Myocardial ischemia in generalized coronary artery–left ventricular microfistulae

International Journal of Cardiology 63 (1998) 47–52

Myocardial ischemia in generalized coronary artery–left ventricular microfistulae ¨ Ulrich Stierle*, Evangelos Giannitsis, Abdolhamid Sheikhzadeh, Jurgen Potratz ¨ Curschmann Klinik, 23669 Timmendorfer Strand, Department of Cardiology ( Internal Medicine II) Medical University, Lubeck , Germany Received 1 September 1997; accepted 9 October 1997

Abstract Generalized (multiple) arterio-systemic fistulae are fistulae arising from all three major coronary arteries and drain into the left ventricle are rare and the clinical and hemodynamic sequelae are incompletely understood. This communication is based on the clinical and hemodynamic data of a series of patients (eight cases out of 7 262 consecutive patients) incidentally identified at coronary angiography combined with data from cases previously reported in literature. The aim was to assess the role of generalized coronary artery fistulae as a non-atherosclerotic cause of myocardial ischemia by means of a coronary sinus lactate study. Coronary sinus lactate study demonstrated myocardial ischemia in 6 / 7 patients. Mean arterio-coronary venous lactate difference decreased from 0.3160.18 mmol / l (lactate extraction ratio, LER, 29.4613.9%) at rest to 0.0460.13 mmol / l (LER –4.0613.3%) at peak exercise. Five minutes after cessation of pacing, lactate difference increased to 0.2260.21 mmol / l (LER – 20.7613.2%). At peak pacing stress, 4 / 7 patients showed frank lactate production, and two patients presented with a reduced cardiac lactate extraction rate also indicating myocardial ischemia metabolically. In the present study, we demonstrated a possible role of a coronary steal mechanism due to microfistulae pathways in the pathogenesis of myocardial ischemia in patients with generalized coronary artery–left ventricular microfistulae.  1998 Elsevier Science Ireland Ltd. Keywords: Congenital coronary–systemic artery fistula; Myocardial ischemia

1. Introduction Congenital coronary arterial fistulae are the most prevalent hemodynamically significant congenital malformations of the coronary arterial circulation. Most of the fistulae arise from either the right or left coronary artery and drain into the right side of the heart [1]. Generalized (multiple) arterio-systemic fistulae are fistulae arising from all three major coronary arteries and drain into the left ventricle are

*Corresponding author. Tel.: 149 4503-602-0; fax: 149 4503-602660.

exceptionally rare and the clinical and hemodynamic sequelae are incompletely understood [1]. However, myocardial ischemia and diastolic volume overload of the left ventricle resulting from left-to-left-shunt are among the most commonly reported hemodynamic consequences of generalized coronary arteriosystemic fistulae [6]. This communication is based on the clinical and hemodynamic data of a series of patients incidentally identified at coronary arteriography combined with data from cases previously reported in the literature [2,7–15,17]. The aim of the clinical and hemodynamic evaluation was to assess the role of generalized coronary artery fistulae as a non-atherosclerotic cause of myocardial ischemia.

0167-5273 / 98 / $19.00  1998 Elsevier Science Ireland Ltd. All rights reserved. PII S0167-5273( 97 )00280-5

