Journal of the American College of Cardiology Vol. 63, No. 13, 2014� 2014 by the American College of Cardiology Foundation ISSN 0735-1097/$36.00Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jacc.2013.09.060

CLINICAL RESEARCH Coronary Artery Disease

Multicenter Core Laboratory Comparisonof the Instantaneous Wave-Free Ratioand Resting Pd /Pa With Fractional Flow Reserve

The RESOLVE Study

Allen Jeremias, MD, MSC,*y Akiko Maehara, MD,yz Philippe Généreux, MD,yzxKaleab N. Asrress, MA, BM BCH,k Colin Berry, MBCHB, PHD,{ Bernard De Bruyne, MD,#

Justin E. Davies, MBBA,** Javier Escaned, MD,yy William F. Fearon, MD,zz K. Lance Gould, MD,xxNils P. Johnson, MD, MS,xx Ajay J. Kirtane, MD, SM,yz Bon-Kwon Koo, MD,kkKoen M. Marques, MD, PHD,{{ Sukhjinder Nijjer, MBBA,** Keith G. Oldroyd, MBCHB, MD,{Ricardo Petraco, MD,** Jan J. Piek, MD,## Nico H. Pijls, MD,*** Simon Redwood, MD,kMaria Siebes, PHD,## Jos A. E. Spaan, PHD,## Marcel van ’t Veer, MSC, PHD,***

Gary S. Mintz, MD,yz Gregg W. Stone, MDyzStony Brook and New York, New York; Montreal, Quebec, Canada; London, United Kingdom;

Glasgow, Scotland, United Kingdom; Aalst, Belgium; Madrid, Spain; Stanford, California;

Houston, Texas; Seoul, Republic of Korea; and Amsterdam and Eindhoven, the Netherlands

From the *

Center, Ston

New York;

York, New

Montreal, Q

Centre of R

don, United

Jubilee Nati

Objectives T

Division of Cardiovascula

y Brook, New York; yCzDepartment of Medicin

York; xDepartment of M

uebec, Canada; kCardiesearch Excellence, King’

Kingdom; {West of Scot

onal Hospital, Glasgow,

his study sought to examine the diagnostic accuracy of the instantaneous wave-free ratio (iFR) and resting distalcoronary artery pressure/aortic pressure (Pd/Pa) with respect to hyperemic fractional flow reserve (FFR) in a corelaboratory–based multicenter collaborative study.

Background F

FR is an index of the severity of coronary stenosis that has been clinically validated in 3 prospective randomizedtrials. iFR and Pd/Pa are nonhyperemic pressure-derived indices of the severity of stenosis with discordant reportsregarding their accuracy with respect to FFR.

Methods iF

R, resting Pd/Pa, and FFR were measured in 1,768 patients from 15 clinical sites. An independent physiology corelaboratory performed blinded off-line analysis of all raw data. The primary objectives were to determine specific iFRand Pd/Pa thresholds with �90% accuracy in predicting ischemic versus nonischemic FFR (on the basis of an FFRcut point of 0.80) and the proportion of patients falling beyond those thresholds.

Results O

f 1,974 submitted lesions, 381 (19.3%) were excluded because of suboptimal acquisition, leaving 1,593 for finalanalysis. On receiver-operating characteristic analysis, the optimal iFR cut point for FFR �0.80 was 0.90 (C statistic:0.81 [95% confidence interval: 0.79 to 0.83]; overall accuracy: 80.4%) and for Pd/Pa was 0.92 (C statistic: 0.82[95% confidence interval: 0.80 to 0.84]; overall accuracy: 81.5%), with no significant difference between theseresting measures. iFR and Pd/Pa had �90% accuracy to predict a positive or negative FFR in 64.9% (62.6% to67.3%) and 48.3% (45.6% to 50.5%) of lesions, respectively.

Conclusions T

his comprehensive core laboratory analysis comparing iFR and Pd/Pa with FFR demonstrated an overall accuracy ofw80% for both nonhyperemic indices, which can be improved to �90% in a subset of lesions. Clinical outcomestudies are required to determine whether the use of iFR or Pd/Pa might obviate the need for hyperemia in selectedpatients. (J Am Coll Cardiol 2014;63:1253–61) ª 2014 by the American College of Cardiology Foundation

r Medicine, Stony Brook University Medical

ardiovascular Research Foundation, New York,

e, Columbia University Medical Center, New

edicine, Hôpital du Sacré-Coeur de Montréal,

ovascular Division, British Heart Foundation

s College London, St. Thomas’ Hospital, Lon-

land Regional Heart and Lung Centre, Golden

Scotland, United Kingdom; #Cardiovascular

Center Aalst, OLV Clinic, Aalst, Belgium; **International Centre for Circulatory

Health, National Heart and Lung Institute, Imperial College London, London, United

Kingdom; yyHospital Clinico San Carlos, Madrid, Spain; zzDivision of Cardiovascular

Medicine, Stanford University Medical Center, Stanford, California; xxWeatherhead

PET Center for Preventing and Reversing Atherosclerosis, Division of Cardiology,

Department of Medicine, University of Texas Medical School andMemorial Hermann

Hospital, Houston, Texas; kkSeoul National University Hospital, Seoul, Republic of

