Author + information
- Received August 12, 2011
- Revision received October 12, 2011
- Accepted October 28, 2011
- Published online February 1, 2012.
- Salvatore Brugaletta, MD⁎,†,
- Victoria Martin-Yuste, MD†,‡,
- Teresa Padró, PhD§,
- Luis Alvarez-Contreras, MD†,
- Josep Gomez-Lara, MD⁎,
- Hector M. Garcia-Garcia, MD, PhD⁎,
- Clarissa Cola, MD‡,
- Giovanna Liuzzo, MD, PhD∥,
- Monica Masotti, MD, PhD†,
- Filippo Crea, MD, PhD∥,
- Lina Badimon, PhD§,
- Patrick W. Serruys, MD, PhD⁎ and
- Manel Sabaté, MD, PhD†,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Manel Sabaté, Department of Cardiology, Thorax Institute, Hospital Clinic, University of Barcelona, 08036 Barcelona, Spain
Objectives This study sought to assess the vascular function in patients with chronic total coronary occlusions (CTO) immediately after successful percutaneous recanalization and its relation with the pre-existing collateral circulation.
Background CTOs represent a long-acting occlusion of a coronary vessel, in which the progressively developed collateral circulation may limit ischemia and symptoms. However, it is unknown if the coronary segment distal to the occlusion has a preserved vascular function.
Methods We prospectively enrolled 19 consecutive patients, after percutaneous coronary intervention of a CTO. Luminal diameter, measured by quantitative coronary angiography, and coronary blood flow at level of epicardial coronary artery distal to the treated CTO was assessed before and after administration of acetylcholine (Ach), adenosine, and nitroglycerin (NTG). Collaterals were assessed angiographically by grading of Rentrop and of collateral connections (CC1: threadlike continuous connection; CC2: side branch–like connection).
Results Overall, Ach and adenosine caused coronary artery vasoconstriction (p = 0.001 and p = 0.004, respectively), whereas NTG failed to induce vasodilation (p = 0.084). Coronary blood flow significantly decreased with Ach (p = 0.005), significantly increased with NTG (p = 0.035), and did not change with adenosine (p = 0.470). Patients with CC2 collaterals (n = 8) had less vasoconstriction response and reduction in coronary blood flow after Ach (p = 0.005 and p = 0.008, respectively), and better vasomotor response to NTG (p = 0.029) than patients with CC1 collaterals (n = 11).
Conclusions Significant endothelial and smooth muscle dysfunction is present in the distal segments of successfully recanalized CTOs, and that seems to be more pronounced in the presence of a low grading of collateral circulation.
The endothelium is a key regulator of vascular physiology, and a damaged or dysfunctional endothelium is an initiator of vascular atherosclerosis (1,2). Endothelial dysfunction is a predictor of coronary events in patients with coronary artery disease and is also associated with short-term coronary graft performance (1,3).
Severe endothelial dysfunction has been previously demonstrated in the infarct-related occluded coronary artery during the early phase after thrombotic occlusion (4). Compared with infarct-related occlusion, chronic total occlusions (CTOs), defined as obstructions of a native coronary artery of at least 3 months' duration, with no luminal continuity and a Thrombolysis In Myocardial Infarction (TIMI) flow grade 0, have a totally different pathological substrate (sudden thrombotic occlusion vs. slow fibrocalcific occlusion) and a different pattern of collateral circulation development (5). In particular, whereas in an acute thrombotic occlusion collaterals develop suddenly or if pre-existing can be acutely recruited and may not be able to restore adequate blood requirements, in chronic occlusion, they develop gradually during the progression of a coronary lesion to a CTO, limiting ischemia and symptoms (6,7). At present, there are no available data on whether endothelial function is abnormal in the coronary segment distal to a CTO or whether the presence of collaterals is able to limit it.
The aims of our study were: 1) to assess endothelial and smooth muscle cells' function distal to a CTO immediately after successful recanalization by percutaneous coronary intervention (PCI); and 2) to explore its relationship with the pre-existing collateral circulation.
From January 2009 to December 2009, all consecutive patients, who underwent successful recanalization of a CTO by PCI, were screened in a single center and were prospectively included in the study. The duration of the occlusion was defined on the basis of a previous angiogram, the date of a prior myocardial infarction, or the onset of symptoms (5,8). All enrolled patients were symptomatic for angina and/or had documented ischemia/viability by noninvasive tests in myocardial regions supplied by the CTO (9), and a minimum of 20 mm of distal vessel amenable for endothelial function testing. For patients with silent ischemia, a functional test, showing ischemia in the myocardial territory supplied by the CTO, was required to indicate the PCI procedure. Patients with contraindications to dual antiplatelet therapy, known or suspected infections, and/or diseases that may have possible effects on inflammation were excluded. All patients received the Resolute zotarolimus-eluting stent (Medtronic, Santa Rosa, California) to cover all dilated segments. All clinical and angiographic data were prospectively recorded. All patients gave their written informed consent to participate in the study. The ethical committee of the hospital approved the protocol.
