Asian Cardiovasc Thorac Ann 2002;10:47-52
© 2002 Asia Publishing EXchange Pte Ltd
Transcatheter Coil Embolization of Coronary Artery Fistula
Bagrat G Alekyan, MD,
Vladimir P Podzolkov, MD,
Carina E Cárdenas, MD
|
Department of Interventional Cardiology and Angiology Bakoulev Scientific Center for Cardiovascular Surgery Russian Academy of Medical Sciences Moscow, Russia
|
|
|
Bagrat G Alekyan, MD Tel: 7 095 414 7547 Fax: 7 095 414 7708 email: b_alekyan{at}mtu-net.ru Department of Interventional Cardiology and Angiology, Bakoulev Scientific Center for Cardiovascular Surgery, Russian Academy of Medical Sciences, 135 Roublevskoye Shosse, Moscow 121552, Russia.
|
 |
ABSTRACT
|
|---|
Between December 1982 and August 2001, coil embolization of coronary artery-to-cardiac chamber fistula was attempted in 15 patients aged 11 months to 44 years (mean, 7.2 ± 2.5 years). The fistulae connected the left anterior descending artery to the right ventricle in 4 patients, the right coronary artery to the right ventricle in 3, the right coronary artery to the right atrium in 3, the circumflex artery to the right ventricle in 2, the circumflex artery to the right atrium in 2, and the right coronary artery to the trunk of the pulmonary artery in 1. Complete fistula occlusion was achieved in 14 patients (93%); one had a residual shunt and underwent repeat embolization one year later, resulting in complete occlusion. There was one early death (7%) in a 4-year-old girl who developed femoral artery thrombosis and acute renal failure. Complications comprised migration of the coil into the pulmonary artery (2), femoral artery thrombosis (2), and perforation of the vessel wall by the guidewire (1) with immediate thrombosis and occlusion of the fistula (no coil was deployed). The 13 survivors with coils were followed up for 0.5 to 13 years; complete occlusion of the fistula was confirmed in all cases.
|
INTRODUCTION
|
|---|
Congenital coronary artery-to-cardiac chamber fistulae have been recognized since 1865, and account for 0.08% to 0.4% of all congenital heart diseases.1–4 Communi-cations between the right coronary artery (RCA) and the right ventricle (RV) are the most frequent, followed by fistulae between the right atrium (RA) and the pulmonary artery (PA).2–4 Hemodynamic effects depend on the size of the shunt into the cardiac chamber, which can reach up to 50% of cardiac output, and on the reduction of myocardial blood supply.1,2 Myocardial blood flow is determined by the difference between the resistance of the orifice of the fistula and that of the intramural myocardial vessels. As the resistance of the latter is always higher, the largest proportion of arterial blood goes through the fistula, and the myocardium receives insufficient blood supply. Fistulae are usually diagnosed in infancy or early childhood as there is a continuous systolic-diastolic murmur with maximal intensity in the 3rd and 4th intercostal spaces. More than half of the patients are asymptomatic. However, when the fistula is of significant size, signs of heart failure may be present, even in neonates. Fatigue, dyspnea, and chest pain appear at a later stage in up to 80% of patients over the age of 20 years.5 Signs of myocardial ischemia and infarction have been reported in 3% of patients.4,6 Complications such as infectious endocarditis, cardiac arrhythmias, congestive heart failure, and rupture of the thin fistula wall have also been seen.2,7,8 The first successful coil embolization of a coronary artery fistula was performed in 1982 at our center. This method seems to be a good alternative to surgical occlusion.9,10 We describe our 18-year experience and long-term results.
|
PATIENTS AND METHODS
|
|---|
Between December 1982 and August 2001, transcatheter embolization of congenital coronary fistula was attempted in 15 patients. There were 8 females and 7 males, ages ranged from 11 months to 44 years (mean, 7.2 ± 2.5 years), and weights ranged from 9 to 70 kg (mean, 25 ± 3.1 kg). In 14 patients, the fistula was the only defect; 1 patient had an associated patent ductus arteriosus. The condition was diagnosed at birth in all cases. Dyspnea on minimal exertion was the main symptom in 10 patients (67%). Clinical status at the time of diagnosis was New York Heart Association functional class I in 7 patients, class II in 3, class III in 1, and class IV in 2. Five patients (33%) complained of intermittent chest pain, one of whom had myocardial infarction. Investigations included clinical examination, an electrocardiogram (ECG), and radiography. Color Doppler echocardiography was carried out on admission, in the first week after the procedure, and during follow-up. A systolic-diastolic murmur with maximal intensity in the 3rd and 4th intercostal spaces to the left of the sternum was auscultated and registered on a phonogram. The ECG showed volume overload of the right and/or left heart, and there were signs of myocardial ischemia in 5 patients (33%). Preexcitation (Wolff-Parkinson-White type) was noted in one case.
