Objective To evaluate the effectiveness of esketamine versus alfentanil as an adjunct to propofol target‑controlled infusion (TCI) for deep sedation during hepatocellular carcinoma (HCC) radiofrequency ablation (RFA) in elderly patients. Methods Eighty patients [65‒75 years of age, American Society of Anesthesiologists Physical Status (ASA) Ⅱ or Ⅲ scheduled for elective HCC RFA] were involved in the study. Patients were divided into either propofol and esketamine (group E， n=39) or propofol and alfentanil (group A，n=37) sedation according to the random number table method. Patients in both groups were sedated using a TCI of 1% propofol. When the effect concentration (Ce) had reached the target level of 1.5 mg/L, group E received esketamine 0.15 mg/kg while group A received alfentanil 2.0 μg/kg. Two minutes later after administering esketamine or alfentanil, the sedation level was assessed using the Modified Observer's Assessment of Alertness/Sedation (MOAA/S) scale. The target level was the MOAA/S score of 2. If MOAA/S was above 2, propofol TCI target (Cpt) was increased to 2.5 mg/L. If a sudden decrease of the MOAA/S score to less than 1 or airway obstruction was observed, TCI propofol would be stopped. Target Ce to the start of the procedure was 2.5 mg/L, at this point MOAA/S was assessed again. When MOAA/S was 1, Cpt or less was decreased in steps of 0.5 mg/L. If MOAA/S was still above 2 after adjusting Cpt to 2.5 mg/L, additional sedation was provided with increments 0.5 mg/L in plasma target level (Cpt) every 3 min. For every 0.5 mg/L step up of propofol TCI, esketamine 0.05 mg/kg or alfentanil 1.0 μg/kg was added. Maximum dose was 0.5 mg/kg esketamine or 7.5 μg/kg alfentanil. The total amount of propofol, alfentanil, esketamine as well as perioperative hypoxemia, recovery time (duration from transferring to PACU to MOAA/S>4) were recorded. The patient's satisfaction and surgeon's satisfaction were also recorded. Aldrete scores at 30 min (T1), 60 min (T2) and 90 min (T3) after waking up in postanesthesia care unit (PACU) and VAS scores at T1, T2, T3 and 1 day (T4) after discharge were recorded. Occurrence of respiratory adverse events, number of cases requiring airway intervention, occurrence of cardiovascular adverse events, number of cases requiring intervention with vasoactive drugs were recorded. Adverse events such as drowsiness, nausea and vomiting were also recorded. Results The total dosage of propofol in group E was lower than that in group A (P<0.05). The dose of esmketamine in group E was 0.25 (0.20, 0.32) mg/kg, and the dose of alfentanil in group A was 5.5 (4.5, 6.8) μg/kg。Anesthesia recovery time, patient satisfaction, physician satisfaction, VAS scores evaluated by surgeon, Aldrete scores at T1, T2 and T3 and VAS scores at T1, T2, T3 and T4, occurrence of respiratory adverse events, number of cases requiring airway intervention, occurrence of cardiovascular adverse events and number of cases requiring intervention with vasoactive drugs, drowsiness, nausea and vomiting was not significantly difference in two groups (P>0.05). Conclusions Low‑dose esketamine could reduce the total amount of propofol necessary for sedation during HCC RFA in elderly patients without affecting the recovery time, satisfaction of patients and surgeons, side effects and respiratory or cardiovascular adverse events, when compared with alfentanil.