Impact of Lesion Length on Functional Significance in Intermediate Coronary Lesions

Authors


Abstract

Background:

Myocardial fractional flow reserve (FFR) is useful in the evaluation of coronary lesion ischemia. However, the impact of lesion length on FFR has not been adequately assessed.

Hypothesis:

We hypothesized that lesion length would influence functional significance in intermediate coronary lesions.

Methods:

FFR measurements were assessed in 136 patients (163 lesions) with stable angina who had >40% stenotic coronary lesion by quantitative coronary angiography (QCA). One hundred sixty-three lesions were classified as intermediate (40%–70% stenosis; n=107; group I) or significant (≥70%; n=56; group S) by QCA. We assessed the relationships between lesion length, coronary stenosis, and FFR in these 163 lesions.

Results:

Regression analysis revealed an inverse correlation between the percentage of diameter stenosis (%DS) and FFR in group S (r = −0.83, P < 0.0001). In group I, no significant correlation was found between %DS and FFR (r = −0.06, P = 0.55), whereas lesion length was significantly inversely correlated with FFR (r = −0.79, P < 0.0001). Receiver operating characteristic curve analysis demonstrated that the best cutoff value for predicting an FFR value <0.80 was a lesion length >16.1 mm in group I (sensitivity, 86%; specificity, 94%).

Conclusions:

These study findings suggest that lesion length has a physiologically significant impact on intermediate-grade coronary lesions. Clin. Cardiol. 2011 DOI: 10.1002/clc.22076

The authors have no funding, financial relationships, or conflicts of interest to disclose.

Introduction

Coronary lesion severity can be assessed using both anatomical (morphology) and physiological myocardial ischemia methods. Although coronary angiography (CAG) is the conventional gold standard anatomical evaluation method, intravascular ultrasound (IVUS) has markedly improved evaluation accuracy.1–3 Optical coherence tomography, which can provide a detailed evaluation of the plaque nature, has also been developed.4,5 Although physiological assessment, including stress electrocardiogram, echocardiography, and single-photon emission computed tomography (SPECT) cannot be performed simultaneously with anatomical assessment, the assessment of physiological indicators by fractional myocardial flow reserve (FFR) using a pressure wire in recent years has allowed simultaneous evaluation with CAG. An FFR value <0.80 identifies ischemia-causing coronary stenoses with an accuracy of >90%.6,7

The recent remarkable development of multidetector-row computed tomography has facilitated noninvasive assessments in outpatients with coronary artery disease (CAD), but techniques for the physiological assessment of ischemia (such as SPECT) are not often performed when patients undergo invasive CAG. FFR has been used to assess revascularization indications. Several studies have suggested that, compared with angiography-guided percutaneous coronary intervention (PCI), FFR-guided PCI is associated with reduced major adverse cardiac events in patients with multivessel CAD.8–11

Although it is recommended that myocardial ischemia in CAD be determined using both physiological and morphological evaluation methods,12,13 it is often assessed visually from CAG findings in daily clinical practice. However, visual estimation of coronary lesions is inaccurate.14–16 Results obtained from quantitative coronary analysis (QCA) are often misread compared to those obtained from other modalities with regard to functional myocardial ischemia, particularly intermediate coronary lesions (40%–70% stenosis).15,17 Although lesion length is an important geometric parameter in addition to the degree of lesion stenosis in the morphological assessment of coronary ischemia, the impact of lesion length on FFR has never been adequately assessed. The purpose of this study was to investigate the relationships between FFR and lesion length, or coronary stenosis.

Methods

Study Population

The study population comprised 165 consecutive patients (199 lesions) with CAD who had at least 1 de novo lesion (>40% stenosis measured by QCA) and who had undergone FFR for the evaluation of myocardial ischemia in our institution between August 2005 and March 2011. The exclusion criteria for the present study were acute coronary syndrome (ACS), culprit lesion of previous myocardial infarction, previous coronary artery bypass surgery, lesion with a collateral vessel, tandem lesion, in-stent restenosis lesion, cardiogenic shock, low left ventricular systolic function (ejection fraction <40%), and the use of hemodialysis. Given these exclusion criteria, 136 patients with 163 lesions were included in the analysis. This study complied with the Declaration of Helsinki, and all subjects provided informed consent.

