Post‐percutaneous coronary intervention (PCI) fractional flow reserve (FFR) ≥0.90 confers an improved cardiac prognosis. There are currently limited data available to determine how often it is possible to improve an angiographically acceptable but physiologically suboptimal result. A physiology‐guided optimization strategy can achieve a clinically meaningful increase in the proportion of patients achieving a final post‐PCI FFR ≥0.90 compared to standard care. Following angiographically successful PCI procedures, 260 patients will be randomized (1:1) to receive either a physiology‐guided incremental optimization strategy (intervention group) or blinded post‐PCI coronary physiology measurements (control group). Patients undergoing successful, standard‐of‐care PCI for either stable angina or non‐ST‐segment‐elevation myocardial infarction who meet the study's inclusion and exclusion criteria will be eligible for randomization. The primary endpoint is defined as the proportion of patients with a final post‐PCI FFR result ≥0.90. Secondary endpoints include change from baseline in Seattle Angina Questionnaire and EQ‐5D‐5L scores at 3 months and the rate of target vessel failure and its components (cardiac death, myocardial infarction, stent thrombosis, unplanned rehospitalization with target vessel revascularization) at 3 months and 1 year. 260 individual patients were successfully randomized between March 2018 and November 2019. Key baseline demographics of the study population are reported within. TARGET FFR is an investigator‐initiated, prospective, single‐center, randomized controlled trial of an FFR‐guided PCI optimization strategy. The study has completed recruitment and is now in clinical follow‐up. It is anticipated that primary results will be presented in Autumn 2020. ClinicalTrials.gov Identifier: NCT03259815. [Correction added on Apr 3 2020, after first online publication: Clinical Trials identifier added.]
coronary flow reserve
diastolic pressure ratio
fractional flow reserve
hyperemic tran‐stent gradient
instantaneous wave‐free ratio
index of microcirculatory resistance
major adverse cardiovascular events
non‐ST segment elevation myocardial infarction
optical coherence tomography
percutaneous coronary intervention
distal coronary pressure/aortic pressure
physiologically guided incremental optimization strategy
resting full‐cycle ratio
Seattle Angina Questionnaire
ST‐segment elevation myocardial infarction
mean hyperemic transit time
mean resting transit time
The utility of fractional flow reserve (FFR) for assessing the physiological significance of coronary stenoses and the benefits of FFR‐guided decision making prior to percutaneous coronary intervention (PCI) have been well‐established in randomized clinical trials.11, 22 However, FFR is rarely used to assess the final PCI result where standard practice continues to be angiographic assessment alone. From a registry of 750 patients receiving bare metal stents, Pijls et al reported that at 6‐month follow‐up, the lowest event rates occurred in patients with post‐PCI FFR ≥0.90.33 Studies and related meta‐analyses involving drug‐eluting stents have suggested similar cutoff values of post‐PCI FFR to predict improved clinical outcomes.44, 55, 66, 77, 88, 99, 1010, 1111, 1212 These range from 0.89 to ≥0.92 with a large systematic review and meta‐analysis of 7470 patients concluding that a post‐PCI FFR ≥0.90 was associated with a significantly lower risk of repeat PCI and major adverse cardiovascular events.1010 Johnson et al correlated the immediate post‐PCI FFR results from 966 patients with clinical outcomes out to 3 years and demonstrated a significant, inverse relationship between post‐PCI FFR and subsequent clinical events.1313
Published values for overall mean or median final post‐PCI FFR results range from 0.84 to 0.95.33, 55, 66, 77, 99, 1212, 1414, 1515, 1616, 1717, 1818, 1919, 2020, 2121, 2222, 2323, 2424, 2525, 2626, 2727, 2828, 2929, 3030, 3131, 3232, 3333, 3434, 3535, 3636, 3737 The proportion of patients actually achieving a final FFR ≥0.90 varies widely between studies and ranges from 37% to 93%.33, 44, 55, 66, 77, 88, 1616, 1818, 2929, 3838, 3939 Of perhaps greater concern however, is the incidence of suboptimal FFR results after stenting. Where reported, the proportion of patients with post‐PCI FFR values ≤0.80 ranges from 4% to 20%.99, 1111, 2424, 2828, 3434, 3737, 4040 This indicates that, despite angiographically satisfactory results, as many as one in five patients may have a post‐PCI FFR result that remains below the threshold for performing revascularization in the first place. With up to 38% of patients still reporting angina 1 year after PCI procedures,4141 it seems plausible that persistently abnormal post‐PCI FFR results may be associated with symptom recurrence.
