Cycle length variation assists in differentiation of atrial, ventricular, and sinus arrhythmias in horses at rest.

Objective: Criteria for arrhythmia classification in horses are insufficient, particularly when differentiating atrial premature complexes (APCs) from sinus arrhythmia (SA). This study investigated RR and PP interval changes between SA, APCs, and ventricular premature complexes (VPCs). Methods: 30 resting ECGs were reviewed: 10 each containing SA, APC, and VPC. Sequences of 10 RR and PP intervals were recorded for each arrhythmia (20 events/ECG). Percent RR and PP deviations were calculated (1) between sequential intervals (%ΔSEQPP and %ΔSEQRR, respectively) and (2) compared to averaged baseline intervals 1 to 3 (%ΔAVGPP and %ΔAVGRR, respectively). Percent RR and PP deviations were compared between arrhythmias with 2-way ANOVA and receiver operator characteristic curve analysis. Results: Cycle length variation differentiated SA from APCs and VPCs in the degree of prematurity and return cycle length by use of both RR and PP intervals. The %ΔAVGRR, %ΔSEQPP, and %ΔAVGPP performed better than %ΔSEQRR for differentiating SA from APCs in degree of prematurity, while return cycle length was differentiated clearly with all 4 methods. Using %ΔSEQPP, the degree of prematurity and return cycle length of APCs differed from both SA and VPCs. Using %ΔAVGRR and %ΔAVGPP removed the influence of the degree of prematurity on the return cycle length variation. Variability of the sinus cycle lengths (PP7 to 10) following VPCs and SA was noted. Conclusions: The addition of PP intervals to cycle length variation analyses provided additional discriminatory capacity, particularly between APCs and SA. Classification bias was a limitation. Clinical Relevance: Evaluation of cycle length variation was a useful adjunct tool for resting ECG classification of SA, APCs, and VPCs.