Korea; {{Department of Cardiology, VU University Medical Center, Amsterdam, the

Netherlands; ##Departments of Cardi

Academic Medical Center, University

and the ***Department of Cardiol

Netherlands. Dr. Jeremias has served

Bureau for Volcano Corp. Dr. Berr

a research grant from St. Jude Medi

consulting fees from St. Jude Medica

and served as a consultant with lice

Dr. Escaned has served as a member o

Volcano Corp. Dr. Fearon has rece

Dr. Gould holds a nonfinancial, m

Corp. and is a 510(k) applicant for

absolute flow using cardiac PET; a

scholarship fund and the Universit

agreement with Positron Corporation

for royalties; however, Dr. Gould reta

Abbreviationsand Acronyms

CI = confidence interval

ECG = electrocardiographic

FFR = fractional flow reserve

(hyperemic by definition)

iFR = instantaneous wave-

free ratio (nonhyperemic)

LAD = left anterior

descending

NPV = negative predictive

value

Pd/Pa = distal coronary

artery pressure/aortic

pressure (nonhyperemic)

PPV = positive predictive

value

ROC = receiver-operating

characteristic

Jeremias et al. JACC Vol. 63, No. 13, 2014Instantaneous Wave-Free Ratio and Pd/Pa Compared to FFR April 8, 2014:1253–61

1254

Fractional flow reserve (FFR) isan index of the hemodynamicsignificance of a coronary ste-nosis that is calculated directlyfrom measurements of hyper-emic pressure (1,2). The physi-ological basis of FFR has beenextensively validated in animaland human studies, and FFRshows good correlation to non-invasive ischemia testing withperfusion scintigraphy (3) and po-sitron emission tomography (4).

See page 1262

FFR has been shown in 3 ran-domized trials to identify coronarystenoses that will benefit from early

revascularization (those with a positive FFR) (5) and converselythose lesions with a negative FFR for which revascularizationmay be safely deferred (6,7). To measure FFR, a vasodilator(most commonly intravenous or intracoronary adenosine) isadministered to minimize microvascular resistance and theeffect of resting hemodynamics such that coronary pressurebecomes proportional to myocardial flow.

Interest has recently emerged as to whether 2 nonhyperemicmeasures of pressure might be useful to assess the severity ofcoronary stenosis. Pd/Pa is the ratio of distal coronary arterypressure to aortic pressure over the entire cardiac cycle.Conversely, the instantaneous wave-free ratio (iFR) measurescoronary pressure during a specific period of diastole whenresting resistance is the lowest (8). By reducing procedural timeand cost, avoiding patient-related discomfort from pharmaco-logical hyperemia, and allowing continuous online mea-surements (thereby facilitating multivessel interrogation),assessment of the severity of coronary stenosis without induc-tion of hyperemia is intuitively appealing, provided diagnosticaccuracy is preserved. However, in prior reports, the diag-nostic accuracy of iFR compared with FFR has rangedwidely from 60% to 91% (8–11), and its relative accuracy

ology and Biomedical Engineering and Physics,

of Amsterdam, Amsterdam, the Netherlands;

ogy, Catharina Ziekenhuis, Eindhoven, the

as a consultant and member of the Speakers’

y has served as a consultant for and received

cal. Dr. De Bruyne has received institutional

l. Dr. Davies has received study support from

nsed intellectual property for Volcano Corp.

f the Speakers’ Bureau for St. Jude Medical and

ived research support from St. Jude Medical.

utual nondisclosure agreement with Volcano

cfrQuant, a software package for quantifying

ll royalties will go to a University of Texas

y of Texas has a commercial, nonexclusive

to distribute and market cfrQuant in exchange

ins the ability to distribute cost-free versions to

compared with Pd/Pa has been debated. Previous compara-tive studies to date have been limited by different studymethodologies, modest sample sizes, and the use of differentalgorithms to calculate iFR. Given these conflicting reports,we formed a collaborative group of investigators to performa large-scale, physiology core laboratory–based analysis withstandardized methods to compare the diagnostic accuracyof iFR and Pd/Pa with respect to FFR as the referencestandard and to determine the proportion of patients inwhom the accuracy of iFR and Pd/Pa is at least 90%.

Methods

Patient population and study inclusion criteria. Thepresent investigation was an international, multicenter,nonrandomized, retrospective, core laboratory–based anal-ysis in patients with coronary artery disease undergoingphysiological lesion assessment by FFR, iFR, and Pd/Pa. Theprincipal investigators representing all of the publishediFR/FFR comparative studies agreed to collaboratively par-ticipate in this effort, including the ADVISE (ADenosineVasodilator Independent Stenosis Evaluation) study andregistry (8,11), VERIFY (VERification of Instantane-ous Wave-Free Ratio and Fractional Flow Reserve forthe Assessment of Coronary Artery Stenosis Severity inEverydaY Practice) (9), and Johnson et al. (10). In addition,6 other study sites contributed unpublished data to theanalysis. All studies included in this analysis were appro-ved by the institutional review boards of the individual sites.Original raw phasic pressure waveforms from each patientwere submitted digitally to the Physiology Core Laboratoryat the Cardiovascular Research Foundation (New York,New York) for independent off-line analysis. In addition,selected baseline patient demographic and procedural datawere supplied to the core laboratory. This study was aninvestigator-sponsored study by theCardiovascular ResearchFoundation and was supported by funding from VolcanoCorp. (San Diego, California). The funding source wasuninvolved with the design of the protocol and the analysisand interpretation of the study results.