48

U. Stierle et al. / International Journal of Cardiology 63 (1998) 47 – 52

2. Patients and methods The data base consisted of 7 262 consecutive patients who underwent cardiac catheterization with coronary arteriography at the University Hospital of Luebeck. There were 90 cases who were reported as having an anomalous coronary artery (‘angiographic incidence’ 1.2%). Twenty-four patients (0.3%) had congenital fistulous communications between the coronary arterial circulation and a cardiac chamber or the pulmonary artery. Eight patients with generalized (multiple) coronary artery microfistulae arising from all three major coronary arteries emptying into the left ventricle underwent further evaluation for myocardial ischemia by means of exercise stress testing, thallium-201 perfusion study, and coronary sinus lactate study in seven cases. Coronary sinus lactate study for metabolic assessment of exercise-induced myocardial ischemia was performed the day after diagnostic catheterization. All patients gave written consent for the study. A 7F Zucker catheter (USCI, Bard, Ireland) was placed into the coronary sinus for incremental atrial pacing and blood sampling. The tip of the catheter was positioned approximately two to five centimetres from the orifice of the coronary sinus to minimize admixture of right atrial blood in the samples. The correct position was confirmed by hand injection of dye. At the beginning of the study all patients were asymptomatic and a 12-lead ECG did not show any evidence of myocardial ischemia. Incremental atrial pacing was started at 100 beats per minute (bpm) and was increased stepwise by 20 bpm every three minutes and was terminated after three pacing periods at 160 bpm. Paired blood samples for determination of serum lactate were withdrawn from the coronary sinus (CV) and an indwelling catheter in the radial artery (A) at rest, at the end of each pacing step and one, three, and five minutes after cessation of pacing. Per definition, positive values of the arterio-coronary venous lactate difference represent lactate extraction, negative values lactate production. Metabolic evidence of myocardial ischemia was defined as a reduction of cardiac lactate extraction ratio (LER) ,10% or a frank lactate production (LER ,0%, [4]). LER (in %) was calculated according to the equation LER5[(A)2(CV)]3(A)21 3100. Pacing was discontinued prematurely with the advent of incapacitating

anginal symptoms or ST segment depressions .0.4 mV in at least two corresponding ECG-leads. Conventional criteria were employed for exercise stress testing and thallium-201-exercise perfusion scans. For descriptive purposes, data are presented as mean with standard deviation. For the analysis of the relation between baseline values and post-pacing values a repeated measure analysis of variance was used, with post hoc comparisons tested with the Scheffe’s test to identify significant differences. Differences were considered statistically significant at a level of p,0.05.

3. Results Eight of 24 cases with fistulous communications (all female, mean age 50.5 years, range 46–69) had generalized (multiple) coronary artery microfistulae arising from all three major coronary arteries draining exclusively into the left ventricle. Of these patients, six complained of stable angina and three had suffered myocardial infarction; in two cases typical unstable angina was present. Two patients had dyspnea at effort, while none had clinically overt congestive heart failure. Only four patients had a systolic murmur, while none had continuous or diastolic murmur suggestive of a coronary fistula. The electrocardiogram was unsuspicious in all but the three patients who had suffered non-Q myocardial infarction (case 1: terminal negative T waves in leads aVL, V 1–3; case 6: transient terminal negative T waves in leads V 5–6; case 6: poor R progression in leads V 1–3, terminal negative T waves in leads aVL and V 4–6). The chest roentgenogram was unremarkable as was the two dimensional echocardiogram including colour, continuous and pulsed wave Doppler. Five patients underwent exercise testing of whom 4 / 5 developed myocardial ischemia. In both patients with unstable angina no exercise stress testing nor Tl-201 perfusion scan was performed. Right and left heart catheterization did not show any pathological findings of the central hemodynamics. Coronary arteriography revealed multiple fistulae arising from all major branches of the coronary arteries draining exclusively into the left ventricle. Atherosclerotic coronary artery disease was ruled out in all patients. From the anatomic aspect, fistulae were best classified into the

U. Stierle et al. / International Journal of Cardiology 63 (1998) 47 – 52

arterio-sinusoidal type of coronary artery fistulae emptying diffusely into a fine network of non-obliterated intramyocardial sinusoids (‘microfistulae’, Fig. 1). The proximal parts of the coronary arteries were dilated and tortuous in four of the eight cases. Left

49

ventriculograms obtained in two planes were unremarkable in all but two patients with a history of myocardial infarction (case I: antero-apical akinesia; case 6: apical hypokinesia). Coronary sinus lactate study demonstrated myocar-

Fig. 1. (A) Selective injection of the left coronary artery at 608 in the LAO projection. Opacification of the left ventricular cavity through a network of microfistulae originating from the left anterior descending and left circumflex artery is visible. (B) Selective injection of the right coronary artery at 508 RAO projection (same patient as in Fig. 1(A)). There is a stream of dye into the left ventricle via a maze of fine vessels from all coronary arterial branches.

