The analysis of heart rate variability (HRV) is essential for assessing autonomic activity in various clinical conditions. However, standard HRV analyses do not effectively capture sympathetic activations during sleep monitoring, particularly those expected in obstructive sleep apnea (OSA). We recently introduced a HRV method based on Poincaré plot analysis, providing time-resolved estimations of the main HRV rhythms, on top of the changes in the baseline interbeat interval duration, using a robust covariance matrix estimation approach based on Ledoit-Wolf approach. Our method yields the Cardiac Sympathetic Index (CSI), which surpasses the capacity of lowfrequency HRV for tracking sympathetic activity in standard autonomic elicitation conditions. Here we investigated whether CSI can effectively capture physiological changes in response to OSA events. Our findings demonstrate that our CSI estimator outperforms standard metrics of sympathetic activity, such as low-frequency HRV and mean interbeat interval duration, offering a valuable and cost-effective approach for studying autonomic activity through HRV analysis.