Objective <p>Diagnosing prolonged disorders of consciousness (DoC) remains a significant challenge in neurorehabilitation. Heart rate variability (HRV), a noninvasive measure of cardiac autonomic regulation, has been increasingly recognized as relevant to the mechanisms underlying consciousness. However, long-term HRV data in DoC patients remain scarce. This study aimed to investigate the relationship between long-term HRV metrics and the severity of consciousness impairment in patients with prolonged DoC.</p> Methods <p>In this cross-sectional study, 49 patients with prolonged DoC were enrolled. Each participant underwent five assessments using the Coma Recovery Scale-Revised (CRS-R) and one 24-hour electrocardiographic (ECG) monitoring session within one week of enrollment. Based on their highest CRS-R subscale scores, patients were categorized into three groups: unresponsive wakefulness syndrome (UWS, <i>n</i> = 17), minimally conscious state (MCS, <i>n</i> = 19), and emergence from MCS (EMCS, <i>n</i> = 13). HRV time-domain and frequency-domain indices were extracted from the 24-hour ECG recordings. Differences in HRV metrics among the consciousness groups were analyzed, and correlations between HRV parameters and CRS-R scores were examined.</p> Results <p>Significant differences in HRV metrics were observed among the three consciousness groups (<i>P</i> &lt; 0.05), including both time-domain indices (SDNN, SDANN, SDNN Index) and frequency-domain indices (HF, LF, VLF, ULF, and Total power). Furthermore, these HRV indices were significantly correlated with CRS-R scores (<i>P</i> &lt; 0.05), indicating a strong association between autonomic function and the level of consciousness.</p> Conclusion <p>Long-term HRV monitoring reveals significantly greater heart rate variability in EMCS patients compared to those in UWS and MCS, highlighting its potential utility as a supplementary biomarker for assessing consciousness in patients with prolonged DoC.</p>