Angiographic Measurements

A 5-Fr or 6-Fr guiding catheter (without side holes) was used to selectively cannulate the ostium of the target coronary artery. All captured angiographic images were analyzed, and the stenosis degree was visualized according to the American Heart Association classification system18 and offline QCA (QAngio XA 7.2; Medis Medical Imaging Systems, Leiden, the Netherlands) by 2 experienced observers who were unaware of the FFR determination results. The minimal lumen diameter (MLD), distal reference, proximal reference, reference diameter (RD), percentage of diameter stenosis (%DS), and lesion length with the least foreshortening view were measured using standard techniques.19 The coronary lesions were then classified into 2 groups according to the QCA data results as follows: (1) intermediate stenosis (group I): 40% to 70% stenosis; (2) significant stenosis (group S): ≥70% stenosis.

FFR Measurements

After diagnostic coronary angiography was performed, a 5-Fr or 6-Fr guiding catheter without side holes was used to engage the coronary artery. A 3000-U bolus dose of heparin was intravenously administered. A 0.014-in sensor-tipped, high-fidelity pressure wire (Certus; Radi Medical, Uppsala, Sweden) was set at 0, calibrated, advanced through the guiding catheter, and positioned distal to the site of stenosis as described previously.6 After a 100- to 200-µg dose of nitroglycerin was injected intracoronary to prevent vasospasm, adenosine (140 µg·kg1 ·min 1) was administered intravenously to induce maximum hyperemia. During offline analysis, FFR was calculated as the ratio of the mean distal intracoronary pressure to the mean aortic pressure at the time of peak hyperemia as determined by the maximum trans-stenotic gradient. An FFR value <0.80 was considered hemodynamically significant.

Statistical Analysis

In the present study, numerical variables were presented as mean ± standard deviation values, and categorical variables were presented as numbers and percentages. The χ2 and Fisher exact tests were used to analyze categorical variables; the unpaired Student t test and Wilcoxon rank sum test were used to analyze continuous variables. Angiographic characteristics and FFR data were compared between groups with intermediate and significant lesions. The Spearman correlation coefficient between the QCA and FFR results was then calculated for both groups. The area under the receiver operating characteristic (ROC) curve was also estimated for the best cutoff value as a predictor of FFR values of <0.80. All statistical analyses were performed using JMP statistical software for Mac version 9.0.2 (SAS Institute Inc., Cary, NC). P values <0.05 were considered statistically significant.

Results

This study included 163 lesions in 136 patients. QCA analysis results categorized 107 lesions (65.6%) into group I and 56 lesions (34.4%) into group S. Of the 163 lesions, 88 (54.0%) were below the ischemic threshold (FFR value <0.80). Patient and lesion characteristics, including the QCA and FFR determination results, are summarized in Table 1. There were no differences between the 2 groups in baseline clinical or lesion characteristics obtained via diagnostic catheterization. QCA and FFR determination revealed that the mean FFR value for all lesions was 0.76 ± 0.16. The mean lesion length was significantly greater in group I than in group S (13.72 ± 7.99 mm vs 10.15 ± 5.94 mm, P = 0.004), whereas the mean FFR value was significantly lower in group S than in group I (0.61 ± 0.16 vs 0.83 ± 0.08, P < 0.0001).

Table 1. Patient and Lesion Characteristics
 Total N=163Group I n=107Group S n=56P Value (I vs S)
  1. Abbreviations: %DS, percentage diameter of stenosis; ACC, American College of Cardiology, AHA, American Heart Association; FFR, fractional flow reserve; CKD, chronic kidney disease; LAD, left anterior descending artery; LCx, left circumflex artery; LMCA, left main coronary artery; LVEF, left ventricular ejection fraction; MLD, minimal lumen diameter; PCI, percutaneous coronary intervention; QCA, quantitative coronary angiography; RCA, right coronary artery; RD, reference diameter.

