Corrected Bifurcation Angle versus Distal Bifurcation Angle for Prediction of Side Branch Compromise after Provisional Bifurcation Coronary Intervention View PDF

Bifurcation coronary artery disease has a significant burden among coronary intervention cases [1]. Being one of the challenging complex procedures, ongoing studies aim to improve immediate and long term outcomes after bifurcational percutaneous coronary intervention (PCI) [2]. The provisional main vessel (MV) PCI is recommended over the two stent strategy but with the risk of side branch (SB) compromise [3].

Many studies had discussed different strategies to minimize the risk of SB compromise during bifurcational PCI such as; jailed wire and jailed balloon strategies [4,5]. Other studies had discussed the angiographic predictors of SB compromise after MV stenting for better selection of intervention strategy [6].

Among the predictors of SB compromise, distal bifurcation angle (BA), the angle between SB and distal MV, was presented as a reliable predictor [7]. However, studies results were contradicted. Some studies demonstrated that narrow distal BA had increased the risk of SB compromise [8]. More recent studies have concluded that the wider the distal BA the higher the risk of SB compromise [9].

This debate could be due to focusing on the distal BA and neglecting the proximal BA [6,10]. MV is not always a straight vessel. Based on three dimensions Quantitative coronary angiography studies, there are two models of coronary bifurcation; Y and T models [11]. Furthermore, a computed tomography (CT) coronary angiography study demonstrated that the MV could change its curvature at the origin of the SB [12].

Correction of the distal bifurcation angle could be achieved by adding proximal BA. Corrected BA, which is the sum of proximal and distal bifurcation angles, could be a potential predictor of SB compromise during provisional bifurcational PCI. This study aims to compare between corrected BA and distal BA in terms of prediction of SB compromise during provisional bifurcational PCI.

The institutional review board approved this study in Zagazig University Hospitals, Sharkia, Egypt.

Study Design, Setting, and Duration

This prospective cohort study was conducted in Zagazig university hospitals. The study started in March 2019 and ended in March 2020.

Study Population

The study involved Patients with coronary bifurcation lesion and a SB diameter ≥ 1.5 mm who underwent provisional MV PCI.

According to the European Bifurcation Club guidelines, Coronary bifurcation disease was defined as SB involvement in MV lesion with absence of normal area between MV minimal luminal diameter and SB, true coronary bifurcation lesion was defined as involvement of SB ostium with SB diameter being ≥ 2 mm, and significant SB was defined as any SB you do not want to lose [13].

Patients with chronic total occlusions and in stent restenosis were excluded. Also, we excluded procedures that started with SB stenting.

Study Variables, Clinical Assessment

Clinical assessment of all patients included age, gender, cardiovascular risk factors, coronary angiography indication, pulse, blood pressure, and serum Creatinine level.

In terms of an observational study, operators planned PCI procedures and peri-procedure medications according to current guidelines without our interference. Loading doses with 300 mg aspirin and 600 mg clopidogrel were given to stable coronary artery disease patients 12 hrs. before PCI. A loading dose with 180 mg ticagrelor replaced the clopidogrel in acute coronary syndrome (ACS) patients in absence of contraindications.

Two separate cardiologists analyzed coronary angiography films using RadiAnt DICOM Viewer 5.0.1 (64-bit) program. Bifurcation angles were defined as following (Figure 1):

• Distal BA was defines as the angle between distal MV and SB
• Proximal BA was defined as the angle between proximal MV and SB.
• Corrected BA was defined as the sum of proximal and distal bifurcation angles.

We calculated of the bifurcation angles by two dimensions (2D) quantitative coronary analysis (QCA) in the view showing the widest angle. Horizontal left main bifurcation angles were calculated in the spider view. However, Vertical left main bifurcation angles were calculated in the left anterior oblique (LAO) view with cranial angulation. We calculated proximal left anterior descending (LAD) / diagonal (D) bifurcation angles in the spider view and the mid to distal LAD/D bifurcation angles in the LAO view with cranial angulation. We calculated left circumflex (LCX)/obtuse marginal (OM) bifurcation angles in either spider view or right anterior oblique (RAO) view with caudal angulation. The posterior descending artery (PDA) / postro-lateral (PL) branch bifurcation angles were calculated in either postro-anterior (PA) view with cranial angulation or LAO view with cranial angulation. If the bifurcation angle was foreshortened in the standard view, views were modified to get the widest angle.

