Purpose: To evaluate the recovery period of autonomic modulation, through geometric indices of heart rate variability (HRV), on coronary artery disease (CAD) patients submitted to a cardiovascular rehabilitation session (CR), associated with hydration. Methods: Thirty male participants of a CR program, diagnosed with CAD were submitted to the control (CP) and hydration protocol (HP) characterized by a CR session. Only during HP were the participants given 8 equal portions of water. The water amount was determined through the hydric loss measured at the CP. During the protocols, the heart rate was measured beat-by-beat at rest (5-10 minutes[M1]) and at recovery (0-5 minutes [M2], 5-10 minutes [M3], 15-20 minutes [M4], 25-30 minutes [M5], 40-45 minutes [M6], 55-60 minutes [M7]) for the HRV analysis, performed by the geometric indices: TINN, RRTRI, SD1, SD2 and SD1/SD2 ratio. Results: Statistically significant differences were observed between the protocols (SD1, p(value) = 0.022), moments (TINN, p(value) = 0.001; SD1, p(value) = 0.019; SD2, p(value) = 0.001; SD1/SD2, p(value) = 0.001) and moments vs. protocol interaction (SD1, p(value) = 0.019). The SD1 index pointed to acceleration of parasympathetic recovery in the first minutes after exercising (HP recovery after M3 [86.07 +/- 32.31%] vs. CP recovery after M5[86.43 +/- 24.56]) and increase in global variability (TINN-HP remained increased in longer, until M5 (M1 83.10 +/- 55.76 ms to M5 116.82 +/- 67.54 ms) vs. CP that remained increased for a short time, until M2 (M1 77.93 +/- 68.56 ms to M2 134.82 +/- 56.08 ms). Conclusions: In CAD patients, hydration promoted a more efficient recovery on parasympathetic autonomic modulation and increased the global HRV in the recovery period.