  2. Values are mean ± standard deviation or number (percentage).

Age, y69.0 ± 9.169.0 ± 9.468.8 ± 8.60.90
Men, n (%)126 (77.3)83 (77.6)43 (76.8)0.91
Body mass index23.7 ± 3.523.3 ± 3.524.3 ± 3.60.09
Hypertension, n (%)124 (76.1)80 (74.8)44 (78.6)0.59
Diabetes, n (%)99 (60.7)62 (57.9)37 (66.1)0.31
Dyslipidemia, n (%)99 (60.7)64 (59.8)35 (62.5)0.74
Current smoker, n (%)78 (47.9)51 (47.7)27 (48.2)0.95
Family history, n (%)31 (19.0)20 (18.7)11 (19.6)0.88
CKD, n (%)29 (17.8)19 (17.8)10 (17.9)0.99
Previous PCI, n (%)51 (31.3)36 (33.6)15 (26.8)0.37
LVEF, %56.5 ± 7.556.4 ± 7.456.7 ± 7.60.77
Target vessel, n (%)   0.66
 LAD95 (58.3)64 (60.8)31 (55.4) 
 LCx33 (20.3)20 (18.7)13 (23.2) 
 RCA33 (20.3)21 (19.6)12 (21.4) 
 LMCA2 (1.2)2 (1.9)0 (0)  
ACC/AHA classification
 B2+C, n (%)66 (40.5)47 (43.9)19 (33.9)0.22
QCA and FFR    
 Proximal RD (mm)2.89 ± 0.672.92 ± 0.702.83 ± 0.600.39
 Distal RD (mm)2.62 ± 0.652.57 ± 0.672.70 ± 0.620.25
 RD (mm)2.85 ± 2.522.91 ± 3.082.73 ± 0.600.68
 MLD (mm)1.00 ± 0.431.24 ± 0.300.55 ± 0.25<0.0001
 %DS (%)61.62 ± 15.1151.98 ± 6.9980.04 ± 7.23<0.0001
 Lesion length (mm)12.49 ± 7.5213.72 ± 7.9910.15 ± 5.940.004
 FFR0.76 ± 0.160.83 ± 0.080.61 ± 0.16<0.0001

Relationship Between QCA Data and FFR

Overall, although there were significant correlations between diameter stenosis (DS) (visual estimation), %DS, MLD, lesion length, and FFR, the correlation between FFR and lesion length was weak compared with that between FFR and each of the other parameters. In group S, although %DS and MLD were significantly correlated with FFR (r = −0.83, P < 0.0001, and r = 0.75, P < 0.0001, respectively), lesion length was not correlated with FFR (r = −0.20, P = 0.13). In contrast, in group I, %DS was not correlated with FFR (r = −0.06, P = 0.55). MLD was weakly correlated with FFR (r = 0.43, P < 0.0001), whereas lesion length was significantly correlated with FFR (r = −0.79, P < 0.0001; Figure 1). The relationship between FFR and each angiographic parameter is summarized in Table 2.

Figure 1.

Correlation between fractional flow reserve (FFR) and angiographic parameters. Overall, FFR was inversely correlated with percentage of diameter stenosis (%DS) (A), whereas the correlation between FFR and lesion length was weak (B). In group S, %DS was significantly correlated with FFR (C), whereas lesion length was not correlated with FFR (D). In group I, %DS was not correlated with FFR (E), whereas lesion length was significantly correlated with FFR (F).

Table 2. Correlation Between Fractional Flow Reserve and Angiographic Parameters
 rt ValueP Value
  1. Abbreviations: %DS, percentage diameter of stenosis; DS, diameter of stenosis; MLD, minimal lumen diameter.