Patients with stable angina, unstable angina, or non–ST-segment elevation myocardial infarction undergoing

selected collaborators for research. Dr. Johnson holds a nonfinancial, mutual

nondisclosure agreement with Volcano Corp. Dr. Koo has received honorarium and

a research grant from St. Jude Medical. Dr. Oldroyd has served as a member of the

Speaker’s Bureau for St. Jude Medical and Volcano Corp. Dr. Piek has served as

a consultant for MAB Abbott Vascular and Miracor. Dr. Pijls has served as

a consultant for St. Jude Medical; received institutional research grants from St. Jude

Medical; and served as an advisory board member for Heart Flow. Dr. Mintz has

served as a consultant and received grant support from Volcano Corp. Dr. Stone has

served as a consultant for Volcano Corp., InfraReDx, and Boston Scientific. This

study was an investigator-sponsored study by the Cardiovascular Research Foun-

dation and was supported by funding from Volcano Corp. (San Diego, California).

The funding source was uninvolved with the design of the protocol and the analysis

and interpretation of the study results. All other authors have reported that they have

no relationships relevant to the content of this paper to disclose.

Manuscript received July 8, 2013; revised manuscript received September 22, 2013,

accepted September 22, 2013.

JACC Vol. 63, No. 13, 2014 Jeremias et al.April 8, 2014:1253–61 Instantaneous Wave-Free Ratio and Pd/Pa Compared to FFR

1255

coronary angiography with or without percutaneous coro-nary intervention in whom FFR of a single stenosis ina major epicardial coronary artery was performed during theprocedure were considered for inclusion in the study. Two ormore lesions could be present in a single patient if indifferent epicardial vessels. Exclusion criteria included leftmain disease, heart failure as defined by New York HeartAssociation class III or IV, respiratory failure requiringintubation or supplementary oxygen, cardiogenic shock,significant arrhythmia precluding waveform analysis (e.g.,excessive premature ventricular contractions or atrial fibril-lation), and tachycardia with a heart rate >120 beats/min.Pressure measurements and analysis. Physiological mea-surements of coronary stenoses were performed according toexisting study protocols. The RadiAnalyzer Xpress instru-ment with the Certus coronary pressure wire (St. JudeMedical, Uppsala, Sweden), the Volcano s5 imaging systemwith the PrimeWire (Volcano Corp., Rancho Cordova,California), or earlier-generation equipment from thesemanufacturers was used for measurements of coronarypressure. After the pressure sensor was zeroed and equalizedto aortic pressure, it was positioned at least 5 mm distal tothe stenosis and a recording of the baseline distal coronaryand aortic pressures was obtained. After the administrationof intracoronary nitroglycerine as per the operators’ discre-tion, hyperemia was induced by the administration of eitherintravenous adenosine at a dosage of 140 mg/kg/min orintracoronary adenosine at various doses and FFR wascalculated. All pressure tracings were submitted directly tothe Cardiovascular Research Foundation physiology corelaboratory for analysis.

FFR is the ratio of mean distal coronary pressure (Pd) tomean aortic pressure (Pa) during maximum hyperemia. ThePd signal is obtained from a guidewire with a piezoresistivepressure transducer, and the Pa signal is obtained froma fluid-filled guiding or diagnostic catheter. FFR is taken asthe lowest stable value of the Pd/Pa ratio during maximalhyperemia. To ensure accuracy of the analysis, waveformanalysis of all pressure tracings was performed to confirm thatnone of the following exclusion criteria were present: signif-icant arrhythmia that may preclude appropriate waveformanalysis, loss of Pa or Pd pressure signal at any point duringthe run apart from intracoronary vasodilator administration,inappropriate recording of Pa or Pd (e.g., only a flat signal ispresent at some point during the recording), dampened Pa

or Pd waveform, reversed gradient during hyperemia (i.e.,Pd pressure signal elevated above Pa, resulting in an FFR>1.00), or sensor drift defined as FFR �0.97 or �1.03 afterpullback of the pressure wire transducer into the guidingcatheter. In addition to the waveform analysis, the FFRrecording had to have an adequate baseline tracing beforeadministration of adenosine. Specifically, a minimum of 5waveforms of uninterrupted recording adequate for analysiswithout significant artifact of the tracing was required. FFRwas calculated independently from the original readout as thelowest artifact-free Pd/Pa during maximal hyperemia.

iFR is the ratio of Pd/Pa measured during a pre-specifiedperiod in mid to late diastole of the cardiac cycle withouthyperemia (8). The onset of diastole was identified from thedicrotic notch, and the diastolic window was calculatedbeginning 25% into diastole and ending 5 ms before enddiastole. iFR was calculated off-line in the core laboratoryusing the Volcano Harvest software package, which containsthe iFR computational algorithm developed at the ImperialCollege of London (8). All analyses were performed ina fully automated manner, eliminating the need for manualselection of data time points. This automated analysis isbased on a synchronized electrocardiographic (ECG) signalto determine the appropriate diastolic intervals for pressuremeasurements. If the ECG signal was missing, the corelaboratory manually inserted R-wave markings based on thepressure waveform into the baseline tracing from which iFRwas calculated.

Resting Pd/Pa was calculated in similar fashion to iFRexcept that Pd/Pa was time averaged over the entire cardiaccycle, thus including both systole and diastole. In additionto the exclusion criteria for measurement of FFR, iFRand Pd/Pa recordings with any of the following character-istics were excluded from the analysis: insufficient baselinerecording before administration of adenosine (recording hadto contain at least 5 cardiac cycles from the start of therecording to the onset of hyperemia), significant arrhythmiasincluding supraventricular tachycardia or premature ventri-cular contractions within the baseline tracing, or heart rate<50 or >120 beats/min.

