OUP user menu

Transthoracic echocardiography in the detection of chronic total coronary artery occlusion

Alla A. Boshchenko , Alexander V. Vrublevsky , Rostislav S. Karpov
DOI: http://dx.doi.org/10.1093/ejechocard/jen159 62-68 First published online: 1 January 2008

Abstract

Aims The aim of our study was to detect chronic total occlusion of the left anterior descending coronary artery (LAD), circumflex coronary artery (Cx), and right coronary artery (RCA) using transthoracic echocardiography (TTE) in 110 consecutive patients who underwent coronary angiography for investigation of angina.

Methods and results Coronary blood flow direction was assessed in the epicardial collaterals [distal LAD (dLAD), obtuse marginal branches and right posterior descending artery (PDA)] and intramyocardial collaterals [LAD septal branch (SB LAD) and RCA septal branch (SB RCA)]. The sensitivity and specificity of retrograde flow for identification of the occluded LAD by TTE in the dLAD only were 78 and 96%, respectively, and those in both dLAD and SB LAD were 89 and 96%, respectively. The retrograde SB LAD flow detects proximal LAD occlusion with 88% sensitivity and 75% specificity. The sensitivity and specificity of retrograde flow for identification of the occluded RCA by TTE in the PDA only were 79 and 97%, respectively, and those in both PDA and SB RCA were 89 and 97%, respectively. The retrograde SB RCA flow does not allow us to differentiate between proximal and non-proximal RCA occlusion. Transthoracic echocardiography is not a method for diagnosing Cx occlusions as the success in visualizing the Cx epicardial collaterals was achieved in 31% of cases only.

Conclusion TTE is a sensitive and highly specific non-invasive method for diagnosis of LAD and RCA occlusions, based on the detection of the coronary blood flow direction in the epicardial and intramyocardial collaterals.

Keywords
  • Transthoracic echocardiography
  • Chronic total coronary occlusion
  • Coronary blood flow

Introduction

At present, the study of the anatomically manifested coronary atherosclerosis remains a prerogative of invasive diagnostic methods, and quantitative coronary angiography (QCA) and intracoronary flowmetry continue to be the ‘gold standards’ of estimation of the chronic total coronary occlusion (CTO) site and size and determination of the post-occusive collateral blood flow.1 The last decade rapid progress in ultrasound and tomography technologies has pushed off some diagnostic priorities of invasive methods.24 Transesophageal echocardiography has proved to be potential in CTO detection.5 However, only proximal coronary lesion can be diagnosed via transesophageal approach because of the non-visualization of the mid- and distal parts of the main coronary arteries.6 Development of high-frequency transthoracic transducers and technology of the second tissue harmonic imaging have allowed application of transthoracic echocardiography (TTE) as a non-invasive, inexpensive, and widely used in clinical practice method for the diagnosis of coronary narrowing. Certain transthoracic ultrasound markers of stenosis of the left anterior descending artery (LAD) and right coronary artery (RCA) as well as markers of successful intracoronary interventions on the LAD have been established.7,8 Recently, the reliability of TTE in detecting coronary occlusion of the LAD and RCA, basing on registration of the retrograde flow in the epicardial collaterals [in the distal LAD (dLAD) for the LAD and posterior descending artery (PDA) for the RCA] has been reported.9,10 Although this method is accurate for the epicardial collaterals, which grow from the peripheral vessels of the non-occluded (‘donor’) epicardial coronary artery to the distal part of the occluded epicardial coronary artery, intramyocardial collaterals, which run mainly in the interventricular septum, could be missed and the detection of the occlusion site could not be detected.

The aim of our study was a comparative diagnosis of main coronary artery CTO using quantitative coronary angiography (QCA) and TTE, based on the examination of collateral epicardial and intramyocardial coronary blood flow.

Methods

Study group

Hundred and ten consecutive non-selective patients (mean age 51 ± 11 years; 98 men and 12 women) with angina pectoris or atypical chest pain, who were planned for the QCA, were included in the study. The clinical characteristics of the study patients are shown in Table 1. Patients with Q wave or without Q wave myocardial infarction of <1 month duration were excluded. Informed written consent was obtained from all participants according to the guidance of the Ethic Committee of the Tomsk Cardiology Research Institute, Russia.