In all patients, Doppler echocardiography confirmed the diagnosis by locating the coronary ostia and measuring the diameters by color-flow mapping. Localization of the coronary artery was performed in the parasternal projection of the short axis of the posterior aspect of the main artery (usually the aorta), and in the parasternal projection of the long axis of the left ventricle, with the withdrawal of the aortic root and an individually selected section for the location of the fistula.11 Color Doppler echocardiographic mapping was carried out in the same projections. Doppler echocardiographic signs of these fistulae included direct observation of an aneurysm along the length of the coronary artery or aneurysmal dilatation of the ostium, and indirect evidence based on the detection of pathological flow due to abnormal drainage of the coronary artery into the heart chamber. The diagnosis was confirmed during the embolization procedure by radiography which ascertained the location, quantity of drainage, and fistula diameter.
Embolization was performed with standard 0.035 or 0.038-inch Gianturco coils (William Cook Europe, Bjaeverskov, Denmark). The following instruments were used: a 5F introducer and a 5F pigtail catheter for right heart cathe-terization, 5F Judkins right and left coronary catheters for selective coronary angiography, a 5F Judkins catheter to deliver the coils, and a set of standard 0.014 to 0.038-inch guidewires. In first-stage catheterization of the right heart, the pigtail catheter was used, blood samples were taken, and RV and PA pressure curves were registered. Systolic pressure in the RV and PA was normal in 7 patients (47%), and elevated in the other 8 (53%) to between 32 and 68 mm Hg. Left-to-right shunting was confirmed in all patients and ranged from 30% to 65% with a mean of 44.5% ± 2.7%.
The choice of treatment method depended on the anatomical structure of the fistula. In cases of a distal fistula, stenosis of the fistula, and no more than 2 drainage openings, preference was given to embolization. The approach was based on the angiographic findings. The percutaneous arterial transfemoral approach was used in 12 patients (80%). In 3 patients aged 11, 22, and 24 months, the procedure was performed via an isolated right subscapular or right common carotid artery. The fistula was between the left anterior descending coronary artery and the RV in 4 patients, between the RCA and the RV in 3 patients, between the RCA and the RA in 3, between the circumflex artery and the RV in 2, between the circumflex and the RA in 2, and between the RCA and the trunk of the PA in 1. The number of coils implanted ranged from 2 to 20 and was determined by the size of the fistula. The coils were advanced using a straight 0.038-inch guidewire. They were placed sufficiently far from the drainage opening (1 to 1.5 cm) to prevent migration into the RA, RV, or PA. To obtain complete occlusion, coils were implanted so as to merge together forming a conglomeration. Coronary angiography was repeated to confirm the quality of the occlusion. Heparin (100 U•kg-1) was routinely given during the procedure.
|
RESULTS
|
|---|
Complete occlusion of the fistula was achieved in 14 (93%) patients (Figures 1, 2, and 3
). In the majority of cases, closure was achieved by the introduction of 3 to 5 coils. One patient (7%) with a huge RCA aneurysm had a trivial residual shunt and underwent repeat embolization one year after the initial procedure, resulting in complete occlusion of the fistula (Figure 4).
View larger version (332K):
[in this window]
[in a new window]
|
Figure 1. (A) Coronary angiogram showing a fistula between the left anterior descending coronary artery and the right ventricle. (B) After coil embolization, complete fistula occlusion and circumflex artery filling can be seen.
|
|
View larger version (324K):
[in this window]
[in a new window]
|
Figure 2. (A) Fistula between a branch of the right coronary artery and the right ventricle. (B) Complete fistula occlusion by Gianturco coils, and right coronary artery filling.
|
|
View larger version (282K):
[in this window]
[in a new window]
|
Figure 3. (A) Fistula between the circumflex artery and the right atrium. (B) Complete fistula occlusion by Gianturco coils.
|
|
Electrocardiography immediately after the procedure and on the following day showed no signs of myocardial ischemia, even in the 5 patients who had ischemia before embolization. Phonocardiograms and transthoracic Doppler echocardiography were performed on the day after the embolization. The phonocardiograms of 14 patients (93%) did not show the to-and-fro murmur detected before the procedure, and Doppler analysis with color-flow mapping verified occlusion in these 14 cases. The patient with a trivial residual shunt had a significant reduction in the amplitude of the murmur.