We measured the bifurcation angle between the central axes of the two vessels to avoid the bias that could be made by plaque irregularity. In case of vessel tortuosity, the angle was measured at the level of bifurcation core.

We checked the coronary angiograms for other potential angiographic predictors of SB compromise that could be confounding factors as coronary dominance, site of bifurcation, Medina classification, angulation, calcification, Plaque irregularity, lesion thrombus, and both MV and SB TIMI flow grade. Guiding catheter diameters were used to correct vessels diameters as following; 2 mm for 6 French catheter and 2.3 mm for 7 French catheter. Radio-opaque length of coronary wires was used to correct the length of lesions. Stenosis percentage was calculated as following; ([reference diameter-minimal diameter]/reference diameter) ×100.  

Furthermore, PCI procedure was observed for potential steps that impact the risk of SB compromise as leaving jailed wire in SB, pre-dilatation of MV, pre-dilatation of SB, any observed dissections after pre-dilatation, stent diameter/distal MV diameter ratio, and proximal optimization technique (POT).

All patients were followed during PCI procedure case by case after completed analysis of coronary angiography film. The SB was considered compromised after the MV stent deployment when the SB TIMI flow grade decreased.

In order to minimize the risk of bias, we performed the following steps:

• Our study involved all patients with bifurcation coronary lesions who underwent provisional PCI during the study period to avoid selection bias.
• All clinical data were collected from each patient before PCI to avoid missed data in medical records.
• Two separate cardiologists analyzed the coronary angiography film before start of PCI to decrease observer bias. SB compromise was checked during PCI procedure as cine recordings could be lost.

Sample Size Calculation

Previous studies demonstrated that the area under the curve of the distal BA was 0.655 for prediction of SB compromise. Therefore, the minimum sample size required was estimated by SciStat online calculator to be 106 cases to achieve 80% power and 95% confidence interval.

Statistical Analysis

Statistical analysis was performed using the Statistical Package for Social ‎Sciences version 16.0 (SPSS for Windows 16.0, Inc., Chicago, IL, USA). Shapiro wilk test was used to test the quantitative variables for normal distribution. Non parametric Mann-Whitney test or student t-test was used to compare between the studies groups according to the type of data. A chi-square test was used to compare between groups as regarding qualitative variables.

We expressed continuous parameters as mean and standard deviation. Categorical data were presented as the absolute number and percentage within ‎brackets. We used Receiver Operating Characteristic (ROC) curve to calculate area under the curve for both distal and corrected bifurcation angles. Multivariate analysis was used to test the predictive power of bifurcation angles taking other confounding factors in consideration. All P values were based on a 2-tailed distribution; P value ≥ 0.05 was considered non-significant, and P value < 0.05 was considered Significant.

Characteristics of the Study Population

This study rolled in 185 patients who underwent provisional PCI for bifurcational coronary artery disease. Patients were divided according to corrected BA into two groups; straight bifurcation model group which involved 73 patients with corrected BA = 180°, and wide bifurcation model group which involved 112 patients with corrected BA > 180°.

As regarding the basic characteristics, the two-group comparison did not show a significant difference in most variables. However, Patents in the straight bifurcation model had more frequent dyslipidemia (P = 0.009) and prior coronary artery bypass graft (CABG) (P = 0.023).Most patients in both groups presented with stable angina. However, the wide bifurcation model involved more patients with non-ST segment elevation myocardial infarction (NSTEMI) and the straight bifurcation model involved more patients with ST segment elevation myocardial infarction (STEMI) (P = 0.037) (Table 1).

Table 1: Baseline characteristics.

Where: *: Mann Whitney test; **: chi-square test; +: Fisher exact test; NSTE-ACS: non ST elevation acute coronary syndrome; PCI: percutaneous coronary intervention; CABG: coronary artery bypass graft; BPM: beat per minute; NSTEMI: Non ST elevation myocardial infarction; STEMI: ST elevation myocardial infarction and the values are presented as n (%) or mean ± standard deviation.

