Coronary Fistulae

The preceding case report by Kassaian and collaborators on covered-stent treatment of 3 coronary-to-pulmonary fistulae stimulates a few discussion points on a frequent and generally benign condition.

The Present Case

The patient, with angiographic evidence of coronary fistulae (untreated at the time of a prior bypass surgery), was admitted with atypical chest pain and a possibly positive nuclear scintigraphy stress test (inferobasal reversible ischemia is frequently seen after successful bypass surgery). The authors' estimate that the shunt carried half of the systemic output of about 3 L/min—as indicated by a QP/QS of 1.5—is really not supported by the angiographic evidence of small fistulae that the authors provided. Interestingly, the authors imply that the venous hum and the systolic murmur were related to these coronary fistulae (the hum and murmur disappeared after closure of the lesions). After angiographic study, the operators concluded that the fistulae required closure, and they performed a repair with 3 covered stents, which yielded excellent immediate results. The Jostent® that was used is a polytetrafluoroethylene graft sandwiched between 2 thin metallic stents. No long-term angiographic follow-up was provided.

Treatment Indications for Coronary Fistulae

Coronary fistulae display a wide spectrum of sizes and anatomic variants, and each of these carries different clinical implications. In general, coronary fistulae can cause nonspecific symptoms and are fairly frequent in the general population. We have reported that about 1% of a consecutive series of patients studied by coronary angiography has some kind of coronary fistulae, but only a very few of those lesions were considered important enough to treat.¹

In the catheterization laboratory, atypical chest pain is a frequent presenting symptom that can be associated with a number of benign conditions. Coronary fistulae can cause ischemia by 2 possible mechanisms: first, there can be a steal of nutrient blood flow to the fistulous tract from the normal coronary branches; and second, there can be stenosis of nutrient side branches secondary to thrombus associated with fistulous tracts, ulcerations, and atherosclerosis. Neither of these mechanisms appears to be supported by the angiographic imaging in the reported case, especially in view of the facts that both the right and left anterior descending coronary branches had patent grafts and that no coronary steal was documented.

Covered Stents

At this time, covered stents (like the Jostent) are indicated only for perforation of a coronary artery with free bleeding into the pericardium (usually, such a rupture is associated with coronary intervention). Their application is limited by a lack of long-term follow-up provided to the FDA, which approved the marketing of these stents only on the basis of humanitarian use, for use in patients who are inoperable but still need intervention. Mid-to-late results (at 0.5 to 3 years of follow-up) are clearly the major concern associated with covered-stent implantation. A substantial incidence of restenosis has been reported, and this manifests both as gradual obstruction of the treated vessel by intimal fibrocellular growth and as stent thrombosis.² The 2nd mode of stent failure, in particular, would be catastrophic, especially if it should occur in a large nutrient vessel treated by this technique.

Final Comments

Coronary anomalies appear frequently in the practice of every interventionist, but they are often poorly understood, and the treatment of most subgroups of coronary anomalies is not supported by consistent guidelines.³ The fact that many coronary anomalies can be elegantly treated in the catheterization laboratory should not tempt the operator to expand treatment indications without validation.

Paolo Angelini, MD
Department of Cardiology, Texas Heart Institute at St. Luke's Episcopal Hospital, Houston