It has been suggested that non‐hyperemic pressure ratios (NHPRs), such as the instantaneous wave‐free ratio (iFR), also have potential to be used as objective measures of improvement in physiology following PCI.2525 A recent study employing blinded post‐PCI iFR assessments reported residual ischemia (iFR <0.90) in nearly one in four patients despite angiographically successful stenting results. The authors concluded that 81.6% of these cases were due to inapparent focal lesions potentially amenable to treatment with additional PCI.4242 While an NHPR‐guided PCI optimization strategy might be more appealing to clinicians as it could facilitate multiple physiological assessments without the need to repeatedly induce hyperemia, data on the prognostic value of post‐PCI NHPR values are currently lacking. The original NHPR, the ratio of distal coronary to aortic pressure (Pd/Pa) is routinely available with all diagnostic guidewires. Recently, two new resting physiology indices have been developed which have diagnostic equivalence to iFR: the diastolic pressure ratio (dPR) and the resting full‐cycle ratio (RFR).4343, 4444
One of the reasons clinicians do not routinely measure post‐PCI FFR is that there are currently limited data available to determine how often it is possible to improve an angiographically acceptable but physiologically suboptimal result.
Given the potential prognostic significance of post‐PCI FFR, but general lack of adoption, there is a clear need for randomized controlled trials to:
TARGET FFR is an investigator‐initiated, prospective, single‐center, randomized controlled trial of an FFR‐guided PCI optimization strategy. The primary hypothesis is that this strategy will result in a clinically meaningful increase in the proportion of patients achieving a final post‐PCI FFR ≥0.90 compared to standard care. Following angiographically successful PCI procedures, 260 patients will be randomized (1:1) to receive either a physiology‐guided incremental optimization strategy (PIOS intervention group) or blinded post‐PCI coronary physiology measurements (control group). Patients undergoing successful, standard‐of‐care PCI for either stable angina or Non‐ST segment elevation myocardial infarction (NSTEMI) who meet the study's inclusion and exclusion criteria (Table 1) will be eligible for randomization. All patients will complete the Seattle Angina Questionnaire (SAQ‐7) and EQ‐5D‐5L questionnaire both prior to, and 3 months after, their procedure. Longer term outcomes will be assessed using record linkage. The study flowchart is depicted in Figure 1.
The primary endpoint is defined as the proportion of patients with a final post‐PCI FFR result ≥0.90.
Secondary endpoints include:
A complete list of secondary outcome measures is provided in Table 2.
|Patient‐reported outcome measures|
|3‐month and 1‐year follow‐up|
|Additional coronary physiology outcomes|
PCI procedures will be performed according to standard clinical practice. Treatment decisions (including the use of adjunctive intracoronary imaging such as OCT or IVUS) and the definition of an angiographically acceptable PCI result will be left entirely to the operator's judgment. The study protocol mandates that all enrolled patients should have pre‐PCI coronary physiology assessments performed, however, the decision whether to then use the pressure wire for PCI or employ an alternative angioplasty wire is at the operator's discretion. Decisions on choice and duration of antiplatelet medications and/or combination with anticoagulant therapy will also be left to the operator's clinical judgment. Patients will only be randomized and post‐PCI study measurements/interventions performed after the operator has declared the standard care procedure to be complete.
If post‐PCI FFR is <0.90, the coronary physiology results and hyperemic pullback assessment will be disclosed to the operator. Based on the interpretation of the pullback recording, the operator will attempt to obtain the target optimal post‐PCI FFR result by following the steps of the PIOS algorithm outlined below:
In keeping with standard care, the operator will determine procedural completeness and success based on the angiographic and/or intracoronary imaging appearances. Post‐PCI coronary physiology measurements will be performed but not disclosed to the operator. No further intervention or optimization measures will be undertaken.
