Zhenhua XingI; Junyu PeiI; Jiabing HuangI; Xinqun HuI; Shan GaoII
ABSTRACTIntroduction: Quantitative flow ratio (QFR) is a novel method enabling efficient computation of FFR from three-dimensional quantitative coronary angiography (3D QCA) and thrombolysis in myocardial infarction (TIMI) frame counting. We decided to perform a systematic review and quantitative meta-analysis of the literature to determine the correlation between the diagnosis of functionally significant stenosis obtained by QFR versus FFR and to determine the diagnostic accuracy of QFR for intermediate coronary artery stenosis.
3D = Three-dimensional
3D QCA = Three-dimensional quantitative coronary angiography
AUC = Area under the curve
CI = Confidence interval
DOR = Diagnostic odds ratio
EAPCI = European Association of Percutaneous Cardiovascular Interventions
ESC = European Society of Cardiology
FFR = Fractional flow reserve
FP = False positive
FAVOR = Functional Diagnostic Accuracy of Quantitative Flow Ratio in Online Assessment of Coronary Stenosis study
FN = False negative
LR = Likelihood ratio
LR+ = Positive likelihood ratio
LR- = Negative likelihood ratio
OR = Odds ratio
MI = Myocardial infarction
PCI = Percutaneous coronary intervention
PRISMA = Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols
QFR = Quantitative flow ratio
QUADAS-2 = Quality Assessment of Diagnostic Accuracy Studies 2
sROC = Summary receiver operating characteristic
STEMI = ST-elevation myocardial infarction
TIMI = Thrombolysis in myocardial infarction
TN = True negative
TP = True positive
Accurate evaluation of coronary artery disease, especially intermediate coronary artery stenosis, is crucial for the evaluation of myocardial ischemia and next treatment. The gold standard for diagnosis and confirmation of functional significance of a stenosis is the fractional flow reserve (FFR). Previous studies have demonstrated that FFR-guided coronary revascularization increases the ratio of event-free survival when compared with a coronary stenosis-guided strategy[1,2]. Despite these advantages, the clinical application of FFR has been variable and slow. FFR requires not only the hyperemic state, but also additional cost, time, and efforts.
Quantitative flow ratio (QFR) is a novel method enabling efficient computation of FFR from three-dimensional quantitative coronary angiography (3D QCA) and thrombolysis in myocardial infarction (TIMI) frame counting. Compared with FFR, QFR does not require any invasive physiological measurements, pharmacological hyperemia induction, and additional cost. The recently FAVOR (Functional Diagnostic Accuracy of Quantitative Flow Ratio in Online Assessment of Coronary Stenosis) II China study showed solid results for QFR computation in identifying the presence of functionally significant stenosis in eligible patients. Several studies have been published in the literature addressing the correlation between the assessment of functionally significant stenosis obtained by QFR versus FFR and addressing the diagnostic accuracy of QFR for intermediate coronary artery stenosis[5,6]. The purpose of our study was to perform a systematic review and quantitative meta-analysis of the literature to determine the correlation between the diagnosis of functionally significant stenosis obtained by QFR versus FFR, and to determine the diagnostic accuracy of QFR for intermediate coronary artery stenosis.
This protocol is reported following the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA) guidelines. We searched PubMed, Embase, and Web of Science published before March 15, 2018. The keywords used for search were "QFR or Quantitative flow ratio". Results were limited to trials published in English. We manually searched reference lists of relevant studies and reviews, editorials, and letters to identify further articles. We used Endnote (Thompson ISI ResearchSoft, Philadelphia, USA) to manage relevant articles and remove duplicated articles.
The inclusion criteria for studies in the analysis were as follows: 1) The design was a diagnostic accuracy study; 2) The study assess the diagnostic performance of QFR compared with invasive FFR as the standard procedure; 3) Data from true positive (TP), false positive (FP), true negative (TN), false negative (FN), sensitivity and specificity can be retrieved or calculated. When relevant data were missing, authors were contacted by e-mail, before excluding the study due to inaccessibility of data.