Angiographic assessment of collaterals
In all angiographic diagnostic studies, the views with the least foreshortening of the collateral connection were selected for analysis. The angiograms were analyzed as follows:
1. Collateral filling of the recipient artery was assessed according to the Rentrop classification. Briefly, Rentrop grade is categorized as follows: 0—no filling of any collateral vessels; 1—filling of side branches of the epicardial segment; 2—partial filling of the epicardial artery by collateral vessels; and 3—complete filling of the epicardial artery by collateral vessels (10) (Fig. 1).
2. The size of the collateral connection (CC) diameter was semiquantitatively assessed by 3 grades: CC0—no continuous connection between donor and recipient artery; CC1—continuous, threadlike connection; and CC2—continuous, small side branch–like size of the collateral through its course. To provide a size estimate, the collaterals were measured, as previously shown, with an electronic caliper on enlarged still images: CC1 collaterals had diameters ≤0.3 mm and CC2 ≥0.4 mm (11,12) (Fig. 1).
Two independent observers, blinded to clinical characteristics and post-PCI angiographic response to vasodilators performed this grading. In case of disagreement, a consensus was reached by a third independent observer.
Assessment of endothelial function in the CTO vessel
Long-acting oral vasoactive drugs were discontinued ≥24 h before the study. Endothelium-dependent and -independent coronary vasomotion was studied as previously described (13,14). After successful recanalization of the CTO and implantation of stent, a 0.014-inch Doppler wire with a 12-MHz piezoelectric transducer mounted on the tip (FlowWire, Volcano, Therapeutics Inc., Rancho Cordova, California) was advanced into the recanalized coronary artery, approximately 5 to 10 mm distal to the distal edge of the implanted stent. As the stents covered the entire injured segment during the PCI, the endothelial function study was performed distal to the stents in a noninjured segment. The electrocardiogram and arterial blood pressure were continuously recorded.
An infusion microcatheter (Finecross, Terumo Corporation, Tokyo, Japan) was advanced over a guidewire and placed at least 15 mm proximal to the tip of the flow wire, after which the guidewire was removed. After obtaining a stable Doppler signal, saline solution was infused through the microcatheter for 1 min to determine the baseline vasomotion tone. This was followed by a baseline angiogram, taken in the 2 orthogonal projections that best showed the artery, without the overlapping of side branches and with minimal foreshortening. From the Doppler flow velocity spectra, the average peak velocity (APV) was calculated as the time-averaged value of the instantaneous peak velocity over 2 consecutive cardiac cycles. A 2 min, selective infusion of acetylcholine (Ach) (10−6 mol/l), a 120-μg bolus of adenosine (Ade) and a 200-μg bolus of nitroglycerin (NTG) were administered. A washout period of at least 5 min between the administration of each drug was mandatory. At peak velocity, blood flow velocity in response to each vasoactive substance was measured and study angiograms in identical projections as to those performed at baseline were recorded (Fig. 2).
Offline quantitative coronary angiography analysis was performed by 1 independent and blinded observer (Cardialysis, Rotterdam, the Netherlands) using the CAAS system (Pie Medical, Maastricht, the Netherlands). Calibration of the system was based on dimensions of the guide catheter filled with contrast medium. The mean luminal diameter (average of the 2 orthogonal projections) was determined after infusion of each substance in the coronary segment distal to the edge of the stents and to the tip of the flow wire. Intraobserver variability showed a correlation coefficient (R2) of 0.97. Changes in diameter were calculated as percentage of change in mean luminal diameter compared with the baseline measurement.
We estimated the volumetric coronary blood flow (CBF) before and after administration of vasodilators according to the validated formula: cross-sectional area × APV × 0.5 (15). For the cross-sectional area of the epicardial vessel, we used the luminal diameter at baseline and after the infusion of Ach, Ade, and NTG. Changes in flow volume have been calculated as percentage of change in mean flow volume compared with the baseline measurement.
In experimental models and in humans, ACh is known to cause endothelium-dependent vessel relaxation (16,17); therefore, paradoxical vasoconstriction after its infusion was used as an indicator of endothelial dysfunction (13).