Core laboratory analyses were performed in a blindedfashion at 3 separate workstations by different technicians insequential, independent phases. First, a thorough waveformanalysis was performed of all baseline and hyperemic trac-ings, and pressure recordings meeting any of the previouslyoutlined exclusion criteria were removed from the analysis.Second, an independent calculation of FFR was performedblinded to the original FFR readout. Third, fully auto-mated, computerized calculations of Pd/Pa and iFR wereperformed by a physician unaware of the waveform analysisand computation of FFR. All tracings were over-read bya physician experienced in physiology measurements (A.M.,P.G., or A.J.) to ensure data quality. FFR, iFR and Pd/Pa

data were recorded on separate case report forms that werenot merged until the completion of the blinded analyses.Study endpoints. The primary objective of this study wasto evaluate the level of diagnostic accuracy of iFR and Pd/Pa

compared with FFR in a variety of clinical settings in thelargest population studied to date using rigorous, pre-specified core laboratory–based processes. Using FFR asthe reference standard, the primary endpoint of the studywas to identify the iFR thresholds that most strongly cor-related with an FFR cut point of 0.80 and to determine theproportion of lesions for which these thresholds applied.Thresholds with �90% diagnostic accuracy were calcula-ted (pre-specified as representing the minimal thresholdsrequired for the potential clinical utility of iFR), and the

Jeremias et al. JACC Vol. 63, No. 13, 2014Instantaneous Wave-Free Ratio and Pd/Pa Compared to FFR April 8, 2014:1253–61

1256

proportions of lesions that fell beyond those thresholds weredetermined (defined as the adenosine-free zone).

Secondary study objectives included determining theiFR thresholds necessary to achieve >90% to 99% diag-nostic accuracy, construction of receiver-operating charac-teristic (ROC) curves for iFR to assess the optimal cutoffpoint with respect to the clinical threshold of FFR �0.80,assessment of the overall correlation between iFR and FFRusing regression techniques, and assessment of the sensi-tivity, specificity, positive predictive value (PPV), negativepredictive value (NPV), and overall diagnostic accuracy atthat cutoff point. All of the preceding analyses were alsoperformed with the cycle-averaged resting pressure ratioPd/Pa, and iFR and Pd/Pa were directly compared withrespect to their diagnostic accuracy. In addition, subgroupanalyses were performed for both iFR and Pd/Pa with respectto coronary vessel (left anterior descending [LAD] coronaryartery vs. non-LAD), route of adenosine administration(intravenous vs. intracoronary), and study site (to assesscenter variability).Statistical analysis. Data are summarized by descriptivestatistics. Pearson’s correlation and linear regression anal-ysis were performed to examine the relationship betweeniFR and FFR and Pd/Pa and FFR, respectively. ROCcurves were constructed to identify the concordance bet-ween FFR, iFR, and Pd/Pa. Agreement between themethods was assessed by Bland-Altman plots with corre-sponding 95% limits of agreement. Sensitivity, specificity,PPV, NPV, and overall diagnostic accuracy of measure-ment of iFR and Pd/Pa relative to an FFR cutoff of �0.80were determined, C statistics were generated, and optimalcutoff values for iFR and Pd/Pa were computed based onmaximizing the sum of sensitivity plus specificity. Binaryvariables were compared using chi-square testing. Fromthe raw data examining the relationship between iFR(or Pd/Pa) and FFR, separate iFR (Pd/Pa) thresholds weredetermined for which the PPV and NPV were each �90%(corresponding to an FFR of�0.80 and>0.80, respectively),and the proportion of lesions meeting these criteria wasdetermined. Similar analyses were performed using differentthresholds from �90% to �99%. SAS software version 9.1(SAS Institute, Cary, North Carolina) was used for allanalyses, and a 2-tailed p value of <0.05 was regarded asstatistically significant.

Figure 1Scatter Plot Showing the Relationship BetweeniFR and FFR and Pd/Pa and FFR

The dashed blue line represents the line of best fit. (A) The horizontal dashed line

notes the optimal iFR cutoff of 0.90 on the basis of ROC analysis. (B) The

horizontal dashed line notes the optimal Pd/Pa cutoff of 0.92. FFR¼ fractional flow

reserve; iFR ¼ instantaneous wave-free ratio; Pd/Pa ¼ distal coronary artery

pressure/aortic pressure; ROC ¼ receiver-operating characteristic.

Results

Patient demographics and procedural data. A total of1,768 patients with 1,974 lesions from 15 clinical sites weresubmitted for analysis. Of these lesions, 381 (19.3%) met atleast one of the pre-defined core laboratory exclusion criteria,leaving 1,593 lesions for final analysis. The most commonreasons for exclusion were insufficient baseline recording orartifact during recording (n ¼ 227), lesions not meetingstudy entry criteria (n ¼ 56), pressure drift or incorrectcalibration (n ¼ 42), and other technical factors (n ¼ 56).

The mean age of the population was 63.4 � 10.3 years,and 74.9% were male. A total of 21.2% had prior myocar-dial infarction and 28.1% had diabetes mellitus, and 29.4%were current smokers. A small fraction had prior coronaryartery bypass grafting (3.4%), chronic kidney disease (8.3%),and congestive heart failure (6.3%). The clinical presentationwas most commonly chronic stable angina (68.6%), with14.4% having unstable angina and 8.4% non–ST-segmentelevation myocardial infarction. More than half of thepopulation had multivessel coronary artery disease (53.8%);the LAD coronary artery was the most commonly interro-gated target lesion (63%), followed by the right coronaryartery (20%) and the left circumflex artery (17%). FFRstudies were performed with intravenous adenosine in 80.1%of cases and intracoronary adenosine in the remainder.Relationships between FFR, iFR, and Pd/Pa. Themedian (interquartile range) FFR, iFR, and Pd/Pa in thisstudy population was 0.79 (0.70 to 0.86), 0.90 (0.83 to

Figure 3 Bland-Altman Analysis

Bland-Altman plots of differences against the means are displayed for (A) iFR and

(B) Pd/Pa. The zero line is displayed in red. The mean bias is represented by the

solid blue line (with the 95% confidence interval represented by the dashed blue

line). Abbreviations as in Figure 1.