Bifurcation site analysis demonstrated that LAD/D bifurcations were more frequent in both groups. The straight bifurcation model involved more patients with LCX/OM and PDA/PL bifurcations. However, LM bifurcation was absent in the straight bifurcation model (P = 0.002). Other qualitative coronary angiography measures are presented in details in table 2.

Table 2: Qualitative coronary angiography parameters.

Where: *: Mann Whitney test; **: chi-square test; +: Fisher exact test; TIMI: thrombolysis in myocardial infarction; LAD/D: left anterior descending artery/diagonal bifurcation; LCX/OM: left circumflex/obtuse marginal bifurcation; LM: left main bifurcation; PDA/PL: posterior descending artery/postrolateral artery bifurcation; PCI: percutaneous coronary intervention and the values are presented as n (%) or mean ± standard deviation.

As regarding quantitative parameters of coronary angiography, the corrected BA was 180° for the straight bifurcation model and the mean corrected BA was 195.3° ± 8.2° for the wide bifurcation model. Also, the mean distal BA was 53.5° ± 18.5° for the straight bifurcation model and 59.6° ± 19.4° for the wide bifurcation model. Furthermore, mean proximal BA was 126.2° ± 18.2° for the narrow bifurcation model and 134.8° ± 20.6° for the wide bifurcation model.

Difference between the Study Groups

The statistical analysis did not reveal any significant difference between two groups as regarding qualitative coronary angiography measures or PCI procedure parameters including SB wiring, pre-dilatation, stent diameter/distal MV diameter ratio, and POT. The only reported significant difference was at level of distal MV lesion length which was longer in the wide bifurcation model (P = 0.029). Details of quantitative coronary angiography parameters and PCI procedure parameters are presented in table 3 and table 4 respectively.

Table 3: Quantitative coronary angiography parameters.

Where: *: Mann Whitney test; mm: Millimetre; Values are presented as mean ± standard deviation.

Table 4: PCI procedure and outcome parameters.

Where: *: Mann Whitney test; **: chi-square test; +: Fisher exact test; TIMI: thrombolysis in myocardial infarction; PCI: percutaneous coronary intervention and the values are presented as n (%) or mean ± standard deviation.

The Study Outcome

The incidence of SB compromise was substantially higher in the straight bifurcation model as compared to the wide bifurcation model group (52.1% vs. 15.2%; P < 0.001) (Figure 2). The frequency of total occlusion of the SB with TIMI flow grade 0 was higher in the straight bifurcation model (11% vs. 3.6%; P <0.001).

Comparison between the Distal and the Corrected BAs

ROC curve was constructed for both the distal and the corrected BA as predictors of the SB compromise (Figure 3).

As regarding the distal BA, the area under the curve was 0.580 with 95% CI 0.495-0.666 (P = 0.084).  The best cut off value of the distal BA was ≤ 64° with sensitivity 70.3% (95% CI 58.5%-80.3%), specificity 52.7% (95% CI 43.0%-62.2%), positive predictive value 49.5% (95% CI 43.4%-55.6%), negative predictive value 72.8% (95% CI 64.4%-79.9%), and accuracy 59.7% (95% CI 52.3%-66.8%).

On the other hand, the area under the curve for the corrected BA was 0.711 with 95% CI 0.623-0.800 (P < 0.001). For prediction of the SB compromise, the corrected BA (180 ?) had a sensitivity of 71.6% (95% CI 59.9%-81.5%), specificity of 73.2% (95% CI 64.0%-81.1%), positive predictive value of 63.9% (95% CI 55.8%-71.2%), negative predictive value of 79.6% (95% CI 72.8%-85.1%), and accuracy of 72.6% (95% CI 65.6%-78.9).

Comparison between the two ROC curves was performed revealing that the area under the curve of the corrected BA was significantly higher than that of the distal BA (P = 0.023). Furthermore, the diagnostic accuracy of the corrected BA was better than that of the distal BA.

Multivariate Analysis

Step wise approach logistic regression analysis revealed that the corrected BA was among the independent predictors of SB compromise after MV stent deployment. The independent predictors were the corrected BA, Ostial SB stenosis percentage before PCI, stent diameter/distal MV diameter ratio, and jailed wire in SB, lesion thrombus. The distal BA was not identified among the best predictors of SB compromise (Table 5).