Continuous data are expressed as mean ± SD or median (interquartile range), according to their normal or not normal distribution, whereas categorical data are expressed as percentages and counts. Analysis of the normality of continuous variables was tested using the Kolmogorov-Smirnov test. Changes of parameters from baseline to post-administration of the various vasoactive drugs were evaluated by Wilcoxon test for paired data. Differences between 2 groups were analyzed by the Mann-Whitney U test. No adjustments were made for multiple comparisons. A 2-sided p value of <0.05 indicated statistical significance. Statistical analyses were performed by the use of SPSS software (version 16.0. SPSS Inc., Chicago, Illinois).
Baseline and procedural characteristics
A total of 19 patients were enrolled in the study (Fig. 3).Table 1 shows the baseline clinical and procedural characteristics. All patients were on dual antiplatelet therapy (acetylsalicylic acid + clopidogrel). Prior PCI were all performed in a non-CTO vessel, without evidence of restenosis at the time of the CTO-PCI. Patient with a left internal mammary artery on the left anterior descending artery had a CTO of the right coronary artery. Out of the 7 patients having a previous myocardial infarction, only 3 had had it in the same territory of the CTO. Of note, only 1 patient exhibited necrosis in the same region of the CTO, but with residual viability by means of functional testing.
Assessment of collateral pathways
Collateral connection grade CC2 was observed in 8 patients (42%) and CC1 in 11 patients (58%). Conversely, Rentrop grade was scored as 3 in 6 patients (31%), as 2 in 7 patients (38%), and as 1 in the remaining 6 patients (31%). There was no correlation between the CC grading and the Rentrop grading (r = 0.13, p = 0.590). No differences in clinical and procedural variables were observed between the patients, when stratified according to Rentrop (Rentrop 3 vs. Rentrop <3) or CC grades (CC2 vs. CC1) (Online Table 1 and Online Table 2).
The interobserver variability yielded a good concordance for Rentrop and CC grades (kappa = 0.74 and 0.91, respectively).
Overall coronary vasomotion and blood flow evaluation
Overall, there was a significant reduction in mean luminal diameter with Ach (p = 0.001) and Ade (p = 0.004), whereas no significant increase was observed with NTG (p = 0.084). Of note, 18 patients demonstrated vasoconstriction response after Ach (−45 ± 20% of change in mean luminal diameter), whereas 1 patient did not exhibit change in mean luminal diameter (+0.2% of change). Conversely, only 2 and 3 patients experienced vasodilation following administration of Ade and NTG, respectively (Table 2).
APV did not change after infusion of Ach (p = 0.172), whereas it significantly increased after Ade and NTG administration (p = 0.02 and p = 0.001, respectively).
Eventually, CBF decreased after infusion of Ach (p = 0.005), increased after infusion of NTG (p = 0.035), and did not change after infusion of Ade (p = 0.470).
Coronary vasomotion according to collateral pathways
Patients with a CC2 grade showed significantly less reduction in mean luminal diameter after infusion of Ach (p = 0.005) and greater increase in mean lumen diameter to NTG (p = 0.029) as compared to patients with a CC1 grade. No difference was found after infusion of Ade (p = 0.491) (Fig. 4).
Patients with a Rentrop grade 3, compared with patients with Rentrop grade <3, also exhibited a lower reduction in mean luminal diameter after infusion of Ach (p = 0.002). At the same time, they showed a trend toward a smaller change in mean luminal diameter in response to NTG (p = 0.143) compared with the other group. No difference was found after infusion of Ade (p = 0.773) (Fig. 4).
Coronary blood flow according to collateral pathways
APV was not different in patients with CC2 grade compared with those with CC1 grade after acetylcholine (p = 0.573), Ade (p = 0.414), or NTG infusion (p = 0.662). No differences were found categorizing the patients also according to Rentrop 3 versus <3 (Tables 3 and 4).
Patients with CC2 exhibited a smaller reduction in CBF after Ach (p = 0.008), compared with patients with CC1. CBF responses to Ade and NTG were similar between CC2 and CC1 (Fig. 5). Patients with Rentrop 3 exhibited smaller reduction in CBF after infusion of Ach (p = 0.008) and greater increase after Ade (p = 0.188) and NTG (p = 0.036) compared with patients with Rentrop grade <3 (Fig. 5).