There was one incident of early mortality in a 4-year-old girl who developed femoral artery thrombosis after the procedure, probably related to the use of an 8F introducer; unfortunately, no 5F introducer was available at that time. Femoroiliac thrombectomy was performed twice. However, acute renal failure requiring hemodialysis developed, and the patient subsequently died.
Early complications included: migration of a coil into the PA in 2 patients (13%), all migrated coils were removed from the PA using a basket, and the fistulae were closed with additional coils; femoral artery throm-bosis in 2 patients (13%); and perforation of the wall of the fistula by the guidewire in 1 case (Figure 5), which resulted in immediate development of limited hemopericardium and thrombosis of the fistula. It was decided not to implant coils in this patient because of angiographic and echocardiographic confirmation of total occlusion of the fistula; there were no ECG changes in the first postprocedural week, and trans-thoracic Doppler echocardiography showed complete closure of the fistula.
Patients were followed up in the outpatient clinic by ECG, chest radiography, phonocardiography, and trans-thoracic Doppler echocardiography. In some cases, selective coronary angiography and myocardial radio-nuclide scintigraphy were also carried out. Complete normalization of the ECG was noted in 12 patients (80%). The patient with pre-procedural Wolff-Parkinson-White syndrome had the same short PR interval. No cardiac murmurs were noted on the phonocardiograms. Trans-thoracic Doppler echocardiography showed fistula occlusion in all patients. Radionuclide scans of the myocardium were performed in 7 patients under loading conditions. Loading was increased to achieve submaximal heart rate or the onset of limited symptoms. In the last minute of maximal loading, thallium-201 was injected intravenously to determine myocardial blood flow. In all cases, the loading was halted because of fatigue; the threshold of loading was 87.0 ± 5.2 W. In all children, the result of the loading test was negative; there was a significant improvement of perfusion in areas supplied by the anomalous coronary artery. Selective coronary angiography in 4 patients after 2 to 13 years confirmed occlusion of the fistula in all of them. The 13 survivors with coils were followed up for 6 months to 13 years. Follow-up was 100% complete. All patients were symptom-free and leading an active life.
|
DISCUSSION
|
|---|
Although congenital coronary-cardiac fistulae are rare with an incidence of approximately 1:500 coronary angiographies, the defect is usually hemodynamically significant.3,8 Because of this, diagnosis of coronary-cardiac fistula is considered to be an indication for surgery. Early closure is necessary to prevent complications such as refractory congestive heart failure, myocardial ischemia, or infective endocarditis, which usually develop in the second or third decades of life.6,12 The transcatheter coil closure technique is a valuable alternative to open surgery nowadays. However, the closure of such fistulae remains one of the most complicated interventional procedures, and there are few reports regarding this form of treatment, due to the rarity of the defect.13–16
Technical problems related to the use of coils include difficulties in accurate identification of the site for adequate occlusion to avoid the risk of closure of normal coronary branches during embolization.14 Visualization of these arteries when the fistula is still functioning is often impossible because of the steal phenomenon.4 Distal passage of a guidewire and catheter through a tortuous coronary artery with a thin wall and stable placement of a deployment catheter in the target area are difficult and sometimes risky, as seen in this experience where perforation of the fistula occurred in one case.14 Because of catheter manipulation within the coronary arteries, there is the potential risk of coronary complications.16 Continuous monitoring of the ECG can detect transient myocardial ischemia during catheterization, and beta blockers or nitroglycerin can be injected if there is persistent ischemia. Some reports are available to support the low incidence of intra-procedural coronary complications.6,10,13–15
For closure of fistulae, Gianturco coils, platinum microcoils, and independent silicon and latex balloons or a combination of coils and balloons have been used.9,15–17 Reidy and colleagues4,16 reported closure of fistulae in 7 patients, using independent balloons in 3, platinum microcoils in 3, Gianturco coils in 1, and a combination of balloons and microcoils in 1 case. Embolization was successful in 6 patients and unsuccessful in one because of balloon damage. The need for catheters with large diameters, and the possibility of early emptying or deployment of the balloon are drawbacks of this technique. Inadequate embolization or migration of coils into the pulmonary system are also potential complications.16 Coaxial embolization by platinum microcoils is considered effective and allows a simple distal occlusion.16 The Gianturco coil is a well-known commercially available embolization device.10 Good immediate and long-term results confirm the efficiency and safety of Gianturco coil occlusion.18
Occlusion of the fistula must be performed distally to prevent closure of normal coronary branches, and also away from the drainage orifice to prevent coil migration into a heart chamber. When passage and stable placement in the target area are technically achievable, the dimension of the implanted coil must exceed the accurately calculated diameter of the fistula by at least 30% to prevent coil repositioning or migration. Embolization of a coronary fistula can be difficult, and complications are likely to occur in the absence of fistula stenosis, with proximal or lateral localization of the drainage orifice, or the presence of several drainage orifices, when there is a large aneurysm of the anomalous coronary artery (especially if localized distally), single coronary artery, or concomitant organic heart disease requiring surgical correction. In these cases, the experience of the surgeon and assessment of the relative risks of endovascular and routine surgical treatment are crucial. In view of the immediate and long-term results of transcatheter coil occlusion, this technique can be considered as an alternative to surgery for coronary fistula. It may decrease the risk of myocardial damage and avoid a thoracotomy and use of cardio-pulmonary bypass in some cases. The technique is also cost effective with a short postprocedural hospital stay and minimal period of rehabilitation.
