Objective <p>The incidence of osteonecrosis of the femoral head (ONFH) in non-elderly patients is increasing. Psychological distress, including anxiety and depression related to both the disease itself and total hip arthroplasty (THA), may adversely affect postoperative outcomes. This study aimed to evaluate the effects of stellate ganglion block (SGB) on perioperative psychological symptoms and postoperative recovery in patients undergoing THA for ONFH.</p> Methods <p>A total of 94 patients undergoing THA were enrolled in this study and randomly divided into the stellate ganglion block group (SGB group) and the control group (CON group). Patients in the SGB group received ultrasound-guided unilateral stellate ganglion block before surgery with an injection of 4 mL of 0.8% lidocaine; patients in the CON group underwent the same procedure, with an equal volume of normal saline replacing lidocaine. On postoperative days 1 and 4, the degree of anxiety and depression, pain intensity, sleep quality, recovery status, and occurrence of complications were evaluated in all patients.</p> Results <p>Compared with the CON group, patients receiving SGB treatment showed favorable trends in GAD-7 scores and improved postoperative recovery quality, as reflected by the primary mixed-effects analysis (<i>P</i> &lt; 0.05) and significantly higher QoR-15 scores. However, additional ANCOVA adjusting for baseline GAD-7 scores did not confirm a significant between-group difference in GAD-7 outcomes. The observed GAD-7 effect was therefore not robust to baseline adjustment. Although HADS scores tended to be lower in the SGB group, the difference did not reach statistical significance. SGB did not significantly affect pain intensity (VAS score) or the incidence of adverse events, including hypotension, dizziness, nausea and vomiting, and nightmares (all <i>P</i> &gt; 0.05).</p> Conclusion <p>Among non-elderly patients undergoing THA, SGB was associated with some favorable trends in perioperative psychological symptoms and postoperative recovery quality. However, the evidence was inconsistent across different psychological assessment scales and sensitive to baseline adjustment. Therefore, these findings should be considered hypothesis-generating rather than conclusive.</p> Trial registration <p>This study was registered in the Chinese Clinical Trial Registry. (ChiCTR2400089542) on 2024/09/10.</p>