The major findings of this study are: 1) the coronary segment distal to a successfully treated CTO exhibits endothelial and smooth muscle cells dysfunction by assessment immediately after percutaneous recanalization; 2) this translates in an impaired flow-mediated vasodilation after PCI that compromises a complete restoration of anterograde blood flow; 3) patients with a pre-procedural low-graded collateral circulation appear to have worse endothelium/smooth muscle cells dysfunction than patients with high graded collateral circulation; 4) relative changes in coronary blood flow seem also to be correlated to the grade of pre-procedural collateral circulation.
Classically, endothelial dysfunction represents the first step of atherosclerosis (2,18) and it has been extensively studied in various types of coronary lesions (3,4). Coronary chronic total coronary occlusions have per definition at least 3-months duration; thus, the vascular segment distally to the occlusion usually become supplied by a collateral circulation, which can more or less restore the blood requirements of the collateral-dependent myocardium and at the same time probably contribute to maintain a normal regional myocardial function (8). Our analysis showed, however, that an endothelial dysfunction is present in this kind of lesion immediately after the recanalization, as suggested by an intense vasoconstriction response to Ach in the vascular segment distal to the occlusion. The same vascular segment exhibits also a smooth muscle cells dysfunction, with an inability to vasodilate after administration of Ade or NTG. Adenosine and nitrates have vasodilatory effect on smooth muscle cells via different mechanisms. Adenosine is a potent arteriolar dilator via stimulation of A2 receptors of smooth muscle cells. Its main effect in the coronary circulation is a drastic decrease of arteriolar resistance, leading to an increase of CBF and, in return, to flow-mediated dilation of epicardial vessels. Conversely, nitrates have a direct vasodilatory effect on smooth muscle cells (19–21). The absence of vasodilation following these drugs may be related to a dysfunction in both mechanisms. Nevertheless, the enhanced vasoconstrictor response to Ach indicates the presence of some preserved vasoreactivity of the smooth muscle cell layer.
Interestingly, we observed a somewhat “paradoxical” response after administration of Ade, with vasoconstriction of the artery, which cannot be explained only by impaired endothelial-dependent flow-mediated dilation and may support direct constriction of smooth muscle cells. Previously in an animal model of CTOs, Heaps et al. (22) demonstrated the selective impairment of cyclic adenosine monophosphate–dependent vasodilation with Ade, which was partially restored by exercise training. In particular, they found that the vasodilation was not impaired by administration of forskolin, a membrane-permeant, receptor-independent vasodilator that stimulates adenyl cyclase directly (23). Similarly, Sellke et al. (24) found in a dog model an impaired relaxation to adenosine diphosphate in collateral-dependent coronary microvessels compared with control subjects. These findings in animal models can suggest either a selective damage/down-regulation of Ade receptors or a defective access of receptor agonist/associated G proteins to the intracellular pools of adenyl cyclase in coronary smooth muscle cells distal to the occlusion, which may in turn activate different intracellular pathways finally causing vasoconstriction. Another possible explanation of this “paradoxical” response to Ade is that the endothelial dysfunction after Ach administration was so intense to mask the flow-mediated dilation after Ade administration, further supporting the intense vascular dysfunction present in the vascular segment distal to a CTO immediately after recanalization. These hypotheses need to be confirmed in a larger study.
Our study analyzed also the blood flow responses after administration of these vasoactive drugs in CTO after recanalization. In particular, infusion of Ach was not able to increase the CBF, due to its marked vasoconstriction effect on epicardial vessels, but conversely it provoked a significant decrease in CBF. Adenosine and NTG correctly increase the APV, but a significant increase in coronary blood flow was detected only after NTG. The lack of flow-mediated dilation to Ade in presence of a correct increase in flow velocity might lead to an elevation in shear stress and could help to identify treated-CTO with high risk of restenosis (25). In clinical practice, many CTO cases show an improvement in distal vessel diameter in the long term. The fact that post-PCI angiography is performed in a context of severe endothelial dysfunction, as shown in our analysis, with a hypothetic positive, flow-mediated vessel remodeling, once the endothelial function improves under nonischemic and anterograde flow conditions, could explain this enlargement in distal vessel diameter. However, further studies are needed to confirm this hypothesis.
