Introduction: Approximate entropy (ApEn) is a widely imposed metric to evaluate a chaotic response and irregularities of RR-intervals from an electrocardiogram. Yet, the technique is problematic due to the accurate choice of the tolerance (r) and embedding dimension (M). We prescribed the metric to evaluate these responses in subjects exhibiting symptoms of chronic obstructive pulmonary disease (COPD) and we strived to overcome this disadvantage by applying different groupings to detect the optimal. Methods: We examined 38 subjects split equally: COPD and control. To evaluate autonomic modulation the heart rate was measured beat-by-beat for 30 min in a supine position without any physical, sensory, or pharmacological stimuli. In the time-series obtained the ApEn was then applied with set values for tolerance, r and embedding dimension, M. Then, the differences between the two groups and their effect size by two measures (Cohen’s ds and Hedges’s gs) were computed. Results: The highest value of statistical significance accomplished for any effect size statistical combinations undertaken was -1.13 for Cohen’s ds, and -1.10 for Hedges’s gs with embedding dimension, M = 2 and tolerance, r = 0.1. Conclusion: ApEn was capable of optimally identifying the decrease in chaotic response in COPD. The optimal combination of r and M for this were 0.1 and 2, respectively. Despite this, ApEn is a relatively unpredictable mathematical marker and the use of other techniques to evaluate a healthy or pathological condition is encouraged.