Follow‐up will be performed by clinical research nurses blinded to both the assigned treatment arm and the final FFR result. All patients will be contacted by letter and/or telephone 3 months after their PCI procedure to repeat the SAQ‐7 and EQ‐5D‐5L questionnaires. In cases where an adverse event or clinical endpoint has occurred, additional information will be retrieved from the patient's electronic health record or general practitioner. Clinical outcomes will be reviewed again at 1 year using electronic health record linkage.
Pilot data from 50 patients who underwent post‐PCI FFR assessment at our institution revealed that an initial post‐PCI FFR result ≥0.90 was achieved in only 16/50 (32%). Additional optimization measures (further postdilation, stenting or both) were attempted in just nine patients (18%). Initial post‐PCI FFR increased from a median of 0.83 (0.80‐0.86) to a final FFR of 0.88 (0.86‐0.89) in these patients. The results of pullback measurements performed in all 50 vessels, however, revealed at least one target for additional optimization measures was present in up to 40% of patients (HTG ≥0.05 in 19/50 (38%); focal step‐up ≥0.05 either proximal or distal to the stented segment in 21/50 (42%). It is hypothesized that systematically applying the PIOS intervention will result in a 20% absolute increase in the proportion of patients with a final post‐PCI FFR ≥0.90 compared to the control group. A sample size of 130 patients per group would be required to have 90% power to detect a 20% difference between groups at the 5% significance level; therefore, 260 patients will be randomized. Patients presenting with stable angina or NSTEMI who attend our institution for diagnostic coronary angiography proceed to PCI during the same procedure in approximately 40% of cases. It is therefore estimated that approximately 650 patients will be enrolled in the study in order to randomize 260 patients following their standard‐of‐care PCI.
The study data will be summarized for the randomized population overall, and by randomized group. The number of observations and number of missing values will be reported. Continuous variables will have normality tests applied and be summarized using the mean ± SD or median and interquartile range according to their distribution. Differences in continuous variables between randomized groups will be assessed using Student's t tests or Mann‐Whitney U tests as appropriate. The Pearson correlation coefficient will be applied to parametric variables while correlation between nonparametric variables will be assessed using Spearman's rank correlation. Categorical variables will be summarized with frequencies and percentages. Differences in categorical variables between randomized groups will be evaluated using Chi‐square tests or Fisher's exact tests.
Where relevant, changes from baseline will be summarized. Multivariate logistic regression analyses will be employed to assess for clinical predictors of post‐PCI FFR values ≥0.90 and ≤0.80. The primary outcome will be summarized in the full analysis set as a whole and by treatment group. A test and 95% CI for two proportions (adjusted Wald method) will be employed, together with Fisher's exact test. Additional secondary analyses on this outcome will use logistic regression to investigate whether any of the baseline characteristics affect the outcome. This will be performed by first investigating each characteristic on its own (together with the treatment group). Any variables that are significant here will be added to build a larger model, bearing in mind sample size limitations. For the binary categorical secondary outcomes, the same analysis approach will be used as with the primary outcome. For quantitative secondary outcomes, two sample t tests or Mann Whitney U tests will be used as appropriate, as well as further analyses using regression to investigate whether any of the baseline characteristics affect the outcome. All tests will be two sided and a P ‐value of <.05 will be considered significant. Efficacy analyses will be carried out according to the intention to treat principle, that is, in relation to randomized treatment allocation, rather than treatment received. Safety analyses will be carried out in relation to treatment received. Details of subgroup and additional analyses are provided in the Supplementary Appendix.
The study received ethical approval from the West of Scotland Research Ethics Service (17/WS/0153) and will be conducted in accordance with the International Conference on Harmonization Good Clinical Practice (ICH GCP) Guideline and the Declaration of Helsinki (64th World Medical Association General Assembly, Fortaleza, Brazil, October 2013). The study sponsor is the NHS National Waiting Times Centre Board (Golden Jubilee National Hospital). Details of the Trial Steering Committee, Clinical Endpoints Committee, and Physiology Core Lab are provided in the Supplementary Appendix.