U. Stierle et al. / International Journal of Cardiology 63 (1998) 47 – 52

50

dial ischemia in 6 / 7 patients. Mean AV–lactate difference decreased from 0.3160.18 mmol / l at rest to 20.0460.13 mmol / l at peak exercise ( p,0.05). Five minutes after cessation of pacing, mean AV– lactate difference increased to 0.2260.21 mmol / l. At peak pacing stress, 4 / 7 patients showed frank lactate production indicating a change of the myocardial metabolism due to myocardial ischemia. Additionally, two patients presented with a reduced cardiac lactate extraction also indicating myocardial ischemia. Only one patient showed an unchanged arterio-coronary venous lactate difference according to a normal, non-ischemic metabolism at peak pacing stress. Table 1 shows the calculated cardiac lactate extraction ratios (LER) at rest, peak pacing stress and 5 min post cessation of pacing in 7 / 8 cases. Cardiac lactate extraction ratio decreased from 29.4613.9% at rest to 24.0613.3% at peak pacing rate ( p,0.05). Five minutes after cessation of pacing, cardiac lactate extraction ratio increased to 20.7613.2%. At peak pacing rate, 5 / 7 patients suffered from typical angina pectoris, only two patients were free of chest pain and reported only palpitations which abolished immediately after cessation of pacing. A survey of the clinical features and diagnostic tests performed in the eleven cases reported in the literature and the eight study cases is given in Table 2. Table 1 Arterial–coronary venous (AV) lactate (in mmol / l) differences and lactate extraction ratio (LER, in %) at rest, peak pacing stress and five minutes after cessation of pacing Rest Differences in AV lactate Case 2 0.50 Case 3 0.16 Case 4 0.18 Case 5 0.12 Case 6 0.58 Case 7 0.32 Case 8 0.31 Mean6SD 0.3160.18 LER Case 2 34 Case 3 17 Case 4 19 Case 5 14 Case 6 43 Case 7 28 Case 8 51 Mean 29.4613.9

Peak pacing stress 0.14 20.17 20.10 0.03 0.09 20.20 20.04 20.0460.13 16 218 212 3 8 218 27 24.0613.3

5 min post pacing 0.13 0.14 0.11 0.14 0.69 0.23 0.08 0.2260.21 17 16 13 15 50 21 13 20.7613.2

4. Discussion In consecutive angiographic series, congenital coronary arterial fistulae are reported with an incidence of 0.08–0.3% [3]. Among these coronary fistulae, only about 10% arise from either coronary artery and drain into the left heart chambers (arterio-systemic fistulae, [1]). Anatomic and clinical findings vary depending on the clinical situation in which coronary artery fistulae are identified [5]. In angiographic surveys, coronary artery fistulae are more commonly incidental findings at the time of angiography, and anatomic and clinical features are different compared with data from case reports. Fistulae identified incidentally are more often multiple, originate from the left coronary and drain predominantly into the pulmonary artery or the left ventricle, whereas fistulae described in case reports are more frequently arteriovenous with draining into the right side of the heart, have an audible cardiac murmur and are more likely to be symptomatic [5]. Multiple coronary artery fistulae arising from all three major coronary arteries emptying into the left ventricle are extremely rare and little is known about the anatomic and clinical features and the hemodynamic consequences of multiple left-to-left-shunts. A total of eleven similar cases have been described in the literature ( [2,7–15,17]) and another eight are added in this report. In general, three types of fistulous communication can be distinguished morphologically: 1) the arterialluminal type – the most common – where a coronary artery enters directly the cardiac chamber; 2) the arterio-sinusoidal type, where communication is through a myocardial sinusoidal network; and 3) the arterio-capillary type, where the arterial vessel drains into the capillaries [2]. The former two forms bypass capillaries for oxygen delivery to the myocardium and may give rise to myocardial ischemia. Since the normal circulation usually impacts a greater resistance to flow than the fistula, coronary steal may be essential in the pathogenesis of myocardial ischemia. All cases of our series had multiple fistulous communications of the arterio-sinusoidal type. Angina and myocardial infarction were the most predominant symptoms suggesting presence of myocardial ischemia in the absence of atherosclerotic coronary artery disease. In most cases evidence for