Data Collection and Quality Assessment
Relevant data were initially extracted by two independent reviewers (Zh Xing and Jy Pei). Disagreements were resolved by consensus or by a third investigator (XQ Hu). We abstracted the following data from the selected articles: first author, publication date, study design, patient demographics; FFR threshold used to describe ischemia; and the data of TP, FP, TN, and FN. When different flow models of QFR were performed, contrast-flow QFR was preferred. Contrast-flow QFR was more accurate for predicting FFR ≤0.80 as compared with fixed-flow QFR. Included studies were analyzed by the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2).
The inter-reviewer agreement regarding the quality assessment of included studies was assessed by the Cohen kappa test. Our meta-analysis was performed using the DerSimonian and Laird random effects model to determine sensitivity, specificity, positive likelihood ratio (LR+), negative likelihood ratio (LR), and diagnostic odds ratio (DOR). The sROC was used to determine diagnostic test accuracy. An area under the curve (AUC) between 0.75 and 0.92 represented a high degree of diagnostic accuracy, and an AUC between 0.93 and 0.96 was considered more accurate. In order to assess heterogeneity among the studies, the statistic was used. For, a value >50% was considered of severe heterogeneity. The Spearman correlation coefficient was calculated to evaluate diagnostic threshold variation among the included studies.
We also performed a meta-regression analysis to identify predefined potential sources of heterogeneity. All statistical analyses were completed using Meta-DiSc (version 1.4).
Study Selection and Characteristics
The flowchart of our search and selection process was presented in Figure 1. Our combined search strategy identified possible relevant studies. Nine studies were retrieved for a more detailed evaluation. Finally, 9 studies consisting of 1175 vessels in 1047 patients met our inclusion criteria[4-6), (9-15]. Characteristics of included studies were shown in Table 1. Clinical heterogeneity was mostly due to different inclusion criteria. Mejia-Renteria and Emori included patients with myocardial infarction. Emori included two different populations: patients with previous myocardial infarction (MI) and patients without previous MI. Spitaleri included patients with ST- elevation myocardial infarction (STEMI) and multivessel disease. Four studies were performed in Japan, two in China, one each in Spain, Italy, and Netherlands. The mean (SD) age was 63.2 years, and 68.1% of the patients were male. The quality assessment of included studies according to QUADAS-2 was presented in Supplementary Figure 1. In general, there was low risk of bias and low concern regarding applicability of all included studies.
|Study||Design||Country||Mean age (years)||Males (%)||Patients' characteristics||Cutoff|
|Emori et al.||Multicenter||Japan||-||-||Intermediate stenosis with FFR||0.8|
|Xu et al.||Prospective, multicenter||China||61||73.7||Intermediate stenosis with FFR||0.8|
|Yazaki et al.||Retrospective, single-center||Japan||72.5||29.6||Intermediate stenosis with FFR||0.8|
|Emori et al. MI*||Retrospective, single-center||Japan||69||83||Patients with previous MI undergoing CAG and FFR||0.8|
|Emori et al. non-MI*||Retrospective, single-center||Japan||70||54||Intermediate stenosis with FFR||0.8|
|Tu et al.||Prospective, multicenter||China||65.8||61||Intermediate stenosis with FFR||0.8|
|Kameyama et al.||Multicenter||Japan||-||-||ACS patients with CAG and FFR||0.8|
|van Rosendael et al.||Prospective, single-center||Netherlands||64||71||Intermediate stenosis with FFR||0.8|
|Mejia-Renteria et al.||Multicenter||Spain||-||-||Patients with CAG and FFR||0.8|
|Spitaleri et al.||Prospective, multicenter||Italy||62||28||STEMI patients with MVD and FFR||0.8|
|Spitaleri et al.||Prospective, multicenter||Denmark||61||67||Intermediate stenosis with FFR||0.8|
* Emori 2018 contained two groups: patients with previous myocardial infarction (MI) and patients with no previous MI.
