Then, from 694 contemporary patients in whom the decision to revascularize was based on angiography only, we matched 212 as comparator (angio-guided group).
Inclusion criteria were the presence of at least one intermediate stenosis (diameter stenosis: 50% to 70%) of a major coronary artery at the time of angiography; aortic valve area ≤1.5 cm 2 , and/or aortic mean pressure gradient ≥20 mm Hg. Severe AS was defined with a valve area ≤1 cm 2 and/or aortic mean pressure gradient ≥40 mm Hg.
All patients underwent left/right-sided heart catheterization. Aortic valve area was calculated with the Gorlin formula. Coronary angiography was performed with 6F diagnostic catheters. Experienced operators not involved in data analysis assessed stenosis severity. Multivessel disease was defined as the presence of stenosis in ≥2 major coronary arteries.
FFR was measured as previously described and left to the operator’s discretion. Percutaneous coronary intervention (PCI) and surgical interventions were left to the operator’s discretion. Coronary artery bypass grafting (CABG) and aortic valve replacement (AVR) performed within 6 months from the diagnostic coronary angiography were referred to the index procedure.
Primary end point was the rate of major adverse cardiac events (MACE), defined as overall death, myocardial infarction (MI), and repeat revascularization up to 5 years. Secondary end points were all the individual end points included in MACE and AVR. Follow-up was obtained through telephone contacts or outpatient visits. Date of death was retrieved from Belgium national death registry. Informed consent to the use of personal data was obtained from each patient.
All analyses were performed with SPSS 21.0 (IBM Inc, New York, New York) and with Prism GraphPad 5.0 (GraphPad Software Inc, La Jolla, California). Nonparsimonious propensity matching with logistic regression modeling was used to select controls with an FFR-guided to angio-guided ratio of 1:2 without replacement (greedy matching algorithm), using a caliper width equal to 0.2 of the SD of the logit of the propensity score. Standardized bias was compared between the unmatched and matched groups to verify the efficacy of this strategy in selecting well-matched controls. In the propensity score, the following were included: gender, age, body mass index, smoking habit, diabetes mellitus, peripheral vascular disease, hypertension, hyperlipidemia, previous CABG, a family history of CAD, atrial fibrillation, logistic EuroSCORE, acute coronary syndrome at presentation, aortic mean pressure gradient, pulmonary hypertension, the number of diseased vessels, left ventricular ejection fraction, and New York Heart Association class.
Categorical variables are reported as frequencies and percentages. Normal distribution was tested with the Kolmogorov–Smirnov test. Comparisons between continuous variables were performed with the Student t test or Mann–Whitney test. Comparisons between categorical variables were evaluated with the Fisher’s exact test or the Pearson chi-square test as appropriate. Difference in survival rate was calculated by applying Kaplan–Meier curves, and survival rates are reported at 5 years. Log-rank tests were used to compare the 2 groups.
Probability values were 2-sided, and values of p <0.05 were considered significant.
Results
From January 2002 to December 2010, 318 patients were included in the study: (1) 106 patients with AS and at least one intermediate stenosis, in whom revascularization strategy was decided according to FFR value and (2) 212 matched patients, in whom the decision to revascularize was based on angiography only. The 2 groups were well matched with respect to clinical characteristics. AS severity did not differ between the groups ( Table 1 ).
| Variable | FFR-guided (n=106) | Angio-guided (n=212) | P value |
|---|---|---|---|
| Men | 76 (72%) | 147 (69%) | 0.66 |
| Age (years) | 73 ± 10 | 73 ± 9 | 0.89 |
| Body mass index (kg/m 2 ) | 27 ± 4 | 27 ± 4 | 0.94 |
| Logistic euroSCORE (%) | 12 ± 7 | 13 ± 7 | 0.68 |
| Smoke | 40 (38%) | 67 (32%) | 0.27 |
| Diabetes Mellitus | 25 (24%) | 50 (24%) | 1 |
| Peripheral vascular disease | 12 (11%) | 23 (11%) | 0.9 |
| Hypertension | 63 (59%) | 118 (56%) | 0.52 |
| Hyperlipidemia | 59 (56%) | 114 (54%) | 0.75 |
| Previous coronary artery bypass graft | 7 (7%) | 9 (4%) | 0.36 |
| Family history | 28 (26%) | 57 (27%) | 0.93 |
| Atrial Fibrillation | 4 (5%) | 4 (2%) | 0.31 |
| Pulmonary hypertension | 17 (16%) | 33 (16%) | 0.91 |
| Acute coronary syndrome | 9 (8%) | 17 (8%) | 0.88 |
| Symptomatic aortic stenosis | 60 (57%) | 115 (54%) | 0.69 |
| Severe aortic stenosis | 70 (66%) | 146 (69%) | 0.61 |
| Aortic valve area (cm 2 ) | 0.77 ± 0.27 | 0.79 ± 0.29 | 0.72 |
| Aortic peak gradient (mmHg) | 59 ± 24 | 62 ± 25 | 0.56 |
| Aortic mean gradient (mmHg) | 38 ± 19 | 40 ± 20 | 0.46 |
| New York Heart Association Class | 2 ± 1 | 2 ± 1 | 0.43 |
| Left ventricular ejection fraction (%) | 69 ± 17 | 69 ± 17 | 0.91 |
Procedural characteristics and initial treatment strategy are summarized in Table 2 . More patients in the FFR-guided group underwent PCI, whereas there was a trend toward more CABG in the angio-guided group. Patients treated with PCI had lower AS severity as compared with patients treated with CABG (aortic valve area: 0.95 ± 0.29 vs 0.76 ± 0.28, p <0.01; aortic mean gradient: 27 ± 15 vs 41 ± 20; p <0.01). At baseline angiogram, the number of diseased vessels was similar in the 2 groups. After functional assessment with FFR, the number of diseased vessels was significantly downgraded within the FFR-guided group (p <0.01) and compared with the angio-guided group (p <0.01). In patients who underwent CABG, the number of arterial grafts and anastomoses per patient was similar between the 2 groups, whereas significantly less venous conduits were used in the FFR-guided group, along with less venous anastomoses. A trend toward less AVR was observed in the FFR-guided group.