View this table:
Table 1

Clinical characteristics of the study patients (n = 110)

Parameters%
Q wave or without Q wave myocardial infarction in the past (>1 month)50
Angina pectoris or atypical chest pain56
Obesity (>30 kg/m2 for men, >24 kg/m2 for women)21
Hyperlipidaemia (total cholesterol >5.0 mmol/L or low-density lipoprotein cholesterol >3.5 mmol/L)74
Arterial hypertension (blood pressure >140/90 mmHg)56
Diabetes mellitus or elevated fasting glucose level8
Smoking42
Peripheral or cerebral occlusive arterial disease9
Conduction disorders (left bundle-branch block, right bundle-branch block, left anterior fascicular block, etc.)10
Rhythm disorders (chronic atrial fibrillation, ventricular or/and supraventricular extrasystoles, sinus tachycardia >90 bpm, etc.)13
Clinical heart failure with left ventricular ejection fraction <40%7
Constant beta blockers or calcium antagonists therapy49
Antiarrhythmia therapy excluding beta blockers4

Coronary angiography

Multiplane QCA was performed through femoral approach [Judkins’ standard method (1967); COROSKOP Plus angiographic complex, Siemens, Germany] to search occlusions of coronary arteries, CTO site detection and to determine characteristics of collateral coronary blood flow within 1 week after TTE. Coronary occlusion was estimated as the focal absence of anterograde flow.

Transthoracic echocardiography

Transthoracic echocardiography was performed without contrast enhancement with the patients positioned in the left lateral decubitus position using ultrasound diagnostic system Vivid 7 GE Healthcare, coronary artery preset and an M3S narrow-band sector transducer (1.7–3.4 MHz). All studies were continuously recorded on hard-drive and CD-disks for off-line analysis. Blood pressure and heart rate were recorded simultaneously by Bosotron 2 (Bosch+Sohn, Germany). The mean duration of the TTE study was 15 ± 4 min.

First, colour Doppler mapping with the Nyquist limit set at 13–27 cm/s was used for the search of coronary arteries; second, the direction of coronary blood flow was identified; and next, blood flow velocity patterns were registered by pulsed wave Doppler using a sample volume (2.0–3.0 mm) placed on the colour signal. We tried to align the ultrasound beam direction to vessel flow as parallel as possible. Only the presence and waveform of systolic and diastolic coronary blood flow was assessed and the time T-F was calculated as a time between the end of the T wave on ECG (the start of electric diastole) and the start of coronary blood flow diastolic phase. First, epicardial collateral vessels were studied: dLAD, first or second obtuse marginal branch (OMB) as a distal circumflex artery (Cx) and PDA as a distal RCA. Next, intramyocardial collateral vessels (septal branches—SB) were estimated.

The dLAD was examined from a low left parasternal position to a modified apical two-, four-, or five-chamber position at varying levels using different short- and long-axis views in the anterior interventricular groove. The normal anterograde blood flow in the dLAD was identified as a red linear colour signal on colour Doppler map (Figure 1A) and as a waveform above zero, which reflected the direction of the flow from base to apex of the left ventricle, at spectral Doppler recording (Figure 1B). The distal Cx and distal RCA were searched from the fourth and fifth intercostal spaces in the apical long-axis position: first or second OMB, in a modified four- or five-chamber view at the lateral wall of the left ventricle (Figure 1C), and PDA, in a modified two- or three-chamber view with caudal angulation of the probe in the posterior interventricular groove (Figure 1D). The direction of the normal OMB flow and normal PDA flow was the same, as well as that of the normal LAD flow.

Figure 1

Examples of normal anterograde coronary blood flow in distal epicardial coronary arteries: (A) colour Doppler image of the distal part of the left anterior descending coronary artery (dLAD); a modified apical two-chamber view. (B) Spectral Doppler tracing of the anterograde coronary blood flow in the dLAD; S, systole; D, diastole. (C) Colour Doppler image of the first obtuse marginal branch (I OMB) as a distal part of the circumflex coronary artery; a modified five-chamber view. (D) Colour Doppler image of the posterior descending artery (PDA) as a distal part of the right coronary artery; a modified two-chamber view with caudal angulation of the probe.