Diagnostic Accuracy of QFR
The results of the included study were presented in Table 2. The accuracy ranged from 80% in Kameyama and 94% in Spitaleri, et al.. Sensitivity ranged between 74% in Tu, et al. and 100% in van Rosendael, et al. and the specificity ranged from 79% in Emori, et al. to 97% in Spitaleri, et al.. The correlation between QFR and FFR ranged from r = 0.69 to r = 0.94.
|Study||Included vessels (n)||Sensitivity (%)||Specificity (%)||Accuracy (%)||Correlation (r)|
|Emori et al.||73||82||79||81||0.69|
|Xu et al.||328||94.6||91.7||92.7||0.86|
|Yazaki et al.||151||88.7||89.1||88.7||0.84|
|Emori et al. MI||75||92||82||87||0.88|
|Emori et al. non-MI||75||95||88||92||0.94|
|Tu et al.||84||74||91||86||0.77|
|Kameyama et al.||25||80||80||80||0.63|
|van Rosendael et al.||15||100||79||80||0.78|
|Mejia-Renteria et al.||300||88||86||87||-|
|Spitaleri et al.||49||88||97||94||0.90|
In pooled data weighted by the number of vessels, QFR had a combined sensitivity and the specificity of QFR for diagnosis of functional significance of a stenosis according to FFR were 0.89 (95% CI: 0.86-0.92) and 0.88 (95% CI: 0.86-0.91), respectively, using a random effects model (Figure 2). No heterogeneity was found for both sensitivity (I2=38.3%, P=0.10) and specificity (I2=24.1%, P=0.22). The pooled estimate of positive likelihood ratio (LR+) and negative likelihood ratio (LR-) was 6.86 (95% CI: 5.22-9.02) and 0.14 (95% CI: 0.10-0.21) (Figure 3). For QFR, Spearman's correlation coefficients were 0.619 (P=0.102), indicating that the diagnostic threshold effect did not exist in QFR data. The area under the ROC curve was 0.94 (Figure 4) and the diagnostic OR was 53.05 (95% CI: 29.75-94.58) (Supplementary Figure 2).
Meta-regression Analysis and Subgroup Analysis
Meta-regression was performed using the potential sources of heterogeneity among studies (age, country, sex, different inclusion criteria). We found no factor effecting the diagnostic accuracy.
Emori, et al. and Mejia-Renteria, et al. included patients with previous MI that might affect the diagnostic accuracy. Exclusion of these two trials slightly improved the specificity (0.90, 0.87-0.92), but did not affect sensitivity.
In the present study, we performed a systematic review of the diagnostic performance of QFR for functional significance of intermediate coronary artery stenosis compared with invasive FFR. Data from 1175 vessels in 1047 patients showed the situation in which QFR is helpful for surgeons to determine whether stents should be implanted at no additional cost, time, and effort. Our study is the first systemic review and meta-analysis that evaluates the diagnostic accuracy of QFR for the assessment of functionally significant stenosis confirmed by FFR.
FFR-guided percutaneous coronary intervention (PCI) is associated with a better outcome compared with revascularization based on angiographic stenosis severity alone in patients with intermediate coronary artery stenosis[2,16,17]. FFR has the highest recommendation (class I, level A) in the European Society of Cardiology guideline on myocardial revascularization. Although FFR has better outcomes in patients with intermediate coronary artery stenosis, the clinical application has been limited due to the invasive procedure with a pressure wire, the cost of pressure wire, and the side effects associated with induction of hyperemia. QFR, an angiographic index of coronary stenosis severity based on 3D QCA and thrombolysis in myocardial infarction (TIMI) frame counting, estimates FFR without invasive procedure. FAVOR Pilot Study and FAVOR II China study have shown good agreement of QFR with invasive FFR[4,5]. However, the simple size was too small. Our meta-analysis included studies of which the majority were performed in China and Japan from 2016 to 2018, with a total of 1175 vessels in 1047patients, and demonstrated that the diagnostic accuracy of QFR for functionally significant stenosis confirmed by FFR was high, with a summary sensitivity and specificity of 0.89 (95% CI: 0.86-0.92) and 0.88 (95% CI: 0.86-0.91), respectively.
Our studies included different study populations (stable coronary artery disease, suspected coronary artery disease, STEMI, previous MI). However, our meta-regression showed that different study populations did not affect our diagnostic accuracy. QFR was more often used in patients with stable coronary artery disease. A recent study has found that QFR may be a safe and reliable tool to guide revascularization in patients with STEMI and multivessel disease. Furthermore, Spitaleri found that functional complete revascularization evaluated by QFR showed a good 5-year outcome.
