Coronary Artery to Pulmonary Artery Fistula With Concomitant Atrial Fibrillation
Coronary artery fistulas are a rare entity and involve an abnormal connection between a coronary artery and another vessel or an adjacent chamber. We report a patient presenting with new onset heart failure in which a fistula between the left anterior descending and the pulmonary artery was incidentally diagnosed. Issues regarding the diagnosis, management and long-term surveillance of such fistulae are addressed.
A coronary artery fistula is an abnormal communication between coronary vessels and cardiac chambers or a great vessel. This was first reported in 1865 by Krauss et al.1 It is a rare cardiac anomaly, estimated to be present in 0.002% of the general population2 and constitutes <15% of all congenital coronary artery anomalies. The majority of patients with such coronary abnormalities are asymptomatic and the anomalies are usually found incidentally during coronary angiography. They have an estimated incidence of <1% in angiographic series.3
We present a case of left anterior descending artery (LAD) to pulmonary artery fistula discovered incidentally during the workup of a patient presenting with new-onset heart failure.
A 53-year-old otherwise healthy man with no previous medical history presented to our outpatient clinic for evaluation of new-onset exertional dyspnea over several months. He denied chest pain, palpitations, orthopnea, paroxysmal nocturnal dyspnea, syncopal episodes, or lower limb edema. The patient did not smoke or consume alcohol. On examination, his blood pressure was 102/60 mmHg, pulse rate was 80/min and irregular, and oxygen saturation was 98% on room air. Cardiac examination revealed an irregular heart rhythm, normal S1-S2, and no murmurs. The chest was clear on auscultation with no signs of jugular venous distension or lower extremity edema. The remainder of his physical examination was unremarkable.
A 12-lead electrocardiogram (EKG) showed atrial fibrillation (AF) at a rate of 100 beats per min. A complete blood count, basic metabolic profile, thyroid function tests, and cardiac enzymes were all within normal limits. The patient was started on metoprolol and aspirin. Echocardiography revealed a mildly enlarged left ventricle and severely reduced left ventricular systolic function with an estimated ejection fraction of 25%. There was diffuse global hypokinesia with apical akinesis, in addition to moderately reduced right ventricular systolic function with normal right-sided pressures and moderate bi-atrial enlargement. A subsequent stress test showed no evidence of ischemia and normal perfusion, and confirmed the reduced ejection fraction.
Despite a normal perfusion study, cardiac catheterization was pursued, given the reduced ejection fraction and regional wall motion abnormalities in the setting of new-onset heart failure. Coronary angiography revealed a coronary fistula connecting the left anterior descending artery (LAD) to the pulmonary artery (Figure 1). Mild atherosclerotic changes with significant ectasia were seen in the left circumflex and right coronary arteries, while the LAD itself had no significant atherosclerotic disease. Right heart catheterization was performed to assess for significant left-to-right shunt from the fistula. The cardiac output (7.56) and cardiac index (3.71) were normal with no evidence of shunting; the Qp/Qs (pulmonary flow/systemic flow) ratio was 1. The right-sided pressures and wedge pressure were within normal limits.
To further evaluate the size and course of the coronary fistula, given a relevance to further management, the patient underwent a gated cardiac computed tomography (CT) angiogram (Figure 2; Video 1). This showed an abnormal tortuous communication from the LAD to the anterior aspect of the pulmonary outflow tract with tortuosity seen at the distal end. The findings confirmed a small left coronary to pulmonary outflow tract fistula. No other congenital abnormalities were detected.
Given the normal filling pressures, absence of a significant shunt, and the presence of new onset AF with atrial enlargement, the patient was diagnosed with rate-related cardiomyopathy and a coincidental coronary-to-pulmonary artery fistula. As such, therapy was directed toward controlling his arrhythmia and the fistula was to be followed conservatively. After appropriate anticoagulation and failed attempts at rhythm control, including cardioversion and sotalol initiation, we elected to pursue a rate-control strategy with metoprolol. Subsequent repeat evaluation after two months showed that the patient remained in AF with adequate rate control. His symptoms had essentially resolved and a repeat echocardiogram demonstrated recovery of left ventricular systolic function with an ejection fraction of 55%. The right-sided chamber enlargement had also improved and the patient tested positive for obstructive sleep apnea, for which he is being treated.
Serial monitoring of the fistula with repeat CT imaging is planned in one to two years and an echocardiogram will be repeated every three to five years, or earlier with symptoms, to monitor for chamber enlargement or reduction in ejection fraction, as recommended by the American College of Cardiology/American Heart Association guidelines.4
The case demonstrates that not all coronary artery fistulae require intervention or closure; the patient’s functional status, symptoms, and results of non-invasive and/or invasive testing should guide management.