Between March 8, 2018 and November 22, 2019, we successfully randomized 260 patients meeting the trial's inclusion and exclusion criteria. Preliminary baseline characteristics of the study population are presented in Table 3. The trial is now in clinical follow‐up and it is anticipated the primary results will be presented in Autumn 2020.
|Total (n = 260)||PIOS (n = 131)||Control (n = 129)|
|Male||226 (86.9%)||117 (89.3%)||109 (84.5%)|
|Age||59 (54‐66)||58 (54‐66)||60 (55‐68)|
|BMI||29 (27‐32)||29 (26‐32)||29 (27‐32)|
|Hypertension||116 (44.6%)||58 (44.3%)||58 (45%)|
|Hypercholesterolemia||146 (56.2%)||72 (55%)||74 (57.4%)|
|Diabetes||49 (18.8%)||24 (18.3%)||25 (19.4%)|
|OHAs||42 (85.7%)||21 (87.5%)||21 (84%)|
|Insulin||5 (10.2%)||3 (12.5%)||2 (8%)|
|Atrial fibrillation||19 (7.3%)||10 (7.6%)||9 (7%)|
|OAC||13 (68.4%)||6 (60%)||7 (77.8%)|
|2||6 (31.6%)||3 (15.8%)||3 (15.8%)|
|3||4 (21.1%)||3 (30%)||1 (11.1%)|
|4||4 (21.1%)||2 (20%)||2 (22.2%)|
|5||4 (21.1%)||2 (20%)||2 (22.2%)|
|6||1 (5.3%)||0||1 (11.1%)|
|Previous TIA/stroke||17 (6.5%)||8 (6.1%)||9 (7%)|
|CKDa||5 (1.9%)||3 (2.3%)||2 (1.6%)|
|Family history of CAD||172 (66.2%)||88 (67.2%)||84 (65.1%)|
|History of smoking||183 (70.4%)||92 (70.2%)||91 (70.5%)|
|Current||50 (27.3%)||28 (30.4%)||22 (24.2%)|
|Within past year||41 (22.4%)||22 (23.9%)||19 (20.9%)|
|Ex‐smoker >1 y||92 (50.3%)||42 (45.7%)||50 (54.9%)|
|Thyroid dysfunction||20 (7.7%)||9 (6.9%)||11 (8.5%)|
|Heart failure||44 (16.9%)||28 (21.4%)||16 (12.4%)|
|NYHA class 1||29 (65.9%)||19 (67.9%)||10 (62.5%)|
|NYHA class 2||15 (34.1%)||9 (32.1%)||6 (37.5%)|
|HFrEF||43 (97.7%)||28 (100%)||15 (93.8%)|
|Previous MI||95 (36.5%)||50 (38.2%)||45 (34.9%)|
|Previous PCI||100 (38.5%)||54 (41.2%)||46 (35.7%)|
|Previous CABG||1 (0.4%)||1 (0.8%)||0|
|Valvular heart disease||8 (3.1%)||2 (1.5%)||6 (4.7%)|
|Aortic stenosis||6 (2.3%)||1 (0.8%)||5 (3.9%)|
|Mitral regurgitation||2 (0.8%)||1 (0.8%)||1 (0.8%)|
|Angina||215 (82.7%)||107 (81.7%)||108 (83.7%)|
|CCS class 1||58 (27%)||28 (26.2%)||30 (27.8%)|
|CCS class 2||101 (47%)||51 (47.7%)||50 (46.3%)|
|CCS class 3||55 (25.6%)||27 (25.2%)||28 (25.9%)|
|CCS class 4||1 (0.5%)||1 (0.9%)||0|
|Single APT||253 (97.3%)||128 (97.7%)||125 (96.9%)|
|Dual APT||185 (71.2%)||97 (74.1%)||88 (68.2%)|
|OAC||16 (6.2%)||8 (6.1%)||8 (6.2%)|
|Statin||250 (96.2%)||127 (97%)||123 (95.4%)|
|Beta blocker||237 (91.2%)||121 (92.4%)||116 (89.9%)|
|CCB||52 (20%)||22 (16.8%)||30 (23.3%)|
|ACEI||175 (67.3%)||91 (69.5%)||84 (65.1%)|
|ARB||23 (8.9%)||11 (8.4%)||12 (9.3%)|
|Diuretic||30 (11.5%)||13 (9.9%)||17 (13.2%)|
|GTN spray||123 (47.3%)||61(46.6%)||62 (48.1%)|
|Used daily||30 (24.4%)||13 (21.3%)||17 (27.4%)|
|Used weekly||67 (54.55)||34 (55.7%)||32 (51.6%)|
|Used monthly||27 (22%)||14 (23%)||13 (21%)|
|Oral nitrate||69 (26.5%)||26 (19.9%)||43 (33.3%)|
|Nicorandil||22 (8.5%)||14 (10.7%)||8 (6.2%)|
|Ivabradine||5 (1.9%)||3 (2.3%)||2 (1.6%)|
|No. anti‐anginal meds|
|0||9 (3.5%)||4 (3.1%)||5 (3.9%)|
|1||99 (38.1%)||55 (42%)||44 (34.1%)|
|2||114 (43.8%)||55 (42%)||59 (45.7%)|
|3||31 (11.9%)||13 (9.9%)||18 (14%)|
|4||7 (2.7%)||4 (3.1%)||3 (2.3%)|
|Stable angina||88 (33.9%)||40 (30.5%)||48 (37.2%)|
|Staged PCI||16 (18.2%)||8 (20%)||8 (16.7%)|
|ACS‐NSTEMI||101 (38.8%)||50 (38.2%)||51 (39.5%)|
|Days post‐MI||21 (12‐28.5)||20 (7‐26.3)||23 (16‐31)|