However, the diagnostic accuracy of QFR for assessing the functional severity of coronary stenosis might be affected in coronary arteries related to previous MI. Microcirculatory resistance may affect this phenomenon. Mejia-Renteria, et al. found that the diagnostic accuracy of QFR was lower in patients with high microcirculatory resistance, which is supported by our subgroup analysis. Due to its specific algorithm based on QCA and TIMI frame counting, coronary collateral circulation may reduce its accuracy. Therefore, corrective measures need to be developed to improve the diagnostic accuracy in patients with previous MI or high microcirculatory resistance. Furthermore, coronary calcification or thrombus may lead to angiographic haziness which undoubtedly reduces the diagnostic accuracy of QFR. A hybrid QFR-FFR approach may be a way to overcome these limitations. Yazaki, et al. found that FFR should be performed in stenosis with QFR 0.75-0.85. This hybrid approach may allow clinicians to get the best of both worlds by ensuring diagnostic accuracy while reducing cost and side effects.
It should be noted that our conclusion should be seen in the context of its limitation. First, the simple size is relatively small. Second, although there was no apparent heterogeneity in statistics, the heterogeneity in clinical and methodology was inevitable.
1. van Nunen LX, Zimmermann FM, Tonino PA, Barbato E, Baumbach A,Engstrøm T, et al. Fractional flow reserve versus angiography forguidance of PCI in patients with multivessel coronary artery disease (FAME):5-year follow-up of a randomised controlled trial. Lancet. 2015 Nov7;386(10006):1853-60. doi: 10.1016/S0140-6736(15)00057-4.
2. De Bruyne B, Pijls NH, Kalesan B, Barbato E, Tonino PA, Piroth Z, etal. Fractional flow reserve-guided PCI versus medical therapy in stable coronarydisease. N Engl J Med. 2012 Sep 13;367(11):991-1001. doi: 10.1056/NEJMoa1205361.Epub 2012 Aug 27. Erratum in: N Engl J Med. 2012 Nov;367(18):1768.Mobius-Winckler, Sven . [MedLine]
3. Dattilo PB, Prasad A, Honeycutt E, Wang TY, Messenger JC.Contemporary patterns of fractional flow reserve and intravascular ultrasounduse among patients undergoing percutaneous coronary intervention in the UnitedStates: insights from the National Cardiovascular Data Registry. J Am CollCardiol. 2012 Dec 4;60(22):2337-9. doi:10.1016/j.jacc.2012.08.990.
4. Tu S, Westra J, Yang J, von Birgelen C, Ferrara A, Pellicano M, etal. Diagnostic Accuracy of Fast Computational Approaches to Derive FractionalFlow Reserve From Diagnostic Coronary Angiography: The International MulticenterFAVOR Pilot Study. JACC Cardiovasc Interv. 2016 Oct 10;9(19):2024-35. doi:10.1016/j.jcin.2016.07.013. [MedLine]
5. Xu B, Tu S, Qiao S, Qu X, Chen Y, Yang J, et al. Diagnostic Accuracyof Angiography-Based Quantitative Flow Ratio Measurements for Online Assessmentof Coronary Stenosis. J Am Coll Cardiol. 2017 Dec 26;70(25):3077-87.doi:10.1016/j.jacc.2017.10.035. [MedLine]
6. Yazaki K, Otsuka M, Kataoka S, Kahata M, Kumagai A, Inoue K, et al.Applicability of 3-Dimensional Quantitative Coronary Angiography-DerivedComputed Fractional Flow Reserve for Intermediate Coronary Stenosis. Circ J.2017 Jun 23;81(7):988-992. doi: 10.1253/circj.CJ-16-1261. [MedLine]
7. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC,Ioannidis JP, et al. The PRISMA statement for reporting systematic reviews andmeta-analyses of studies that evaluate healthcare interventions: explanation andelaboration. BMJ. 2009 Jul 21;339:b2700. doi:10.1136/bmj.b2700.
8. Whiting P, Rutjes AW, Reitsma JB, Bossuyt PM, Kleijnen J. Thedevelopment of QUADAS: a tool for the quality assessment of studies ofdiagnostic accuracy included in systematic reviews. BMC Med Res Methodol. 2003Nov 10;3:25.