U. Stierle et al. / International Journal of Cardiology 63 (1998) 47 – 52

51

Table 2 Survey of clinical findings and the results in stress testing in cases reported in literature (n511) and the present series (n58) Author

Age / gender

Angina

Tl-201

Lactate study

Reddy [7] Kinard [8] Rose [9] Cha [17] Vogelbach [10] Vogelbach [10] Chia [11] Sheikhzadeh [12] McLellan [13] Berberich [2] Yokawa [15] Case 1 Case 2 Case 3 Case 4 Case 5 Case 6 Case 7 Case 8

58 / M 60 / F 62 / M 54 / M 59 / F 53 / F 43 / M 55 / F 48 / M 65 / F 62 / F 49 / F 46 / F 68 / F 67 / F 57 / F 53 / F 67 / F 69 / F

1 n.d. 1 2 2 n.d. 2 2 1 1 1 n.d. 1 1 1 1 1 2 1 n.d. 2 n.d. 1 n.d. 1 1 1 1 1 n.d. 1 1 1 1 1 2 1 n.d. Cases reported in literature

Ergometry

n.d. n.d. n.d. n.d. 2 2 n.d. 2 n.d. n.d. n.d. n.d. 2 1 n.d. 1 1 n.d. n.d. Present series

n.d. n.d. n.d. n.d. n.d. n.d. n.d. 1 n.d. n.d. n.d. n.d. 2 1 1 1 1 1 1

Angina typical atypical unstable

9 / 11 1 / 11 2

6/8 2 2/8

Myocardial infarction Sudden cardiac death Dyspnea Palpitations Heart failure

1 / 11 1 / 11 3 / 11 3 / 11 2

3/8 2 2/8 3/8 2

Cardiac murmur systolic diastolic

4 / 11 1 / 11

4/8 2

ECG changes Non-Q-infarction Left ventricular hypertrophy Unspecific STT changes

1 / 11 3 / 11 9 / 11

3/8 2 4/8

Chest X-ray Cardiomegaly Pulmonary venous congestion

2 / 11 2

2 2

Angiography Dilated coronary arteries Left ventricular dilatation

9 / 11 3 / 11

4/8 2

M – male; F – female, n.d. – not done; 1 – present or positive result; 2 – absent or negative result.

myocardial ischemia was provided by thallium scan, ergometry and assessment of myocardial metabolism during pacing-induced ischemia by means of coronary sinus lactate sampling. In the absence of concomitant atherosclerotic coronary artery disease or left ventricular hypertrophy causing an oxygen demand-supply imbalance, coronary steal appears to be of major importance in the

pathogenesis of myocardial ischemia in cases with generalized arterio-systemic microfistulae. As a second major hemodynamic consequence, left-to-left shunt into the left ventricle bears the potential to cause diastolic volume overload [6]. However, in our series, absence of ventricular enlargement or hypertrophy as well as the absence of a continuous or diastolic cardiac murmur indirectly