|ACS‐unstable angina||3 (1.2%)||2 (1.5%)||1 (0.8%)|
|Staged PCI/completion of revascularization||68 (26.2%)||39 (29.8%)||29 (22.5%)|
|Post‐STEMI||46 (67.7%)||29 (74.4%)||17 (58.6%)|
|Days since MI||68.8±29.5||70.4±30.9||66.1±27.6|
|Post‐NSTEMI||22 (32.4%)||10 (25.6%)||12 (41.4%)|
|Days since MI||67 (54‐98)||64 (54‐86.8)||79.5 (53.3‐110.8)|
|LAD||149 (57.3%)||75 (57.3%)||74 (57.4%)|
|RCA||67 (25.8%)||28 (21.4%)||39 (30.2%)|
|LCx||33 (12.7%)||20 (15.3%)||13 (10.1%)|
|OM||10 (3.8%)||8 (6%)||2 (1.6%)|
|Diagonal||1 (0.4%)||0||1 (0.8%)|
Prior studies of post‐PCI FFR have recruited patients with either stable angina or recent NSTEMI. However, while in general enrolling heterogeneous populations enhances external validity, in this case, the performance of the diagnostic test may differ according to the nature of the coronary artery disease. By measuring CFR and IMR with FFR we aim to take account of microvascular dysfunction to help inform the interpretation of the primary and secondary physiology outcomes. The sample size calculation is based on pilot data which identified potential targets for additional intervention following an index PCI procedure. The indication for the PIOS intervention may be more limited if a high proportion of patients have a post‐PCI FFR ≥0.90 or if there is a higher incidence of the diffuse gradient disease pattern in the population than predicted. This could result in the study being underpowered. Currently, there are limited data on the potential to further optimize the physiological result of a PCI procedure. Agarwal et al reported that 137 of the 664 lesions (20.6%) in their patient cohort underwent additional PCI based on the presence of a persistently ischemic, or if not ischemic, “unsatisfactory” (as determined by the operator) initial post‐PCI FFR.77 The mean initial FFR for these lesions was 0.78 ± 0.07. 58/137 lesions (42%) received further postdilation of the implanted stent with a bigger balloon size and higher pressure and duration of inflation. 45/137 (33%) had another stent implanted while 24/137 (18%) underwent both additional stenting and balloon postdilation. These subsequent interventions led to an improvement in FFR in this subgroup from 0.78 ± 0.07 to 0.87 ± 0.05. Overall, suboptimal initial post‐PCI FFR prompting subsequent intervention led to an increase in lesions with final FFR >0.91 from 34% to 43% (≥0.86 from 60% to 74%) and decreased persistently ischemic lesions (≤0.81) from 21% to 9%. Of note, however, only 74/118 (63%) of those with initial post‐PCI FFR ≤0.81 actually had further intervention attempted. In a cohort of 13 patients who fulfilled both functional and OCT‐defined criteria for suboptimal stent results, Wolfrum et al increased the mean post‐PCI FFR in this group from 0.80 ± 0.02 to 0.88 ± 0.01 through a combination of additional stent postdilation (46%), additional stenting (39%) or a combination of both strategies (15%). Larger increases in FFR value were observed in the seven patients who received additional stents.4545