|
REFERENCES
|
|---|
-
Kirklin JW, Barratt-Boyes BG. Cardiac surgery. New York: Wiley, 1987:945–55.
-
McNamara JJ, Gross RE. Congenital coronary artery fistula. Surgery
1969;65:59–69.[Medline]
-
Baltaxe HA, Wixson D. The incidence of congenital anomalies of the coronary arteries in the adult population. Radiology
1977;122:47–52.[Abstract]
-
Reidy JF, Anjos RT, Qureshi S. Transcatheter embolization in the treatment of coronary artery fistulas. J Am Coll Cardiol
1991;18:187–92.[Abstract]
-
Burakovsky VP, Bockeria LA. Cardiovascular surgery. Moscow: Medicine, 1996:359–62.
-
Rittenhouse EA, Doty DB, Ehrenhaft JL. Congenital coronary artery cardiac chamber fistula: review of operative management. Ann Thorac Surg
1975;20:468–85.[Abstract]
-
Haberman JH, Howard ML, Johnson ES. Rupture of the coronary sinus with hemopericardium. Circulation
1963;28:1143–4.[Abstract/Free Full Text]
-
Wilde P, Watt I. Congenital coronary artery fistulae: six new cases with a collective review. Clin Radiol
1980; 31:301–11.[Medline]
-
Alekyan BG. Endovascular surgery in the treatment of some congenital heart diseases [Russian]. Ann Surg
1996;3:54–63.
-
Alekyan BG, Podzolkov VP, Cárdenas CE, Mitina IN. Transcatheter coil embolization of the coronary fistulae. Ninth complex coronary angioplasty course book (endovascular therapy course, coronary and peripheral). Paris: Europa Edition, 1998:767–77.
-
Velvis H, Schmidt KG, Silverman NH. Diagnosis of coronary artery fistula by two-dimensional echocar-diography, pulsed Doppler ultrasound and color-flow imaging. J Am Coll Cardiol
1989;14:966–76.
-
Levin DC, Fellows KE, Abrams HL. Hemodynamically significant primary abnormalities of the coronary arteries. Circulation
1978;58:2503–4.
-
Bennett JM, Maree E. Successful embolization of a coronary artery fistula. Int J Cardiol
1989;23:405–6.[Medline]
-
Issenberg HJ. Transcatheter coil closure of a congenital coronary arterial fistula. Am Heart J
1990;120:1441–3.[Medline]
-
Moskowitz WB, Newkumet KM, Albrecht GT, Goble MM, Schieken RM. Case of steel versus steal: coil embolization of congenital coronary arteriovenous fistula. Am Heart J
1991;121:909–11.[Medline]
-
Reidy JF, Tynan MJ, Qureshi S. Embolization of a complex coronary arteriovenous fistula in a 6-year-old child: the need for specialised embolisation techniques. Br Heart J
1990;63:246–8.[Abstract/Free Full Text]
-
Gianturco C, Anderson W, Wallace S. Mechanical devices for arterial occlusion. Am J Roentgenol
1975;124:428–35.[Abstract]
-
Kienast W, Emmrich K, Schultz J, Kroger JC, Francke A, Lenschow U. Coronary artery fistula — surgical or percutaneous embolization treatment [German]? Z Kardiol
1993;82:436–42.[Medline]
TOP
ABSTRACT
INTRODUCTION