Coronary artery fistulas can be acquired or congenital, and the majority of patients are usually asymptomatic. However, they can present with symptoms of angina, congestive heart failure, or arrhythmia. Ischemia is commonly seen in patients with coronary fistulas due to the “coronary steal phenomenon”5 where blood is shunted away from the coronary beds to the low-pressure pulmonary circulation during diastole, resulting in ischemia in the affected territories.
Our patient presented with new-onset heart failure and AF. The coronary fistula was discovered coincidentally, and given the normal filling pressures and lack of significant shunt, the likelihood that the fistula was the cause of his presentation was felt to be very low. It is theoretically possible that the AF was caused by shunting of blood from LAD to the pulmonary outflow tract and pulmonary venous dilation. However, normal right-sided chamber and wedge pressures with lack of a significant right-to-left shunt would make this scenario improbable.
The duration of the patient’s AF is unclear. In all likelihood, the patient had been in AF for some time, given the structural changes on echocardiography and since he presented after developing tachycardia-induced cardiomyopathy and subsequent heart failure symptoms.
The treatment of asymptomatic coronary-to-pulmonary fistulas is currently debatable; however, the main factors to consider when deciding on management include the relative size of the fistula, age of the patient, presence of symptoms, and the risk versus benefit of an invasive approach.6
Treatment options for coronary artery fistulae include surgical closure or trans-catheter embolization/closure. Armsby et al reported similar early effectiveness, morbidity, and mortality between surgical and transcatheter closure of coronary fistulas.7 The different endovascular options include different types of coils, detachable balloons, umbrella devices, covered stents, and dedicated vascular plugs.7 However, in cases with extreme tortuosity of the vessels, multiple draining sites or coronary branches at the site of the fistula7, the percutaneous approach may not be feasible and surgery is required.
In conclusion, coronary fistulae are rare entities that are usually coincidental findings. Patients are often asymptomatic, but may present with a variety of symptoms including angina, congestive heart failure, myocardial infarction, and arrhythmias. Therapeutic decision-making should be guided by the patient’s clinical presentation, results of testing, and patient preference. In general, a conservative route should be preferred in small, asymptomatic, coronary artery fistulas. However, the patient should be followed closely with repeat imaging to serially monitor for development of symptoms, size enlargement of the fistula, or deleterious effects of the fistula on surrounding structures.
- Krause W. Uber den ursprung einer akzessorischen a. coronaria aus a. pulmonalis [in German]. Z Ratl Med. 1865; 24: 225-227.
- Dodge-Khatami A, Mavroudis C, Baker CL. Congenital heart surgery nomenclature and database project: anomalies of the coronary arteries. Ann Thorac Surg. 2000; 69: 270-297.
- Angelini P, Velasco JA, Flamm S. Coronary anomalies: incidence, pathophysiology, and clinical relevance. Circulation. 2002; 105: 2449-2454.
- Warnes CA, Williams RG, Bashore TM, et al. ACC/AHA 2008 Guidelines for the Management of Adults with Congenital Heart Disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (writing committee to develop guidelines on the management of adults with congenital heart disease). Circulation. 2008 Dec 2; 118(23): e714-e833. doi: 10.1161/CIRCULATIONAHA.108.190690.
- Brooks CH, Bates PD. Coronary artery-left ventricular fistula with angina pectoris. Am Heart J. 1983 Aug; 106(2): 404-406.
- Latson LA. Coronary artery fistulas: how to manage them. Catheter Cardiovasc Interv. 2007 Jul 1; 70(1): 110-116.
- Armsby LR, Keane JF, Sherwood MC, et al. Management of coronary artery fistulae. Patient selection and results of transcatheter closure. J Am Coll Cardiol. 2002 Mar 20; 39(6): 1026-1032.
1Division of Cardiovascular Medicine, University of Toledo Medical Center, Toledo, Ohio; 2Research volunteer, University of Toledo Medical Center, Toledo, Ohio; 3Department of Radiology, University of Toledo Medical Center, Toledo, Ohio
Disclosures: The authors report no conflicts of interest regarding the content herein.
The authors can be contacted via Ehab Eltahawy, MD, MPH, FACC, FSCAI, FSVM, Program Director, Interventional Cardiology Fellowship; Medical Director, Cardiac Catheterization Laboratories; Assistant Professor of Medicine
Division of Cardiovascular Medicine, The University of Toledo Medical Center
Toledo, Ohio, at firstname.lastname@example.org.