To record the coronary flow velocity pattern in the SB, the transducer was placed at the left parasternal position. A modified parasternal short axis B-view of the left ventricle at the papillary muscle level was obtained. Inclining the probe from base to apex of the left ventricle, SB were searched as the linear intramyocardial colour Doppler signals in the interventricular septum. The SB LAD was defined as a mid-LAD branch perforating the interventricular septum from the anterior to the posterior interventricular groove. The normal anterograde blood flow in the SB LAD was identified as a blue linear colour signal on colour Doppler map and as a waveform below zero at spectral Doppler recording (Figure 2A). Septal branch of the Cx was not identified. The SB RCA was defined as a mid- or distal RCA branch perforating the interventricular septum from the posterior to the anterior interventricular groove toward the SB LAD. The normal anterograde blood flow in the SB RCA was identified as a red linear colour signal on colour Doppler map and as a waveform above zero at spectral Doppler recording (Figure 2B).

Figure 2

Examples of colour Doppler mapping (top) and spectral Doppler tracing (bottom) of the normal anterograde coronary blood flow in the intramyocardial branches of the left anterior descending artery (LAD) and of the right coronary artery (RCA); a modified parasternal short axis B-view at the level of papillary muscles. (A) Septal branch of the LAD (SB LAD); mLAD, mid-part of the LAD. (B) Septal branch of the RCA (SB RCA); mRCA, mid-part of the RCA.

Retrograde coronary blood flow was detected in the case of inversion of the normal coronary blood flow direction. We defined retrograde dLAD flow or retrograde SB LAD flow as a Doppler marker of LAD occlusion indicating at epicardial or intramyocardial collateral blood flow, respectively (Figure 3). Retrograde OMB flow was assessed as epicardial collateral blood flow being a Doppler marker of Cx occlusion (Figure 4). We considered retrograde PDA flow or retrograde SB RCA flow as a Doppler marker of RCA occlusion indicating at epicardial or intramyocardial collateral blood flow, respectively (Figure 5).

Figure 3

Proximal occlusion of the left anterior descending coronary artery (LAD). Examples of colour Doppler mapping (top) and Doppler spectrum (bottom) of the retrograde coronary blood flow. (A) Epicardial collateral vessel: distal LAD (dLAD); (B) intramyocardial collateral vessel: septal branch of the LAD (SB LAD).

Figure 4

Patient N. Mid occlusion of the circumflex coronary artery. Examples of colour Doppler mapping (top) and Doppler spectrum (bottom) of the retrograde coronary blood flow in the epicardial branch (second obtuse marginal branch—OMB).

Figure 5

Patient M. Proximal occlusion of the right coronary artery (RCA). Examples of colour Doppler mapping (top) and Doppler spectrum (bottom) of the retrograde coronary blood flow. (A) Epicardial collateral vessel: posterior descending artery (PDA); (B) intramyocardial collateral vessel: septal branch of the RCA (SB RCA).

Every patient was assessed and the Doppler images were analysed by two experienced specialists, who were blinded to the angiographic data.

Statistical analysis

STATISTICA package version 6.0 for Windows software (StatSoft Inc., USA) was used for statistical analysis. Unpaired t-test for independent variances was used for the analysis of the parametric data at the patients with coronary occlusion and patients without coronary occlusion. Analysis of the χ2 test was performed to compare the rate of non-parametric variances. A P-value of <0.05 was considered statistically significant. For comparative analysis of TTE and QCA, sensitivity and specificity of TTE in diagnosis of occlusions were determined using standard formulas.

Results

Coronary angiography

Eighteen LAD occlusions (eight proximal, eight mid, and two distal), 11 Cx occlusion (five proximal, four mid, and two distal) and 28 RCA occlusions (8 proximal, 17 mid, and 3 distal) were determined in 47 patients by coronary angiography. Thirty-seven patients had an occlusion of one vessel (12 LAD, 6 Cx, and 19 RCA), 10 patients had an occlusion of two vessels (five LAD and RCA, four Cx and RCA, and one LAD and Cx). Sixty-three patients had no occlusions. Significant differences in the age and rate of myocardial infarction in the past were observed between the group of patients with occlusion and patients without occlusion (Table 2).