Total   
 DS (visual)−0.75−14.05<0.0001
 %DS−0.74−14.11<0.0001
 MLD0.7614.94<0.0001
 Lesion length−0.19−2.530.01
Group I   
 DS (visual)−0.08−0.460.65
 %DS−0.06−0.600.55
 MLD0.434.84<0.0001
 Lesion length−0.79−13.14<0.0001
Group S   
 DS (visual)−0.66−6.39<0.0001
 %DS−0.83−10.87<0.0001
 MLD0.758.28<0.0001
 Lesion length−0.20−1.530.13

Optimal Angiographic Parameter for Predicting an FFR Value of <0.80

Using ROC analysis of the overall cohort of 163 lesions, MLD was the best parameter for predicting an FFR value of <0.80 (area under the curve [AUC], 0.87; cutoff value, 1.0 mm; sensitivity, 78.4%; specificity, 85.3%; positive predictive value [PPV], 86.3%; negative predictive value [NPV], 77.1%). In group S, %DS was the best parameter for predicting an FFR value <0.80 (AUC, 0.87; cutoff value, 72.9%; sensitivity, 84.9%; specificity, 100%; PPV, 100%; NPV, 27.3%). In group I, however, lesion length was the best parameter for predicting an FFR value <0.80 (AUC, 0.94). A lesion length of 16.1 mm was identified as the best cutoff value for predicting a categorized FFR cutoff value of 0.80 (sensitivity, 85.7%; specificity, 94.4%; PPV, 88.2%; NPV, 93.2%; Figure 2). The ROC analysis for predicting an FFR value <0.80 is shown in Table 3.

Figure 2.

Receiver operating characteristic (ROC) curve analysis for predicting a fractional flow reserve (FFR) value <0.80. (A) ROC analysis of percentage of diameter stenosis (%DS) for predicting an overall FFR value <0.80. (B) ROC analysis of lesion length for predicting an overall FFR value <0.80. (C) ROC analysis of %DS for predicting a group I FFR value <0.80. (D) ROC analysis of lesion length for predicting FFR of <0.80 in group I.

Table 3. Receiver Operating Characteristic Analysis for Predicting Fractional Flow Reserve <0.80
 AUCCutoff ValueSensitivity (%)Specificity (%)PPV (%)NPV (%)
  1. Abbreviations: %DS, percentage diameter of stenosis; AUC, area under the curve; LL, lesion length; MLD, minimal lumen diameter; NPV, negative predictive value; PPV, positive predictive value.

Total      
 %DS0.7865.74%64.893.391.969.3
 MLD0.871.02 mm78.485.386.377.1
 LL0.6816.58 mm42.196.092.558.5
Group I      
 %DS0.5448.99%51.469.445.074.6
 MLD0.741.02 mm48.688.968.078.0
 LL0.9416.12 mm85.794.488.293.2
Group S      
 %DS0.8772.9%84.910010027.3
 MLD0.740.55 mm56.610010011.5
 LL0.6310.35 mm30.21001007.5

Discussion

The present study showed that the degree of stenosis has significant functional importance for myocardial ischemia in angiographically detected severe stenotic lesions, and lesion length has a significant effect on intermediate stenosis. In particular, a lesion length >16 mm was the best cutoff value for predicting an FFR value <0.80 for intermediate lesions; this criterion had good sensitivity (86%), specificity (94%), PPV (88%), and NPV (93%).

To date, only a few studies have focused on lesion length in coronary stenosis. Takayama and Hodgson20 reported that lesion length (measured by 3-dimensional [3D] IVUS) had a positive correlation with pressure gradient in 17 lesions, and minimal lumen area (MLA) or lesion length measured by 3D IVUS was the only significant independent determinant of FFR. However, the sample size of the study was relatively small (17 lesions) to observe the relationship between lesion length and FFR. Briguori et al21 reported that lesion length measured by IVUS had a weak inverse correlation with FFR (r = −0.41) in 53 intermediate lesions. ROC analysis revealed that a 10-mm lesion length was the best IVUS cutoff value for predicting an FFR value <0.75. This IVUS cutoff value had high specificity (80%) but low sensitivity (41%). Brosh et al22 focused on lesion length in 63 angiographically determined intermediate lesions and found a weak inverse correlation between lesion length and FFR (r = −0.31) and a moderate correlation between %DS and FFR (r = −0.55). The correlation between %DS and FFR improved only when the lesion length was >10 mm (r = −0.78). The Intravascular Ultrasound Diagnostic Evaluation of Atherosclerosis in Singapore study analyzed small coronary arteries (RD <3 mm),23 and demonstrated that a lesion length >20 mm, in addition to an MLA <2 mm and a plaque burden >80% as measured by IVUS, were the best cutoff values for predicting an FFR value <0.75.

