This article describes a very rare vascular condition in a dog. To the authors’ knowledge, this is the first report of a coronary arterial shunt from the LCA to the MPA in a dog. The diagnosis of this congenital abnormality was achieved through a multimodal imaging approach including TTE and CT imaging. Cases of CAF in veterinary medicine have been described for canine, feline, bovine, and camelid patients [1,2,3,4,5,8,9,11,12,13]. Fistulous connections between the coronary arterial system can communicate directly to the cardiac chambers (coronary cameral fistula), to a great vessel (coronary arteriovenous shunt) [8] such as the MPA or the cava vein or can finish inside the cardiac myocardium (aortocardiac fistula) [5,11,13]. The last type seems to be very serious in horses [11,13], where the CAF may finish into the interventricular septum, which results in ventricular arrhythmias, left-sided congestive heart failure (CHF) or wall rupture [11,13]. Further classification of CAFs may include the size and number of fistulae present, with solitary macro fistulas being more common than multiple micro fistulas [14]. We presume that its origin may be congenital because there was no evidence of trauma or surgery of the thorax, neither local nor systemic infection nor neoplastic process from the history and other examinations including radiographs, CT scan, abdominal ultrasound or blood tests. In general, congenital CAFs arise secondary to embryonic developmental disturbances resulting in the persistence of the vascular network which nourishes embryonic ventricular walls. These remnants are known as Thebesian vessels, which provide direct connections between the coronary vasculature and the heart chambers, bypassing the coronary sinus [12].
In humans, most CAFs (90%) drain into the right side of the heart, with the other 10% terminating into left-sided chambers [1,8]. In asymptomatic human patients, CAFs typically originate from the LCA and often drain into the MPA [8]; however, this type of connection has not been described in the veterinary literature until now. The presence or absence of clinical signs is thought to be determined by the size, location, and configuration of the fistula itself [1]. In our patient, a murmur was not detected during clinical examination. This was probably due to the small size and flow velocity through the shunt. These anatomic facts can also explain, -at least partially-, why our patient remains without clinical signs and even underwent anesthesia without complications. Most of CAF cases in humans have a continuous or more commonly diastolic murmur [1]. Medium to large CAFs can result in volume overload and chamber remodeling, which can lead to CHF, exercise intolerance, endocarditis, myocardial ischemia [1], fistula rupture [4,9,12] and arrhythmias, particularly atrial fibrillation, if there is atrial enlargement [9]. Furthermore, the dog described in this report had no obvious evidence of coronary steal syndrome (when blood is preferentially shunted through the fistula rather than perfused through the myocardium, which leads to myocardial ischemia), as there were no clinical signs, no arrhythmias, normal left ventricular function [1,12,15] and no elevation of cardiac markers [1].
A final diagnosis was achieved through echocardiography and CT imaging, which are helpful complementary diagnostic tools for obtaining more precise information about coronary vessel anatomy and confirming the lesions. Standard two-dimensional and Doppler echocardiography initially proved useful for evaluating the CAF and providing information on intraluminal fistular flow and its site of drainage. However, spectral Doppler failed to show a more realistic value of systolic and diastolic velocities, although it was able to define flow direction during systole and diastole. On the other hand, color Doppler echocardiography demonstrated intraluminal fistular blood flow and its communication with the MPA. The continuous mosaic color signal observed within the fistula indicated flow signal aliasing as a result of turbulent blood flow. Other turbulent jets between the proximal aorta and MPA can be observed in cases of aortopulmonary window; however, the lesion in our patient was located at the level of the aortic valve, where there is a dilated and interrupted Valsalva sinus. A supracristal ventricular septal defect and coronary cameral fistula were also ruled out because the flow finished directly in the MPA. However, echocardiography may not be sufficient for the assessment of extracardiac structures and due to the size of our patient, we chose to perform CT imaging in lieu of cardiac catheterization with angiography [1]. Two-dimensional reformation and three-dimensional reconstruction provided an accurate depiction of the morphologic structures of the aortic root, coronary arteries and CAF; thus the presence of more distal fistulae or an aortic tunnel (abnormal tubular extracardiac communication between the aortic root and other heart structures) were also ruled out [5,6,16].
Coronary angiography may help in further characterization of this lesion and is considered the gold standard for the diagnosis of CAFs in humans. Transesophageal, and 3D echocardiography, and magnetic resonance imaging have also proven useful [15]; however, these modalities were unavailable at the time of diagnosis.
No therapy was necessary. There was no evident dilation of the aorta or MPA, volume overload, or other changes compatible with ongoing pulmonary hypertension based on her hemodynamics (the Qp:Qs was estimated to be 1.19). Treatment options include surgical ligation of the fistula under cardiopulmonary bypass, percutaneous transcatheter closure, and medical management of the resulting volume overload and myocardial ischemia. Long-term medical management with antiplatelet drugs is indicated when aneurysmal dilation of the CAF causes blood stagnation and thrombosis [1,2]. None of the anterior splints were necessary for our patient who was free of clinical signs.