Murmurs and arrhythmias are commonly detected in equine athletes. Assessing the relevance of these cardiovascular abnormalities in the performance horse can be challenging. Determining the impact of a cardiovascular disorder on performance, life expectancy, horse and rider or driver safety relative to the owner's future expectations is paramount. A comprehensive assessment of the cardiovascular abnormality detected is essential to determine its severity and achieve these aims. This consensus statement presents a general approach to the assessment of cardiovascular abnormalities, followed by a discussion of the common murmurs and arrhythmias. The description, diagnosis, evaluation, and prognosis are considered for each cardiovascular abnormality. The recommendations presented herein are based on available literature and a consensus of the panelists. While the majority of horses with cardiovascular abnormalities have a useful performance life, periodic reexaminations are indicated for those with clinically relevant cardiovascular disease. Horses with pulmonary hypertension, CHF, or complex ventricular arrhythmias should not be ridden or driven.

Background: In humans, atrial fibrillation (AF) induces electrical, contractile, and structural remodeling leading to AF stabilization. Little is known about AF-induced atrial remodeling in horses. Hypothesis: Induced AF produces rapid atrial electrical and contractile remodeling in horses. Animals: Six horses, 5 animals completed the study. Methods: Each horse was instrumented with a pulse generator and pacemaker to maintain AF by burst pacing and to study atrial and ventricular electrophysiology (AF cycle length [AFCL], AF duration, and atrial/ventricular effective refractory period [AERP/VERP] at different pacing cycle lengths [PCL]). Left atrial and ventricular contractile remodeling were assessed echocardiographically by calculation of fractional changes in atrial and ventricular dimensions, respectively, during the cardiac cycle. Measurements were performed at baseline, a 7-day AF period and a 2-day recovery period. Results: Atrial electrical and contractile remodeling could be demonstrated after 4 and 12 hours of AF, respectively. A progressive shortening of the AERP (261 ± 39–171 ± 18 ms at a PCL of 1,000 ms, P < .0001), an attenuation of the AERP rate adaptation, a decrease in AFCL (239 ± 39–194 ± 7 ms, P < .0001), and a decrease in atrial FS (12 ± 3% to 0 ± 2%, P < .05) occurred. AF duration increased progressively and became persistent in 2 animals. VERP did not change significantly. Upon restoration of sinus rhythm, values returned to baseline within 48 hours. Conclusions and Clinical Importance: Atrial electrical and contractile remodeling appears rapidly. After 7 days of AF, reverse remodeling occurred within 2 days. These observations suggest that early conversion of AF might be beneficial for success rate and early return to training.

Automatic or semi-automatic analysis of the equine electrocardiogram (eECG) is currently not possible because human or small animal ECG analysis software is unreliable due to a different ECG morphology in horses resulting from a different cardiac innervation. Both filtering, beat detection to classification for eECGs are currently poorly or not described in the literature. There are also no public databases available for eECGs as is the case for human ECGs. In this paper we propose the use of wavelet transforms for both filtering and QRS detection in eECGs. In addition, we propose a novel robust deep neural network using a parallel convolutional neural network architecture for ECG beat classification. The network was trained and tested using both the MIT-BIH arrhythmia and an own made eECG dataset with 26.440 beats on 4 classes: normal, premature ventricular contraction, premature atrial contraction and noise. The network was optimized using a genetic algorithm and an accuracy of 97.7% and 92.6% was achieved for the MIT-BIH and eECG database respectively. Afterwards, transfer learning from the MIT-BIH dataset to the eECG database was applied after which the average accuracy, recall, positive predictive value and F1 score of the network increased with an accuracy of 97.1%.