The proposed target for an “optimal” post‐PCI FFR result stems from a meta‐analysis which included several early studies that assessed post‐PCI FFR using relatively small bolus doses of intracoronary adenosine to induce hyperemia. Smaller doses of adenosine may not have achieved maximal hyperemia and could potentially have overestimated the final FFR results. As such, a post‐PCI FFR value of ≥0.90 may not be realistic target, particularly in the left anterior descending artery which typically subtends a larger myocardial mass. It has in fact been suggested that optimal cutoff values of post‐PCI FFR are different according to the target vessels involved.4646 Consequently, the PIOS intervention may not be effective at increasing the FFR value to such an extent, or at least, not in the proportions desired.
There are potential concerns regarding both performance and confirmation bias with this trial design and as such the following steps were taken to minimize their effects.
By its very nature, the PIOS group potentially receives more focused attention from the interventionalist than the control group. Despite the operator's best efforts, this does not necessarily translate into higher final FFR values (the primary endpoint), however, and that is the question the trial seeks to answer—does routinely applying a post‐PCI physiology‐guided optimization strategy actually achieve a clinically meaningful difference in the proportion of patients with optimal final FFR results?
Following PCI procedures, the treating interventionalists reassured all patients that they received the highest standard of care, regardless of their randomization group. Specific results of the final physiology measurements were not disclosed to patients in either group.
In an effort to mitigate the potential Hawthorne (“observer”) effect, the study could have on local PCI practices, other than cases randomized to the PIOS intervention where post‐PCI FFR was <0.90, operators were blinded to all post‐PCI physiology results, 15 different interventional cardiologists were enlisted to participate in the trial in order to: (a) assess a wider variety of practice and (b) dilute the exposure any one physician had to unblinded post‐PCI physiology results (on average each operator would have performed less than 10 unblinded PIOS cases over a 20‐month period). Further details on blinding procedures are contained in the Supplementary Appendix.
Furthermore, we posit that rather than being subject to confirmation bias, the primary results of the trial will actually challenge such bias regarding the functional results of PCI procedures.
TARGET FFR is an investigator‐initiated, prospective, single‐center, randomized controlled trial to determine the feasibility and efficacy of using a coronary physiology‐guided optimization strategy to achieve final post‐PCI FFR results ≥0.90.
Damien Collison has received speaker fees from Abbott. Colin Berry is employed by the University of Glasgow which holds consultancy and research agreements with Abbott, AstraZeneca, Boehringer Ingelheim, GSK, HeartFlow, Menarini, Novartis, Ospens, and Siemens Healthcare. Keith G. Oldroyd has received consultant and speaker fees from Abbott and Boston Scientific. John D. McClure has no conflicts of interest to disclose.
This work was supported by NHS National Waiting Times Centre Board endowment funds. Colin Berry has received grant support from the British Heart Foundation (RE/18/6134217; FS/14/15/30661; FS172632744; PG‐17‐25‐32884).