Table 5: Multivariate analysis using step wise approach logistic regression analysis to pick up independent predictors of side branch compromise.

Where: CI: Confidence interval.

Importance of the Study

In clinical practice, Coronary bifurcation coronary artery disease is frequent scenario requiring PCI [1]. Current guidelines recommend provisional MV PCI over two stent strategy in terms of reduction of target lesion failure and cardiovascular mortality [3,14, and 15]. Therefore, it is better to predict the risk of SB compromise before proceeding with provisional PCI. The relation between BA and fate of SB during provisional PCI is a current hot debate.

Summary of Key Findings

Our study demonstrated that bifurcation lesions in the straight bifurcation model had a higher incidence of SB compromise in comparison with lesions in the wide bifurcation model (P < 0.001). The corrected BA was better that the distal BA in prediction of SB compromise after provisional PCI with statistically significant difference (P = 023). Furthermore, the corrected BA was among the best independent predictors of SB compromise in association with high SB ostial stenosis percentage prior to PCI, stent diameter/distal MV diameter ratio > 1, absence of jailed wire in SB, and presence of lesion thrombus.

Previous Research

Previous studies demonstrated that the risk of SB compromise had increased with narrow distal BA [8,10,16, and 17]. Our study was concordant with these studies and the risk of carina shift or plaque shift with narrow distal BA could explain such finding [18]. Other equivocal studies, including intravascular ultrasound (IVUS) guided studies, did not demonstrate any association between the distal BA and the incidence of SB compromise [19-21].

On the other hand, more recent studies concluded that the risk of SB compromise had increased with wider distal BAs [6,7, and 9]. They explained such finding with the Pressure drop in SB with wider distal BA [22], the circular shape of SB ostium in the wide distal BA with subsequent small ostium circular area which was believed to be a sensitive predictor of SB compromise [2,23], and the increase in bifurcation core plaque burden with the wide distal BA due to decrease in the wall shear stress [24-27].

The impact of corrected BA on the risk of SB compromise during provisional PCI is not studied yet. Our study demonstrated that corrected BA was among the independent predictors of SB compromise. Previous research studied only the distal BA considering that the MV is always a straight vessel [6,7]. However, MV could change its angle at site of bifurcation. In real life, we have two models of coronary bifurcation; model A that involves wide corrected BA when the sum of distal and proximal bifurcation angles is more than 180°, and model B that involves straight bifurcation model with corrected BA ≤ 180° (Figure 4). In our study, the risk of SB compromise was significantly high in patients with straight bifurcation model. We could explain such new result with the direction of stent strata opening. Theoretically, the direction of stent strata opening could open the SB otium in model A and could occlude it in model B.

To the best of our knowledge, we are the first study to compare between the corrected BA and the distal BA regarding the risk of SB compromise. The area under the curve of the corrected BA was better than that of the distal BA with statistically significant difference. 

Many studies confirmed that stent oversize in the distal MV could increase the risk of carina shift and SB compromise which is concordant to our study [28]. However, Zhang et al study reported that stent diameter/distal MV reference diameter did not affect the fate of SB [6]. The importance of Jailed wire was demonstrated by many studies to decrease the risk of carina shift, to decrease the risk of SB compromise, and to be a marker for the re-cross if SB was compromised [3,29, and 30]. Our study was concordant with such finding as non-wiring of the SB was among the best predictors of SB compromise. However, our study was discordant with many studies that reported that jailed wire in SB did not impact the fate of SB [6,20].

In our study, the increased pre-procedure ostial SB stenosis percentage was one of the predictors of SB compromise which is concordant with many studies that proposed bifurcation core stenosis and SB stenosis as independent predictors of SB compromise [9,20].

Many studies had recommended ACS as a strong predictor of SB compromise [20]. Furthermore, IVUS guided studies concluded that unstable plaques at bifurcation lesions are more liable for SB compromise [31]. This is concordant with our study that demonstrated that presence of lesion thrombus was one of the best predictors of SB compromise.

Our study has several strength points:

• The study was a prospective cohort study.
• Bifurcation angles were checked by two separate cardiologists to minimize observing bias.
• Bifurcation angles were measured between the central axes of the vessels to minimize the impact of plaque irregularity on the measurement accuracy.
• Follow up was performed for all cases during the PCI procedure to avoid missing recordings. However, the study was observation in nature with small sample size. Furthermore, a dedicated bifurcation quantitative coronary angiography was not available.