View this table:
Table 2

Comparative clinical characteristics of the patients with coronary occlusion and patients without occlusion (n = 110)

ParametersPatients with occlusion (n = 47)Patients without occlusion (n = 63)P-value
Age54 ± 847 ± 12<0.01
HR (bpm)61 ± 1064 ± 11NS
SBP (mmHg)124 ± 17123 ± 19NS
DBP (mmHg)78 ± 1579 ± 19NS
Myocardial infarction in the past (%)7623<0.001
Obesity (%)2421NS
Hyperlipidaemia (%)7678NS
Arterial hypertension (%)6059NS
Diabetes mellitus (%)108NS
Smoking (%)4543NS

Transthoracic visualization of the coronary arteries

A good colour Doppler mapping and Doppler velocity patterns in the dLAD were obtained in 103 (94%), OMB in 34 (31%), and PDA in 105 (95%) of 110 patients. Septal branches of the LAD and SB RCA were visualized rarely in 21 and 16% of cases, respectively.

Interobserver variabilities for Doppler identification of the dLAD, OMB, and PDA were 1.9, 8.8, and 3.8%, respectively.

TTE in LAD occlusion

Retrograde flow by TTE was detected in 16 (dLAD, seven patients; SB LAD, two patient; both dLAD and SB LAD, seven patients) of 18 patients with the occluded LAD, and anterograde flow was detected in 82 of 85 patients without LAD occlusion (Table 3). The sensitivity and specificity of retrograde flow for identification of the occluded LAD by TTE in the dLAD only were 78 and 96%, respectively, and those in both dLAD and SB LAD were 89 and 96%, respectively. Significant differences in the rate of retrograde SB LAD flow were observed between the patients with proximal and non-proximal LAD occlusion (χ2 = 6.35, P < 0.05). Retrograde SB LAD flow was determined in seven of eight (88%) cases of proximal LAD occlusion and in two of eight (25%) cases of non-proximal LAD occlusion only. So, the retrograde SB LAD flow detected the proximal LAD occlusion with 88% sensitivity and 75% specificity.

View this table:
Table 3

Detection of chronic total coronary artery occlusions: comparative analysis of transthoracic echocardiography (TTE) and quantitative coronary angiography (QCA) data

VesselQCA occlusion+ (n)QCA occlusion− (n)TTE occlusion+ (n)TTE occlusion− (n)TP (n)TN (n)FP (n)FN (n)
LAD (n = 103)18851984168232
Cx (n = 34)52943042901
RCA (n = 105)28772778257523
  • LAD, left anterior descending artery; Cx, circumflex artery; RCA, right coronary artery; n, number of cases; TP, true-positive cases; TN, true-negative cases; FP, false-positive cases; FN, false-negative cases.

TTE in Cx occlusion

Retrograde flow by TTE was detected in the OMB in four of five patients with the occluded Cx (Table 3), and anterograde flow was detected in 29 of 29 patients without Cx occlusion. The sensitivity and specificity of retrograde flow for identification of the occluded Cx by TTE in the OMB were 80 and 100%, respectively, on condition that the OMB was successfully visualized by TTE, and 36 and 100%, respectively, if the estimation was performed in all study patients.

TTE in RCA occlusion

Retrograde flow by TTE was detected in 25 (PDA, 14 patients; SB RCA, 3 patients; both PDA and SB RCA, 8 patients) of 28 patients with the occluded RCA, and anterograde flow was detected in 75 of 77 patients without RCA occlusion. The sensitivity and specificity of retrograde flow for identification of the occluded RCA by TTE in PDA only were 79 and 97%, respectively, and those in both PDA and SB RCA were 89 and 97%, respectively. Non-significant differences in the rate of retrograde SB RCA flow were observed between the patients with proximal and non-proximal RCA occlusion (χ2 = 1.63, P > 0.05). Retrograde SB RCA flow was determined in five of eight (63%) cases of proximal LAD occlusion and in 6 of 17 (35%) cases of non-proximal RCA occlusion. The sensitivity and specificity of retrograde SB RCA flow for identification of proximal RCA occlusion were 63 and 65%, respectively.