JACC Vol. 63, No. 13, 2014 Jeremias et al.April 8, 2014:1253–61 Instantaneous Wave-Free Ratio and Pd/Pa Compared to FFR

1257

0.95), and 0.93 (0.86 to 0.96), respectively. A scatter plotbetween iFR and FFR is shown in Figure 1A, demon-strating moderate overall linear correlation between the 2measures, with an R2 of 0.66 (95% confidence interval [CI]:0.64 to 0.70) (p < 0.001). Similarly, the correlation ofresting Pd/Pa and FFR demonstrated an R2 of 0.69 (95%CI: 0.67 to 0.72) (p < 0.001) (Fig. 1B). Although theoverall correlations between Pd/Pa versus FFR and iFRversus FFR were similar, the data points were more clusteredaround the regression line with a flatter slope and greaterintercept for the Pd/Pa versus FFR relationship. The areaunder the ROC curve (C statistic) to predict an FFR �0.80was 0.81 (95% CI: 0.79 to 0.83) for iFR and 0.82 (95% CI:0.80 to 0.84) for Pd/Pa, indicating moderate to gooddiscrimination for both (Fig. 2). The optimal cutoff value foran FFR �0.80 derived from ROC analyses was 0.90 for iFRand 0.92 for Pd/Pa.

Bland-Altman plots for iFR and Pd/Pa are shown inFigures 3A and 3B, respectively. On average, iFR exceededFFR by þ0.10 (95% CI: �0.06 to þ0.26) and Pd/Pa

exceeded FFR by þ0.14 (�0.01 to þ0.29). However, bothiFR and Pd/Pa demonstrated a substantial degree of scatter,particularly below the threshold of 0.80.

The correlation between iFR and Pd/Pa is shown inFigure 4A. There was a strong correlation between these 2parameters (R2 ¼ 0.95; p < 0.001), showing that 95% of thevariation in iFR was accounted for by Pd/Pa. However,Bland-Altman analysis showed that Pd/Pa overestimates iFRon average by 0.04 and substantially more when the iFRis <0.80 (Fig. 4B).Diagnostic accuracy of iFR. The overall sensitivity,specificity, PPV, and NPV for iFR �0.90 versus FFR�0.80 was 78.9%, 82.4%, 85.2%, and 73.3%, respectively,with an overall diagnostic accuracy of 80.4%. To achieve�90% diagnostic accuracy at each extreme, the overall

Figure 2 ROC Curves for iFR and Pd/Pa

Comparisons are made with an FFR at a cut point of 0.80. Abbreviations as

in Figure 1.

iFR range had to be restricted to �0.88 (to predict anFFR �0.80) and �0.97 (to predict an FFR >0.80),comprising 1,034 of the 1,593 study lesions (64.9%). Inother words, if a �90% diagnostic accuracy compared withFFR is deemed sufficient for therapeutic interchangeability,64.9% (95% CI: 62.6% to 67.3%) of the study lesionswould fall within the adenosine-free zone and not requirehyperemia for the diagnosis of ischemia. Figure 5 demon-strates the association between the adenosine-free zoneand diagnostic accuracy. The adenosine-free zone narrowsas increasing diagnostic accuracy of iFR is required, suchthat only 28.6% (26.4% to 30.8%) and 18.0% (16.1% to19.8%) of lesions would achieve �95% and �99% diag-nostic accuracy, respectively.Diagnostic accuracy of Pd/Pa. The sensitivity, specificity,PPV, and NPV for a Pd/Pa �0.92 for an FFR �0.80 was76.3%, 88.1%, 89.2%, and 74.4%, respectively, resulting inan overall diagnostic accuracy of 81.5%. A diagnosticaccuracy of �90% was achieved when the Pd/Pa range wasrestricted to �0.92, with 769 of the 1,593 (48.3%; 45.6%to 50.5%) lesions falling in that range. However, incontrast to iFR, there was no upper boundary of Pd/Pa thatpredicted with �90% accuracy a negative FFR value (i.e.,

Figure 4 Relationship Between iFR and Pd/Pa

(A) The scatter plot demonstrates a highly linear relationship. (B) The Bland-

Altman plot displays differences against the mean, demonstrating substantial

variation between iFR and Pd/Pa. Abbreviations as in Figure 1. Lines as in Figure 3.

Figure 5Association Between Use of Adenosine andDiagnostic Accuracy of iFR and Pd/Pa

An inverse relationship between use of adenosine and diagnostic accuracy is

shown, such that with increasing accuracy the adenosine-free zone decreases in

width for both iFR and Pd/Pa. The blue line displays this association for iFR and the

red line for Pd/Pa. Abbreviations as in Figure 1.