We are in unmet need for future study with a larger sample size and longer follow up duration in order to test the impact of the corrected BA on immediate and long term angiographic and clinical outcomes.

Straight bifurcation model with corrected BA = 180° is one of the independent predictors of SB compromise after provisional bifurcational intervention. Compared to the distal BA, the corrected BA presented a better predictive accuracy with statistically significant difference. We recommend a future study with a larger sample size and longer follow up duration in order to test the impact of the corrected BA on the immediate and the long term angiographic and clinical outcomes.

1. Tsuchida K, Colombo A, Lefevre T, Oldroyd KG, Guetta V, et al. (2007) The clinical outcome of percutaneous treatment of bifurcation lesions in multivessel coronary artery disease with the sirolimus-eluting stent: insights from the Arterial Revascularization Therapies Study part II (ARTS II). Eur Heart J 28: 433-442.
2. Louvard Y, Lefevre T, Morice MC (2004) Percutaneous coronary intervention for bifurcation coronary disease. Heart 90: 713-722.
3. Banning AP, Lassen JF, Burzotta F, Lefevre T, Darremont O, et al. (2019) Percutaneous coronary intervention for obstructive bifurcation lesions: the 14th consensus document from the European Bifurcation Club. EuroIntervention 15: 90-98.
4. Oda H, Tsuchida K, Okamura K, Ozaki K, Takahashi K, et al. (2010) Two-wire protection of side branches to prevent side branch occlusion during stent implantation for bifurcational lesions. Cardiovasc Interv Ther 25: 112-116.
5. Shishido K, Moriyama N, Hayashi T, Yokota S, Miyashita H, et al. (2020) The efficacy of modified jailed balloon technique for true bifurcation lesions. Catheter Cardiovasc Interv 2020.
6. Zhang D, Xu B, Yin D, Li Y, He Y, et al. (2015) How bifurcation angle impacts the fate of side branch after main vessel stenting: a retrospective analysis of 1,200 consecutive bifurcation lesions in a single center. Catheter Cardiovasc Interv 85: 706-715.
7. Zhang D, Xu B, Yin D, Li YP, He Y, et al. (2015) Clinical and angiographic predictors of major side branch occlusion after main vessel stenting in coronary bifurcation lesions. Chin Med J (Engl) 128: 1471-1478.
8. Gil RJ, Vassilev D, Rzezak J, Slysz A, Krzyzewski R, et al. (2009) Influence of parent vessel - side branch distal angle on restenosis rates after main vessel stenting in coronary bifurcation lesion. Kardiol Pol 67: 36-43.
9. Dou K, Zhang D, Xu B, Yang Y, Yin D, et al. (2015) An angiographic tool for risk prediction of side branch occlusion in coronary bifurcation intervention: the RESOLVE score system (Risk prEdiction of Side branch OccLusion in coronary bifurcation interVEntion). JACC Cardiovasc Interv 8: 39-46.
10. Gil RJ, Vassilev D, Formuszewicz R, Rusicka-Piekarz T, Doganov A (2009) The carina angle-new geometrical parameter associated with periprocedural side branch compromise and the long-term results in coronary bifurcation lesions with main vessel stenting only. J Interv Cardiol 22: E1-E10.
11. Reiber JH, Tu S, Tuinenburg JC, Koning G, Janssen JP, et al. (2011) QCA, IVUS and OCT in interventional cardiology in 2011. Cardiovasc Diagn Ther 1: 57-70.
12. Medrano-Gracia P, Ormiston J, Webster M, Beier S, Ellis C, et al. (2017) A study of coronary bifurcation shape in a normal population. J Cardiovasc Transl Res 10: 82-90.
13. Louvard Y, Medina A (2015) Definitions and classifications of bifurcation lesions and treatment. EuroIntervention 11: V23-V26.
14. Ford TJ, McCartney P, Corcoran D, Collison D, Hennigan B, et al. (2018) Single?versus 2?stent strategies for coronary bifurcation lesions: A systematic review and meta?analysis of randomized trials with long?term follow?up. J Am Heart Assoc 7: e008730.