9. Mejia-Renteria H, Lauri F, Lee JM, der Hoeven N, de Waard G, deHoyps A, et al. TCT-70 Influence of microcirculatory resistance on theassessment of coronary stenosis severity with Quantitative Flow Ratio (QFR):Results of an international multicentre study. J Am Coll Cardiol. 2017;70(18Suppl B): B30-1. doi: 10.1016/j.jacc.2017.09.121
10. van Rosendael AR, Koning G, Dimitriu-Leen AC, Smit JM,Montero-Cabezas JM, van der Kley F, et al. Accuracy and reproducibility of fastfractional flow reserve computation from invasive coronary angiography. Int JCardiovasc Imaging. 2017 Sep;33(9):1305-1312.doi:10.1007/s10554-017-1190-3. [MedLine]
11. Kameyama T, Kubo T, Emori H, Ino Y, Matsuo Y, Yamano T, et al.TCT-542 Usefulness of QFR measurement for non-culprit lesion of ACS patients. JAm Coll Cardiol 2016;.68(18 Suppl B):B219. doi:10.1016/j.jacc.2016.09.679
12. Spitaleri G, Tebaldi M, Biscaglia S, Westra J, Brugaletta S,Erriquez A, et al. Quantitative Flow Ratio Identifies Nonculprit CoronaryLesions Requiring Revascularization in Patients With ST-Segment-ElevationMyocardial Infarction and Multivessel Disease. Circ Cardiovasc Interv. 2018Feb;11(2):e006023. doi:10.1161/CIRCINTERVENTIONS.117.006023. [MedLine]
13. Emori H, Kubo T, Kameyama T, Ino Y, Matsuo Y, Kitabata H, et al.Diagnostic Accuracy of Quantitative Flow Ratio for Assessing Myocardial Ischemiain Prior Myocardial Infarction. Circ J. 2018 Feb 23;82(3):807-814.doi:10.1253/circj.CJ-17-0949. [MedLine]
14. Kobayashi Y, Lee JM, Fearon W, Lee J, Zimmermann F, Jung JH, et al.TCT-522 The Independence of Coronary Microvascular Dysfunction from EpicardialDisease Severity: Three-Vessel Invasive Coronary Physiologic Study. J Am CollCardiol 2016 (18 Suppl B):B210-211. doi10.1016/j.jacc.2016.09.659
15. Emori H, Kubo T, Kameyama T Ino Y, Matsuo Y, Tanaka A, et al.TCT-529 The correlation between Quantitative Flow Ratio (QFR) and FractionalFlow Reserve (FFR). J Am Coll Cardiol. 2016;68(18 Suppl B):B213-4. doi: 10.1016/j.jacc.2016.09.666
16. Tonino PA, Fearon WF, De Bruyne B, Oldroyd KG, Leesar MA, Ver LeePN, et al. Angiographic versus functional severity of coronary artery stenosesin the FAME study fractional flow reserve versus angiography in multivesselevaluation. J Am Coll Cardiol. 2010 Jun 22;55(25):2816-21. doi:10.1016/j.jacc.2009.11.096.
17. Pijls NH, van Schaardenburgh P, Manoharan G, Boersma E, Bech JW,van't Veer M, et al. Percutaneous coronary intervention of functionallynonsignificant stenosis: 5-year follow-up of the DEFER Study. J Am Coll Cardiol.2007 May 29;49(21):2105-11.
18. Authors/Task Force members, Windecker S, Kolh P, Alfonso F, ColletJP, Cremer J, et al. 2014 ESC/EACTS Guidelines on myocardial revascularization:The Task Force on Myocardial Revascularization of the European Society ofCardiology (ESC) and the European Association for Cardio-Thoracic Surgery(EACTS)Developed with the special contribution of the European Association ofPercutaneous Cardiovascular Interventions (EAPCI). Eur Heart J. 2014 Oct1;35(37):2541-619. doi:10.1093/eurheartj/ehu278.
No financial support.
No conflict of interest.
Authors' roles & responsibilities
ZX Designed the study and provided methodological expertise; final approval of the version to be published
JP Drafted the manuscript; final approval of the version to be published
JH Drafted the manuscript; final approval of the version to be published
XH Drafted the tables and figures; final approval of the version to be published
SG Designed the study and provided methodological expertise; final approval of the version to be published
Article receive on Wednesday, August 8, 2018