Jeremias et al. JACC Vol. 63, No. 13, 2014Instantaneous Wave-Free Ratio and Pd/Pa Compared to FFR April 8, 2014:1253–61

1258

>10% of lesions with a Pd/Pa of 1.00 had an FFR �0.80).Figure 5 shows the association between the adenosine-freezone and diagnostic accuracy for Pd/Pa. Similar to iFR,there was a trade-off between diagnostic accuracy and thesize of the adenosine-free zone. Only 36.0% (33.7% to38.4%) and 19.5% (17.5% to 21.4%) of lesions wouldachieve a diagnostic accuracy of �95% and �99%,respectively.Subgroup analyses. There was no significant difference inthe diagnostic accuracy of iFR compared with FFR withintravenous versus intracoronary administration of adenosine(81.5% vs. 78.2%; p ¼ 0.07) or among patients presentingwith stable versus unstable angina (80.4% vs. 80.2%;p ¼ 0.97). Similarly, no significant differences in diagnosticaccuracy were noted when LAD coronary artery stenoseswere compared with non-LAD coronary artery stenoses(79.9% vs. 81.9%; p ¼ 0.34) or for tracings with versuswithout an embedded ECG signal (83.7% vs. 80.2%;p ¼ 0.39). Finally, the variation in overall accuracy betweeniFR and FFR at individual study sites ranged from 78.6%to 82.7%, and the correlation varied from an R2 of 0.54 toan R2 of 0.72 (Table 1). For Pd/Pa, the overall accuracy

ranged from 72.6% to 89.5% and the correlation variedfrom an R2 of 0.61 to an R2 of 0.75.

Discussion

In this large, core laboratory–based analysis, the overalllinear correlation between both iFR and Pd/Pa and FFR wasmoderate (R2 ¼ 0.66 and 0.69, respectively), with an overalldiagnostic accuracy ofw80% for both nonhyperemic indices(using the optimal ROC-determined cutoff points of 0.90and 0.92 to predict an FFR �0.80). The diagnostic accuracywas independent of vessel, embedded versus core laboratory–generated ECG gating signal, use of intravenous versusintracoronary adenosine to induce hyperemia, and clinicalsite. Accepting FFR as the reference method (in the absenceof outcome studies with iFR or Pd/Pa), this level of accuracyis insufficient to use either parameter for procedural guid-ance in all cases because w20% of therapeutic decisionswould be discordant from FFR.

Although iFR and Pd/Pa are imperfect surrogates ofFFR close to the clinically used cutoff value of 0.80(11), they may still provide acceptable accuracy at greateror lesser degrees of functional stenosis severity. Thefundamental principle of FFR, justifying pressure-derivedestimation of coronary flow impairment, is that the trans-lesional pressure ratio approximates flow when microvas-cular resistance is minimized (12,13), requiring the use ofa potent vasodilator. However, microvascular resistance isinfluenced by many factors, including capacitive, inertial,and resistive forces as well as the complex effects of systoliccontraction. Nonhyperemic pressure ratios may theoreti-cally have adequate concordance with hyperemic pressuremeasurements when there is a large baseline gradient (i.e.,

Table1

Individu

alDataFrom

Includ

edStudies

andIndividu

alStudy

Sites

Study

/Participa

ting

Site

No.

ofLe

sion

s

iFR

Pd/Pa

CutoffPoint

AUCFrom

ROC

(Cstatistic)

Ove

rallAcc

urac

y(%

)Correlation

(R2)

CutoffPoint

AUCFrom

ROC

(Cstatistic)

Ove

rallAcc

urac

y(%

)Correlation

(R2)

Total

1,593

0.90

0.81

80.4

0.66

0.92

0.82

81.5

0.69

ADVISE

*432

0.91

0.82

81.9

0.71

0.92

0.82

81.9

0.75

VERIFYy

654

0.89

0.80

79.4

0.60

0.92

0.81

79.8

0.65

Seou

lNationa

lUniversity

179

0.92

0.83

82.7

0.68

0.93

0.82

82.1

0.70

Ston

yBrook

University

149

0.93

0.81

79.2

0.54

0.93

0.83

83.2

0.61

Colum

biaUniversity

95

0.91

0.84

82.1

0.62

0.92

0.87

89.5

0.70

AMC/VUMC/K

CL

84

0.90

0.78

78.6

0.72

0.93

0.72

72.6

0.70

*Include

sda

tafrom

theADVISE

stud

yan

dADVISE

registry.yIn

clud

esda

tafrom

theprospe

ctivean

dretrospe

ctiveVE

RIFYcoho

rts.

AMC¼

Acade

mic

Med

ical

Cen

ter,University

ofAmsterda

m;AUC¼

area

unde

rthecurve;

iFR¼

instan

tane

ouswave-free

ratio

;KCL¼

King’sCollege

Lond

on;Pd/P

a¼

distal

corona

ryartery

pressure/aortic

pressure;ROC¼

receiver-ope

ratin

gcharacteris

tic;VU

MC¼

VUUniversity

Med

ical

Cen

ter,Amsterda

m.

JACC Vol. 63, No. 13, 2014 Jeremias et al.April 8, 2014:1253–61 Instantaneous Wave-Free Ratio and Pd/Pa Compared to FFR

1259

obvious impairment of coronary flow) or no gradient atall (i.e., absence of any resting flow disturbance). In thisregard, a recent retrospective analysis of almost 500patients demonstrated a good correlation between Pd/Pa

and FFR with an area under the curve of 0.86 (14). Whenonly translesional resting pressure ratios of <0.88 and>0.95 were considered, the PPV and NPV increased to>95%, with more than half of the study population fallingin these categories. The present larger, multicenter, corelaboratory–based analysis shows that if 90% accuracy com-pared with the FFR reference standard is accepted at themargins (the pre-specified precision limit for therapeuticinterchangeability in the present study), use of iFR andPd/Pa might avoid hyperemia in 65% and 48% of lesions,respectively. If 95% accuracy is required, however, use ofiFR and Pd/Pa might avoid hyperemia in only 29% and36% of lesions, respectively. In addition, the percentage oflesions falling into the adenosine-free zone will varybased on the spectrum of lesions being studied. If onlyintermediate lesions are investigated (i.e., with an FFR near0.80 in a greater proportion of patients), the adenosine-freezone may be smaller compared with the findings of thecurrent study.