15. Zhou Y, Chen S, Huang L, Hildick-Smith D, Ferenc M, et al. (2018) Definite stent thrombosis after drug-eluting stent implantation in coronary bifurcation lesions: A meta-analysis of 3,107 patients from 14 randomized trials. Catheter Cardiovasc Interv 92: 680-691.
16. Vassilev D, Gil R (2008) Clinical verification of a theory for predicting side branch stenosis after main vessel stenting in coronary bifurcation lesions. J Interv Cardiol 21 :493-503.
17. Vassilev D, Gil RJ, Koo BK, Gibson CM, Nguyen T, et al. (2012) The determinants of side branch compromise after main vessel stenting in coronary bifurcation lesions. Kardiol Pol 70: 989-997.
18. Koo BK, Waseda K, Kang HJ, Kim HS, Nam CW, et al. (2010) Anatomic and functional evaluation of bifurcation lesions undergoing percutaneous coronary intervention. Circ Cardiovasc Interv 3: 113-119.
19. Yoshitaka Goto Y, Kawasaki T, Koga N, Tanaka H, Koga H, et al. (2012) Plaque distribution patterns in left main trunk bifurcations: prediction of branch vessel compromise by multidetector row computed topography after percutaneous coronary intervention. EuroIntervention 8: 708-716.
20. Hahn JY, Chun WJ, Kim JH, Song YB, Oh JH, et al. (2013) Predictors and outcomes of side branch occlusion after main vessel stenting in coronary bifurcation lesions: results from the COBIS II Registry (COronary BIfurcation Stenting). J Am Coll Cardiol 62: 1654-1659.
21. Xu J, Hahn JY, Song YB, Choi SH, Choi JH, et al. (2012) Carina shift versus plaque shift for aggravation of side branch ostial stenosis in bifurcation lesions: volumetric intravascular ultrasound analysis of both branches. Circ Cardiovasc Interv 5: 657-662.
22. Sayed Razavi M, Shirani E (2013) Development of a general method for designing microvascular networks using distribution of wall shear stress. J Biomech 46: 2303-2309.
23. Kang SJ, Mintz GS, Kim WJ, Lee JY, Park DW, et al. (2011) Preintervention angiographic and intravascular ultrasound predictors for side branch compromise after a single-stent crossover technique. Am J Cardiol 107: 1787-1793.
24. Huo Y, Finet G, Lefevre T, Louvard Y, Moussa I, et al. (2012) Optimal diameter of diseased bifurcation segment: a practical rule for percutaneous coronary intervention. EuroIntervention 7: 1310-1316.
25. Kimura BJ, Russo RJ, Bhargava V, McDaniel MB, Peterson KL, et al. (1996) Atheroma morphology and distribution in proximal left anterior descending coronary artery: in vivo observations. J Am Coll Cardiol 27: 825-831.
26. Kaazempur-Mofrad MR, Isasi AG, Younis HF, Chan RC, Hinton DP, et al. (2004) Characterization of the atherosclerotic carotid bifurcation using MRI, finite element modeling, and histology. Ann Biomed Eng 32: 932-946.
27. Rodriguez-Granillo GA, Garcia-Garcia HM, Wentzel J, Valgimigli M, Tsuchida K, et al. (2006) Plaque composition and its relationship with acknowledged shear stress patterns in coronary arteries. J Am Coll Cardiol  47: 884-885.
28. Vassilev D, Gil RJ (2008) Relative dependence of diameters of branches in coronary bifurcations after stent implantation in main vessel-importance of carina position. Kardiol Pol 66: 371-378.
29. Hoye A (2017) The proximal optimisation technique for intervention of coronary bifurcations. Interv Cardiol 12: 110-115.
30. Yang JH, Lee JM, Park TK, Song YB, Hahn JY, et al. (2019) The proximal optimization technique improves clinical outcomes when treated without kissing ballooning in patients with a bifurcation lesion. Korean Circ J 49: 485-494.
31. Maehara A, Mintz GS, Bui AB, Walter OR, Castagna MT, et al. (2002) Morphologic and angiographic features of coronary plaque rupture detected by intravascular ultrasound. J Am Coll Cardiol 40: 904-910.