| Variable | FFR-guided | Angio-guided | P value |
|---|---|---|---|
| Narrowed coronary arteries | 1.85 ± 0.97 | 1.80 ± 0.97 | 0.62 |
| Narrowed coronary arteries after reclassification by Fractional flow reserve | 1.48 ± 1 | 1.80 ± 0.97 | <0.01 |
| Diameter stenosis (%) | 65 ± 23 | 67 ± 22 | 0.48 |
| Coronary artery bypass graft | 41 (39%) | 103 (49%) | 0.09 |
| Percutaneous coronary intervention | 25 (24%) | 28 (13%) | 0.019 |
| Coronary arteries treated with percutaneous coronary intervention | |||
| Left main | 2 (7%) | 0 (0%) | 0.04 |
| Left anterior descending | 7 (23%) | 11 (37%) | 0.6 |
| Left circumflex | 7 (23%) | 8 (26%) | 0.26 |
| Right | 14 (47%) | 11(37%) | 0.012 |
| Coronary arteries treated with Coronary artery bypass graft | |||
| Left main | – | – | – |
| Left anterior descending | 22 (38%) | 76 (46%) | 0.36 |
| Left circumflex | 19 (33%) | 49 (30%) | 0.79 |
| Right | 17 (29%) | 40 (24%) | 0.56 |
| Patients treated with drug-eluting stent | 12/25 (48%) | 9/28 (32%) | 0.37 |
| Stent length (mm) | 19 ± 8 | 24 ± 18 | 0.23 |
| Stent diameter (mm) | 3.13 ± 0.74 | 3.21 ± 0.37 | 0.61 |
| Aortic valve replacement | 49 (46%) | 122 (57%) | 0.056 |
| Biological/Mechanical | 42 (81%)/10 (19%) | 106 (86%)/17 (14%) | 0.36 |
| Clamp time (minutes) | 87 ± 30 | 88 ± 37 | 0.88 |
| Perfusion time (minutes) | 118 ± 36 | 122 ± 52 | 0.59 |
| Total grafts | 1.3 ± 1 | 1.5 ± 1 | 0.27 |
| Arterial grafts | 0.80 ± 0.64 | 0.75 ± 0.57 | 0.59 |
| Venous grafts | 0.50 ± 0.68 | 0.73 ± 0.76 | 0.05 |
| Total anastomoses | 1.46 ± 1.2 | 1.78 ± 1.3 | 0.13 |
| Arterial anastomoses | 0.84 ± 0.73 | 0.86 ± 0.72 | 0.87 |
| Venous anastomoses | 0.61 ± 0.85 | 0.94 ± 1 | 0.032 |
| Initial treatment strategy | |||
| Medical therapy | 32 (30%) | 62 (29%) | 0.96 |
| Percutaneous coronary intervention | 25 (24%) | 28 (13%) | 0.019 |
| Aortic valve replacement | 8 (7%) | 19 (9%) | 0.83 |
| Coronary artery bypass grafting and aortic valve replacement | 41 (39%) | 103 (49%) | 0.09 |
Clinical follow-up was available for all patients at a median time of 56 (31-60) months, and it is summarized in Table 3 and in Figure 1 . We found no difference in MACE up to 5 years, overall death, repeat revascularization, and nonfatal MI between the 2 groups. Similar results were found in the subgroup of patients with severe AS and are reported in Table 3 . When stratified by initial treatment strategy, MACE rate was still not significantly different in FFR- and angio-guided groups. Combined CABG and AVR was associated with the best clinical outcome and medical therapy with the worst, whereas intermediate outcomes were reported in patients initially treated with PCI or AVR only. AVR at follow-up was performed in 9% of patients in the FFR-guided group (time to AVR: 31 [22 to 43] months) versus 6% in the angio-guided group (time to AVR: 21 [7 to 54] months; p = 0.28). At the latest follow-up available, cumulative AVR rate was still lower in the FFR-guided group (59 [55%] vs 135 [63%]; p = 0.16).