As the degree of collateral circulation in CTO is extremely variable, we sought to analyze the vasomotor and blood flow responses in relation to the degree of pre-procedural collateral circulation. This was evaluated using both qualitative and semiquantitative indexes, such as Rentrop and CC scores, respectively. In particular, the angiographic grading of the CC has been shown to be closely associated with invasively determined parameters of collaterals' hemodynamics. This allowed us to optimally differentiate collaterals according to their functional capacities without excessively prolonging the CTO-PCI for hemodynamic measurement (12). Conversely, Rentrop grade showed only a weak correlation with invasive parameters of collateral function and, as it is based on a qualitative assessment, has a high interobserver variability (12,26). In line with previous reports, we found a weak correlation between CC and Rentrop grades (12). The endothelial and smooth muscle cells dysfunction appeared to be greater in patients with a baseline CC1 grade than in patients with a baseline CC2 grade. The CC2 grade patients were also found to have smaller decreases in CBF after infusion of Ach compared with CC1 grade patients, which is suggestively due to a higher recruitment of collateral vessels. Despite the limited number of patients, our findings suggest a causal relationship between CTO collateral circulation and vasomotor function of the vessel distal to the occlusion. Future studies should address whether the impaired vasomotor function is really a cause or an effect of the pre-existing collateral circulation of a CTO. Another interesting hypothesis to test would be whether the presence of a previous myocardial infarction in the same territory of the CTO could additionally influence the vascular function distal to the CTO. In our analysis, only 1 patient exhibited necrosis distal to the CTO, so that any statistical comparison would have been meaningless.
The main limitations in our study relate to the small number of patients included. For these reasons, our findings should be considered as exploratory and need to be confirmed in larger studies. In addition, we did not collect any data about the non-CTO vessel and did not include a control group. Time of occlusion definition could be imprecise in such a small population and in the absence of serial angiograms.
The collateral function was not assessed invasively (e.g., collateral flow index) to not prolong excessively the procedural time. The CC angiographic grading, however, has previously been shown to be closely associated with invasively determined parameters of collaterals hemodynamics (12).
We used only 1 dose of Ach (10−6), which is the highest dose used in humans to evaluate the maximal endothelium-dependent vasomotor response (4,14). Coronary flow reserve was not calculated, as epicardial vessel sizes were not constant after the infusion of Ade, as previously indicated (27).
Our study demonstrates that in the coronary segment distal to a CTO, there is an impairment of endothelial and smooth muscle cell function, which seems to be most notable in the presence of low-graded collateral circulation. Larger studies are required to determine the clinical significance of these findings.
The authors thank all the nurses of the interventional cardiology department of “Santa Creu i Sant Pau” Hospital and of “Hospital Clinic” in Barcelona, Spain. The authors would like to thank also Scot Garg and Maria D. Radu, who kindly reviewed the manuscript.
For supplementary information on Rentrop and CC grades, please see the online version of this article.
This study received a grant from Spanish Society of Cardiology (“BECA SEC PARA INVESTIGACIÓN CLÍNICA,” 2008) and it is included as part of the “Beca La Marató de TV3” 2008. The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- average peak velocity
- coronary blood flow
- collateral connection
- chronic total occlusion
- percutaneous coronary intervention
- Thrombolysis In Myocardial Infarction
- Received August 12, 2011.
- Revision received October 12, 2011.
- Accepted October 28, 2011.
- American College of Cardiology Foundation
- Schächinger V.,
- Britten M.B.,
- Zeiher A.M.
- Bonetti P.O.,
- Lerman L.O.,
- Lerman A.
- Iràculis E.,
- Cequier A.,
- Gómez-Hospital J.A.,
- et al.
- Stone G.W.,
- Kandzari D.E.,
- Mehran R.,
- et al.
- Cohen M.,
- Rentrop K.P.
- Werner G.S.,
- Surber R.,
- Ferrari M.,
- Fritzenwanger M.,
- Figulla H.R.
- Werner G.S.,
- Ferrari M.,
- Betge S.,
- Gastmann O.,
- Richartz B.M.,
- Figulla H.R.
- Rentrop K.P.,
- Cohen M.,
- Blanke H.,
- Phillips R.A.
- Carroll R.J.,
- Verani M.S.,
- Falsetti H.L.
- Werner G.S.,
- Ferrari M.,
- Heinke S.,
- et al.
- Sabaté M.,
- Kay I.P.,
- van Der Giessen W.J.,
- et al.
- Doucette J.W.,
- Corl P.D.,
- Payne H.M.,
- et al.
- Hodgson J.M.,
- Marshall J.J.
- Maseri A.
- Lupi A.,
- Buffon A.,
- Finocchiaro M.L.,
- Conti E.,
- Maseri A.,
- Crea F.
- Heaps C.L.,
- Sturek M.,
- Rapps J.A.,
- Laughlin M.H.,
- Parker J.L.
- Seamon K.B.,
- Padgett W.,
- Daly J.W.
- Sellke F.W.,
- Quillen J.E.,
- Brooks L.A.,
- Harrison D.G.
- Pries A.R.,
- Habazettl H.,
- Ambrosio G.,
- et al.