Discussion

Recent studies have demonstrated a high clinical value of early detection of CTO11 as a pathology resulting in a considerable increase of sudden death frequency even in comparison with severe stenosis of coronary arteries.12 It has been established13 that patients with verified diagnosis of single-vessel CTO for the choice of medication therapy have a high mortality rate (25%) over the 10 year prospective period, whereas endovascular interventions (balloon angioplasty or coronary stenting) increase the survival rate in this cohort1416 at а presence of persistent myocardial ischemia.17 However, unlike patients with acute coronary occlusions and severe coronary stenosis, patients with CTO often have neither typical anginal clinical symptoms, nor ischaemic changes on ECG, nor regional wall motion abnormalities of the left ventricle warranting the performance of QCA.

It has been established that visualization of the LAD segments using high-frequency TTE with second tissue harmonic imaging and echocontrast enhancement of Doppler signals is totally possible in 90–100% of patients, RCA in 75–76% of patients, and Cx in 35% of patients.710,1821 Our data on visualization of the main coronary arteries, though not using echocontrast agents, did not significantly differ from the previously obtained data,9,10,18,19 and the detection of the PDA as distal RCA was even more successful, that could be explained by the use of a different ultrasound system and ultrasound probe, and different study tasks. So, we studied the distal epicardial coronary artery only for which the direction and blood flow systole-diastole phasic structure without detailed estimation of blood flow velocities and time indexes were determined, whereas other authors1820 analysed all segments of the coronary artery or considered the quality of the coronary flow pattern.

Recently, possibility of TTE for the detection of LAD and RCA occlusion has been found.9,10,22 As TTE permits detection of only separate segments of coronary arteries, and the study of coronary blood flow is based on the Doppler colour coding of its velocity and direction, Watanabe et al.9 proposed the inversion of the coronary blood flow in the epicardial collateral vessels to be a main ultrasound CTO sign. It has been established that retrograde flow in the dLAD is a marker of LAD occlusion with 98% sensitivity and 100% specificity,9 and retrograde flow in the PDA is a marker of RCA occlusion with 68% sensitivity and 100% specificity.10 In 2006, Pizzuto et al.22 doubted such a high sensitivity of TTE in the detection of LAD occlusion as they have revealed a retrograde flow in the dLAD in only 43% of patients with CTO and normal anterograde flow in more than half the patients (55%). Our data based on the detection of retrograde coronary blood flow in epicardial collaterals only have demonstrated specificity for the LAD and RCA and sensitivity for the RCA consistent with the results of Japanese investigators,9,10 but moderate sensitivity for the LAD which confirm the data of Pizzuto et al.22 Angiographic data demonstrate that in LAD and RCA occlusions, besides collateral flow distal to the occluded region through the connections on the epicardial surface, 63–82% of the patients have intramyocardial collateral channels lying usually in the interventricular septum.2325 Intramyocardial channels providing input of blood to the mid-part of the occluded artery are a case of anterograde flow in their distal part. On the basis of the angiographic studies, we hypothesized that simultaneous determination of the coronary blood direction and type in the distal epicardial arteries and intramyocardial (septal) collaterals might be a more sensitive method of diagnosing CTO than the assessment of the epicardial collateral flow only. Indeed, additional study of the blood flow in the intramyocardial pathways increased the sensitivity of TTE in the detection of both the occluded LAD (from 78 to 89%) and the occluded RCA (from 79 to 89%) without the loss of specificity. Moreover, patients with CTO demonstrated an improvement of visualization of SB LAD and SB RCA as actively working collaterals (for SB LAD from 21% at all study patients to 50% at patients with LAD occlusion, P < 0.001 and for SB RCA from 16% at all study patients to 39% at patients with RCA occlusion, P < 0.01). Characteristics of collateral flow in the epicardial and intramyocardial collateral vessels were different. The diastolic phase of the coronary flow in the epicardial collaterals, having low extravascular compression,24 started within 0–100 ms of the start of the electric diastole on ECG (Figure 4). In the intramyocardial collaterals, having high extravascular compression due to the intramural tract of the vessels, the appearance of the diastolic flow was considerably later in respect to the start of the electric diastole on ECG (100–250 ms) (Figure 5).

The study aimed at investigation of TTE potentials in the diagnosis of Cx occlusions was performed by us for the first time. Unlike CTO of the LAD and RCA, the basis for diagnosing CTO of the Cx was the assessment of the potential epicardial collaterals only. However, even this approach was possible in <1/3 (31%) of the patients. It is due to the fact that in Cx occlusions large intramyocardial collateral pathways are rarely formed and their visualization is technically difficult because of the large angle (40–90°) between the Cx branches and the direction of the ultrasound beam in all main transthoracic views. But when the distal Cx could be located TTE sensitivity was satisfactory and specificity was 100% in the diagnosis of occlusion.