A secondary goal of the present study was to compare andcontrast iFR and Pd/Pa. By restricting measurements toa specific segment of diastole in which the maximumachievable coronary flow occurs during resting conditions,iFR has a theoretical advantage compared with Pd/Pa.However, using FFR as the reference standard, we found nosignificant differences between iFR and Pd/Pa with respectto sensitivity, specificity, PPV, NPV, or diagnostic accuracy.Although modest differences were noted between the iFRand Pd/Pa versus FFR regression patterns, the overall similarresults are consistent with a prior retrospective analysis byJohnson et al. (10). Prospective studies are required todetermine whether the differences between iFR and Pd/Pa

are practically or clinically relevant.Study limitations. The present study has several strengthsbut also some limitations. Prior studies examining therelationship between iFR, Pd/Pa, and FFR showed signi-ficant variability and thus reached strikingly differentconclusions (8–10). In this regard, it is reassuring to notethat by applying a rigorous study methodology, commoninclusion and exclusion criteria, and a standardized physi-ology assessment methodology, the data from these priorstudies showed relatively little variation, with diagnosticaccuracy ranging from 79% to 83%. We have applied linearmodels to our data, although the complete physiologicalrelationship between FFR and iFR or rest Pd/Pa may bestbe described by a curvilinear relationship. RESOLVE isthe first coronary physiology study that used a core labo-ratory for analysis of hyperemic and resting pressure–derived indices of the severity of stenosis. Surprisingly, 19%of measurements were found to be suboptimal and wereexcluded from the analysis (perhaps explaining the reducedsite-to-site variability in the present report compared with

Jeremias et al. JACC Vol. 63, No. 13, 2014Instantaneous Wave-Free Ratio and Pd/Pa Compared to FFR April 8, 2014:1253–61

1260

previously reported individual studies). Future clinical trialsshould consider including core laboratory analysis to assessthe validity of hemodynamic measurements, as is currentlythe standard for quantitative coronary angiography andintravascular ultrasonography. An additional strength is thesize of the present study, encompassing all iFR studiespublished to date as well as several nonpublished clinicalexperiences, which provides incremental power to accu-rately locate point estimates while reducing CI width andaffording subgroup analysis. However, the present retro-spective analysis is limited by nonuniform patient andlesion characteristics at each site and varying FFR acqui-sition protocols. Despite the fact that all studies underwentrigorous analysis by an independent core laboratory toeliminate potential erroneous measurements and minimizevariability, we cannot fully exclude selection bias and othersources of inconsistencies. A final pullback of the pressurewire into the guiding catheter confirming the absence ofpressure drift was not required and was performed in onlya small minority of cases.Clinical implications. As with any diagnostic test FFR,iFR and Pd/Pa have inherent variability (9,15,16). On thebasis of 3 randomized trials showing superior clinicaloutcomes with FFR guidance compared with angiographicguidance alone (5–7), FFR is justifiably accepted as thestandard in both US and European guidelines for invasivephysiological lesion assessment and clinical decision making(17,18). On the basis of the present report and consistentwith prior studies (9,10), the universal adoption of iFR orPd/Pa with use of a single cutoff point cannot be recom-mended (19). However, using a hybrid approach whereinPd/Pa or iFR are accepted at the 2 outer tails of the spectrumwith FFR-based decisions required in the gray area inbetween (20) may be feasible and might avoid the useof hyperemia in approximately 48% to 65% of lesions,respectively, if �90% correlation with an FFR cutoff �0.80is accepted. Although there will always be a trade-off forgreater diagnostic accuracy (e.g., if >99% accuracy comparedwith FFR is desired, the adenosine-free zone would shrinkto <20% of patients), a small (�10%) degree in variabilitybetween nonhyperemic physiological measurements andFFR in a large proportion of patients may be acceptable tomany physicians in daily clinical practice given the cost,inconvenience, and potential side effects associated withadministration of adenosine (21,22) and the relatively lowmajor adverse cardiac event rate around the FFR 0.80 cutpoint (5), where most classification errors are likely to occur.However, the iFR and Pd/Pa cutoff values identified in thepresent retrospective study require validation, and prospec-tive randomized trials are required to determine whethera hybrid strategy results in noninferior clinical outcomes tothe routine use of FFR.

AcknowledgmentsThe authors thank Liang Dong, MD, Lin Wang, MD,Shinji Inaba, MD, PhD, Shigeo Saito, MD, Tomotaka

Dohi, MD, PhD, Nobuaki Kobayashi, MD, PhD, and EliasSanidas, MD, PhD, for their invaluable efforts with dataanalysis.

Reprint requests and correspondence: Dr. Allen Jeremias, Divi-sion of Cardiology, Department of Medicine, Health SciencesCenter, T16-080, Stony Brook, New York 11794-8160. E-mail:allen.jeremias@stonybrook.edu.