52

U. Stierle et al. / International Journal of Cardiology 63 (1998) 47 – 52

suggest a small shunt volume. In support with our findings, Vachon et al. reported on 62 reviewed cases with arterio-systemic coronary fistulae into the left ventricle [6]. Single fistulae and multiple fistulae could be distinguished from each other by the clinical presentation. While diastolic overload of the left ventricle giving the clinical picture of aortic incompetence was predominant in single fistulae, myocardial ischemia and absence of a continuous murmur was more prevalent in multiple coronary artery fistulae [6]. Little is known about the natural history of multiple arterio-systemic coronary artery fistulae. However, the high incidence of myocardial infarction and anginal symptoms supports the potential hazard of coronary steal mediated severe myocardial ischemia. The management of patients with coronary artery fistulae remains controversial. Clear indications for surgical intervention include the presence of large shunts, other major cardiac lesions, concomitant atherosclerotic coronary artery disease, symptoms, or complications including progressive enlargement, bacterial endocarditis, rupture, pulmonary hypertension, and thromboembolism [16]. The natural history of patients with small fistulae is less well defined than that of large, symptomatic fistulae. In general, the incidence of both residual and recurrent fistulae averaged 9% in five series and usually occurred in patients with complex fistulae exhibiting multiple entry points in the receiving chamber [5]. However, there is no experience with surgical correction or percutaneous closure using a detachable balloon in patients with multiple arterio-systemic (left ventricular) coronary fistulae, to date. Medical therapy with nitrates in conjunction with beta-blockers [11,12] or calcium-antagonists [13] proved efficient in the few cases where considered or mentioned. Consecutively, all our patients were treated with beta-blockers and nitrates and were advised as to the need of infective endocarditis prophylaxis. In this study, we demonstrated a possible role of a coronary steal mechanism due to microfistulae pathways in the pathogenesis of myocardial ischemia in

patients with generalized coronary artery–left ventricular microfistulae.

References [1] Perloff JK. Congenital coronary arterial fistula. In: Perloff JK, editor. The clinical recognition of congenital heart disease. Philadelphia: Saunders, 1987:511. [2] Berberich SN, Zager JRS, Herman NP, Eslava RL. Diffuse bilateral coronary artery fistulae entering the left ventricle. Vasc Surg 1978;12:204–9. [3] Yamanaka O, Hobbs RE. Coronary artery anomalies in 126 595 patients undergoing coronary arteriography. Cathet Cardiovasc Diagn 1990;21:28–40. [4] Jackson G. Laboratory diagnosis of myocardial ischemia. In: Schroeder JS, editor. Invasive cardiology. Philadelphia: Davies, 1985:45–65. [5] Sapin P, Frantz E, Jain A, Nichols TC, Dehmer GJ. Coronary artery fistula: an abnormality affecting all age groups. Medicine 1990;69:101–13. [6] Vanchon JM. Les fistules congenitales coronaro-ventriculaires gauches. Ann Cardiol Angiol 1983;32:21–5. [7] Reddy K, Gupta M, Hamby RI. Multiple coronary arteriosystemic fistulas. Am J Cardiol 1974;33:304–6. [8] Kinard S. Hypoplasia of the coronary sinus with coronary drainage into the left ventricle by way of the Thebesian system. Chest 1975;68:384–5. [9] Rose AG. Multiple coronary arterioventricular fistulae. Circulation 1978;58:179–80. [10] Vogelbach KH, Edmiston WA, Stenson RE. Coronary artery–left ventricular communications: a report of two cases and review of the literature. Cathet Cardiovasc Diagn 1979;5:159–67. [11] Chia BL, Chan ALK, Tan KKA, Ng RAL. Coronary artery–left ventricular fistula. Cardiology 1981;68:167–79. [12] Sheikhzadeh A, Stierle U, Langbehn AF, Thoran P, Diederich KW. Generalized coronary arterio-systemic (left ventricular) fistula: case report and review of literature. Jpn Heart J 1986;27:533–44. [13] McLellan BA, Pelikan PCD. Myocardial infarction due to multiple coronary ventricular fistulas. Cathet Cardiovasc Diagn 1988;18:317– 20. [14] Berberich SN, Zager JRS, Herman NP, Eslava RL. Diffuse bilateral coronary artery fistulae entering the left ventricle: a case confirmed surgically. Vasc Surg 1978;12:204–9. [15] Yokawa S, Watanabe H, Kurosaki M. Asymptomatic left and right coronary artery–left ventricular fistula in an elderly patient with a diastolic murmur only. Int J Cardiol 1989;25:244–6. [16] Hobbs RE, Millit HD, Raghavan PV, Moodie DS, Sheldon WC. Coronary artery fistulae: a 10-year review. Clev Clin Q 1982;49:191–7. [17] Cha SD, Maranhao V, Goldberg H. Silent coronary artery–left ventricular fistula: a disorder of the Thebesian system. Angiology 1978;29:169–73.