A recent study suggested that lesion length with a lumen area <3.0 mm2 (measured by IVUS) correlates weakly with FFR (r = −0.47), and that 3.1 mm was the best cutoff value of lesion length with a lumen area <3.0 mm2 to predict an FFR value <0.80.24 Lopez-Palop et al25 focused on lesion length in 106 angiographically determined intermediate-long lesions (>20 mm). That study revealed a moderate inverse correlation between lesion length and FFR (r = 0.63). ROC analysis determined that the best cutoff value for predicting an FFR value >0.80 was a lesion length of 27.1 mm. Their study included 65 patients (63.1%) with ACS in whom the microvascular bed in the infarct zone may not have been uniform or constant or may have had minimal resistance.26 Therefore, due to the theoretical limitations of FFR for ACS lesions, we excluded ACS lesions from the present study. The studies mentioned above included only intermediate coronary lesions, whereas in this study we investigated both angiographically determined severe stenoses and intermediate lesions.

Although the earlier studies adopted either 0.75 or 0.80 as an FFR cutoff value for myocardial ischemia, the current study used an ischemic FFR cutoff value of 0.80. An FFR value <0.75 has been associated with ischemia in stress testing by numerous comparative studies with a sensitivity of 88%, specificity of 100%, PPV of 100%, and overall accuracy of 93%. However, an FFR value >0.80 is associated with negative ischemic testing with a predictive accuracy of 95%.26 An FFR value between 0.75 and 0.80 is in the gray zone and requires clinical judgment. We conservatively considered the upper limit of the small transition zone to limit the number of potentially ischemic lesions left untreated as in the Fractional Flow Reserve versus Angiography for Multivessel Evaluation (FAME) study.9,10,16

As defined by Poiseuille's law of fluid dynamics, pressure gradient is influenced by coronary blood flow and viscosity, minimum radius, and lesion length. In effect, the pressure gradient is inversely proportional to the fourth power of the lesion radius (r4) and is proportional to lesion length.27 The current findings are consistent with this equation.

The current study showed that a 16-mm lesion length cutoff value is a highly accurate indicator of functional coronary ischemia in intermediate coronary lesions. When angiography reveals a severe coronary stenosis (73% in our study) in a subject with suspected CAD, myocardial ischemia may be diagnosed with high probability. For intermediate coronary stenosis, however, assessment of the functional significance of this lesion is required in most cases to enable decision making about revascularization. We suggest that physiological assessment should be performed according to lesion length in intermediate coronary lesions.

Study Limitations

This study has several limitations. First, the number of included subjects was relatively small, particularly in the group with significant stenosis (n = 56). In daily practice, FFR is often performed for angiographically detected intermediate lesions rather than significant stenoses to enable revascularization decision-making. Second, although we excluded tandem lesions from this study, the degree of stenosis is not entirely homogeneous in most coronary lesions. Third, with regard to QCA, visual assessment of DS leads to higher values than QCA assessment in many cases. As such, intermediate stenosis values may differ from visual assessment values. Finally, because this study did not include IVUS data, we could not assess the relationship between FFR and IVUS data such as true vessel size, stenosis area, or target lesion plaque volume.

Conclusion

Lesion length is relevant to the assessment of the physiological significance of intermediate-grade coronary lesions. If the lesions are diffuse with intermediate stenosis, physiological methods should be used to evaluate myocardial ischemia.

Ancillary