REFERENCES

1. Pijls NH, De Bruyne B, Peels K, et al. Measurement of fractional flowreserve to assess the functional severity of coronary-artery stenoses.N Engl J Med 1996;334:1703–8.

2. Pijls NH, Van Gelder B, Van der Voort P, et al. Fractional flowreserve. A useful index to evaluate the influence of an epicardialcoronary stenosis on myocardial blood flow. Circulation 1995;92:3183–93.

3. Christou MA, Siontis GC, Katritsis DG, Ioannidis JP. Meta-analysisof fractional flow reserve versus quantitative coronary angiography andnoninvasive imaging for evaluation of myocardial ischemia. Am JCardiol 2007;99:450–6.

4. De Bruyne B, Baudhuin T, Melin JA, et al. Coronary flow reservecalculated from pressure measurements in humans. Validation withpositron emission tomography. Circulation 1994;89:1013–22.

5. De Bruyne B, Pijls NH, Kalesan B, et al. Fractional flow reserve-guidedPCI versus medical therapy in stable coronary disease. N Engl J Med2012;367:991–1001.

6. Tonino PA, De Bruyne B, Pijls NH, et al. Fractional flow reserveversus angiography for guiding percutaneous coronary intervention.N Engl J Med 2009;360:213–24.

7. Pijls NH, van Schaardenburgh P, Manoharan G, et al. Percutaneouscoronary intervention of functionally nonsignificant stenosis: 5-yearfollow-up of the DEFER Study. J Am Coll Cardiol 2007;49:2105–11.

8. Sen S, Escaned J, Malik IS, et al. Development and validation of a newadenosine-independent index of stenosis severity from coronary wave-intensity analysis: results of the ADVISE (ADenosine VasodilatorIndependent Stenosis Evaluation) study. J Am Coll Cardiol 2012;59:1392–402.

9. Berry C, van ’t Veer M, Witt N, et al. VERIFY (VERification ofInstantaneous Wave-Free Ratio and Fractional Flow Reserve for theAssessment of Coronary Artery Stenosis Severity in EverydaY Prac-tice): a multicenter study in consecutive patients. J Am Coll Cardiol2013;61:1421–7.

10. Johnson NP, Kirkeeide RL, Asrress KN, et al. Does the instantaneouswave-free ratio approximate the fractional flow reserve? J Am CollCardiol 2013;61:1428–35.

11. Petraco R, Escaned J, Sen S, et al. Classification performance ofinstantaneous wave-free ratio (iFR) and fractional flow reserve ina clinical population of intermediate coronary stenoses: results of theADVISE registry. EuroIntervention 2013;9:91–101.

12. Pijls NH, van Son JA, Kirkeeide RL, et al. Experimental basis ofdetermining maximum coronary, myocardial, and collateral blood flowby pressure measurements for assessing functional stenosis severitybefore and after percutaneous transluminal coronary angioplasty.Circulation 1993;87:1354–67.

13. Gould KL, Lipscomb K, Hamilton GW. Physiologic basis for assessingcritical coronary stenosis. Instantaneous flow response and regionaldistribution during coronary hyperemia as measures of coronary flowreserve. Am J Cardiol 1974;33:87–94.

14. Mamas MA, Horner S, Welch E, et al. Resting Pd/Pa measured withintracoronary pressure wire strongly predicts fractional flow reserve.J Invasive Cardiol 2010;22:260–5.

15. Kern MJ, Lerman A, Bech JW, et al. Physiological assessment ofcoronary artery disease in the cardiac catheterization laboratory:a scientific statement from the American Heart Association Committeeon Diagnostic and Interventional Cardiac Catheterization, Council onClinical Cardiology. Circulation 2006;114:1321–41.

JACC Vol. 63, No. 13, 2014 Jeremias et al.April 8, 2014:1253–61 Instantaneous Wave-Free Ratio and Pd/Pa Compared to FFR

1261

16. Spaan JA, Piek JJ, Hoffman JI, SiebesM. Physiological basis of clinicallyused coronary hemodynamic indices. Circulation 2006;113:446–55.

17. Levine GN, Bates ER, Blankenship JC, et al. 2011 ACCF/AHA/SCAI Guideline for Percutaneous Coronary Intervention: a report ofthe American College of Cardiology Foundation/American HeartAssociation Task Force on Practice Guidelines and the Society forCardiovascular Angiography and Interventions. J Am Coll Cardiol2011;58:e44–122.

18. Wijns WKP, Danchin N, Di Mario C, et al. Task Force on MyocardialRevascularization of the European Society of Cardiology (ESC) andthe European Association for Cardio-Thoracic Surgery (EACTS);European Association for Percutaneous Cardiovascular Interventions(EAPCI). Eur Heart J 2010;20:2501–55.

19. Samady H, Gogas BD. Does flow during rest and relaxation suffice?J Am Coll Cardiol 2013;61:1436–9.

20. Petraco R, Park JJ, Sen S, et al. Hybrid iFR-FFR decision-makingstrategy: implications for enhancing universal adoption of physiology-guided coronary revascularisation. EuroIntervention 2013;8:1157–65.

21. Cerqueira MD, Verani MS, Schwaiger M, Heo J, Iskandrian AS.Safety profile of adenosine stress perfusion imaging: results from theAdenoscan Multicenter Trial Registry. J Am Coll Cardiol 1994;23:384–9.

22. Kleiman NS. Bringing it all together: integration of physiology withanatomy during cardiac catheterization. J Am Coll Cardiol 2011;58:1219–21.

Key Words: coronary physiology - fractional flow reserve -

myocardial ischemia.