So far there have been no investigations aimed at topical detection of CTO with the help of TTE. According to angiography studies the SB LAD forming intramyocardial collaterals of the occluded LAD more frequently take off the mid-LAD, whereas the SB RCA can be branches of any RCA segment including its posterior atrioventricular artery. Hence, we supposed that determination of the SB flow direction would enable us to make topical diagnosis of LAD occlusion, but fail to differentiate between proximal and non-proximal RCA occlusion. Indeed, in all patients with LAD occlusion situated more proximal to the site of SB LAD take-off (i.e. with LAD proximal occlusion) the flow direction in the SB LAD was inverted into a retrograde one, whereas in non-proximal LAD occlusion situated more distal to the SB LAD take-off the normal anterograde flow direction was preserved in the SB LAD in 75% of cases. At the same time, the blood flow in the SB RCA was retrograde not only in patients with proximal CTO but in a considerable part of patients (35%) with non-proximal CTO. This fact does not allow us to recommend determination of the flow direction in the SB RCA for the topical diagnosis of RCA occlusion.

Clinical implication

Transthoracic echocardiography can be rather an attractive non-invasive method of CTO detection due to its low cost, little time consumption, and good reproducibility in standardized approach. The method may be used at the routine ultrasound examination of the heart before QCA in diagnostically difficult cases (left bundle-branch block, frequent extrasystoles, non-sinus rhythm, etc.) or at patients with poor or atypical symptoms both for verification or exception of CTO presence only and in combination with all possible cases, with the stress testing for regional wall motion abnormality detection and coronary flow reserve assessment not only in epicardial but in intramyocardial collaterals for identification and topical diagnosis of myocardial ischaemia. The TTE detection of CTO is based on the identification of the coronary blood flow direction only without the assessment of rest Doppler coronary flow velocities and, thus, it is not dependent on the haemodynamic conditions of the patients and does not require discontinuation of the medication therapy, including antianginal, hypotensive, antiarrhythmic drugs, that allows to apply the method at unstable patients, patients with life-threatening rhythm, and conductance disturbances.

Study limitations

The main limitation is a peculiarity of selecting the study patients. As it was necessary to angiographically control TTE findings the study included patients planned for QCA in whom the expected frequency of CTO and diagnostic look-out were high. The number of study patients nor was large. One of main limitations of TTE application for CTO diagnosis in clinical practice is an operator-dependent factor, i.e. the necessity of having a well-trained specialist.

All false-positive results of the TTE were associated with the detection of ‘occlusion’ in patients with subtotal stenosis according to QCA data. Evidently, it can be explained by different methodical approaches. In angiography, the contrast agent is pumped under pressure in the vessel, and it can improve the detection of the artery minimal residual lumen or provoke its functioning. In TTE, coronary blood flow distal to the lesion site, depending on intravascular pressure gradient only without any additional action, is determined.

False-negative results of TTE for all arteries were associated with the wrong topical diagnosis of the vessels due to their fragmentary visualization. So, two cases of the unrevealed LAD occlusion were due to the flow recording not in the occluded dLAD, but in the first diagonal artery which was not occluded. All three cases of the unrevealed RCA occlusion were in patients with normal anterograde flow in the PDA but strongly dominant left coronary artery in which the PDA is a branch of the left and not the RCA.

The peculiarity of the present study is the assessment of the TTE potentiality in the diagnosis of CTO only (over 1 month duration) in which the formation of the collateral coronary pathways is under way or has been completed. The principles of transthoracic diagnosis of acute coronary occlusion are sure to be absolutely different and require an independent study.

Conclusion

Thus, TTE is a sensitive and highly specific non-invasive method for diagnosis of LAD and RCA occlusion, based on the detection of coronary blood flow direction in the epicardial and intramyocardial collaterals, but TTE is not a method for the diagnosis of Cx occlusion as the success in visualization of the Cx epicardial collaterals is very low.

Conflict of interest: none declared.

References

View Abstract