This study was approved by the Ethics Committee of The General Hospital of Ningxia Medical University (no. KYLL-2022-0056) and registered at http://www.chictr.org.cn/index.aspx (no. ChiCTR2200057226). After obtaining written informed consent, female patients aged 30–75 years with American Society of Anesthesiologists (ASA) physical status I/II who were scheduled for MRM were enrolled. Patients with a history of severe heart, lung, liver, kidney, or endocrine diseases, history of allergy to the drug being studied, cognitive impairment, use of antipsychotics, or use of non-steroidal anti-inflammatory drugs (NSAIDs) were excluded.

Patients were randomly assigned to one of four treatment groups (control, low-dose, medium-dose, or high-dose S-ketamine) following a 1:1:1:1 ratio using a computer-generated random number table. The group allocation was concealed in sealed and opaque envelopes, and the corresponding envelopes were only opened when the patient was allocated to the intervention. The study medications were prepared by an attending anesthetist so that the designated researcher was blinded to the intervention. Group L-Sk received S-ketamine 0.25 mg/kg for induction, followed by an infusion of 0.25 mg/kg/h. Group M-Sk received S-ketamine 0.5 mg/kg for induction, followed by an infusion of 0.5 mg/kg/h. Group H-Sk received S-ketamine 0.75 mg/kg for induction, followed by an infusion of 0.75 mg/kg/h. Group C did not receive S-ketamine during anesthesia. Lavender suggested that an induction dose of S-ketamine is 0.5–1 mg/kg, and the maintenance dose is 0.5–3 mg/kg (Lavender et al., 2020). The S-ketamine dose in this study was within a safe dose range.

All patients were intramuscularly injected with 1 mg of penehyclidine 30 min before surgery. Routine monitoring includes an electrocardiography, pulse oximetry, heart rate, non-invasive blood pressure, and bispectral index. The anesthesia induction and maintenance methods among the four groups were as follows: patients in group C were induced with midazolam (0.05 mg/kg), sufentanil (0.3 μg/kg), propofol (2.0 mg/kg), and rocuronium (0.8 mg/kg) and maintained with a target-controlled infusion of propofol (4–6 mg/kg/h) and remifentanil (0.2–0.6 ug/kg/min). Patients in three different doses of the S-ketamine group were induced with midazolam (0.05 mg/kg), S-ketamine (0.25, 0.5, or 0.75 mg/kg), propofol (2.0 mg/kg), and rocuronium (0.8 mg/kg) and maintained with a target-controlled infusion of propofol (4–6 mg/kg/h), remifentanil (0.2–0.6 ug/kg/min), and S-ketamine (0.25, 0.5, or 0.75 mg/kg/h).

The anesthetic administration rate was adjusted to the study protocol’s maintenance dose to maintain a bispectral index range of 40–60 and mean arterial pressure (MAP) within 20% of preoperative baseline values. Ephedrine or norepinephrine was administered if the patient had hypotension (defined as a decrease in arterial blood pressure of more than 20% of the basic value). Urapidil, alpha 1 antagonist, was administered if the patient had hypertension (defined as an increase in arterial blood pressure of more than 20% of the basic value). S-Ketamine infusion stopped 30 min before the end of surgery. Drugs possessing anti-inflammatory effects, such as dexamethasone, etomidate, lidocaine, and dexmedetomidine, were not administered during the perioperative period in all patients. Patients were transferred to the post-anesthesia care unit (PACU) for recovery.

The primary outcomes were cellular immune function and systemic inflammatory markers. A volume of 4 ml of peripheral blood from the elbow of all patients was collected before anesthesia (T0) and at the end of (T1) and 24 h after the surgery (T2). The percentage of T lymphocytes (CD3⁺, CD4⁺, and CD8⁺), B lymphocytes, and natural killer (NK) cells, the ratio of CD4⁺/CD8⁺, and the counts of white blood cells (WBC), neutrophils (NEUT), lymphocytes (LYM), and hypersensitive C-reactive protein (hs-CRP) were compared between the groups. The inflammation-based prognostic scores were calculated, such as neutrophil-to-lymphocyte ratio (NLR), lymphocyte-to-monocyte ratio (LMR), platelets-to-lymphocyte ratio (PLR), systemic inflammation response index (SIRI), and systemic immune-inflammation index (SII). The SIRI and SII were calculated using the following formulas: SIRI = neutrophil count × monocyte count/lymphocyte count and SII = platelet count × neutrophil count/lymphocyte count. Lymphocyte subset analysis and lymphocyte percentage were detected by flow cytometry (FCM) to reflect the cellular immune function.

The secondary outcome included postoperative pain intensity, cumulative opioid and propofol consumption, PACU recovery time, awakening time (it refers to the time from drug withdrawal to extubation), adverse events (e.g., the occurrence of shivering, nausea, vomiting, dizziness, agitation, hypertension, psychotomimetic side effects—hallucinations or nightmares—and other adverse reactions), the incidence of remedial analgesia, and patient satisfaction score. Postoperative pain intensity was evaluated using an 11-point visual analogue scale (VAS) score at the end of the surgery and 4, 6, 24, and 48 h after the surgery. A straight line was divided into nine equal parts, and the two ends represented 0 points of no pain and 10 points of severe pain. The patients used a line to mark the value according to their subjective feelings to reflect the pain degree (Dannenbaum et al., 2011). If the postoperative pain exceeded 3 points, an intramuscular injection of diclofenac sodium of 75 mg was carried out for remedial analgesia. If the patient had moderate-to-severe nausea or vomiting, a serotonin 5-HT3 antagonist (tropisetron) of 2 mg was given intravenously. Patient satisfaction was assessed 48 h postoperatively by patients using an 11-point Likert scale (range, 0–10), where a score of 0 means “entirely unsatisfied,” and a score of 10 means “fully satisfied” (Farooq et al., 2016).

Statistical analysis was performed using SPSS 27.0. The Kolmogorov–Smirnov test and the Q–Q plot were used to assess the normality of all data. Normally distributed variables were analyzed using one-way analysis of variance (ANOVA). A Bonferroni post hoc test was used to compare groups pairwise. Continuous variables not following a normal distribution were analyzed using the Kruskal–Wallis test with a Dunn post hoc test used to compare groups pairwise. The inflammation index and ratio of lymphocyte subsets were compared using ANOVA for repeated measures. Enumerated data were presented as a percentage or ratio, and the chi-square test was used for comparisons. p < 0.05 was considered significant.

Based on our preliminary data (n = 30, 15 cases in group C, 15 cases in group M-Sk), the percentage of CD3⁺ in group C at the end of surgery was (55.51 ± 5.67), that in group M-Sk was (60.25 ± 4.48), the sample size required to detect a significant difference in immune function was calculated with an online program (http://powerandsamplesize.com/Calculators/Compare-k-Means/1-Way-ANOVA-Pairwise), and the results showed that 25 subjects in each group are needed, ensuring that the power of the test is 80% with a type alpha error of 0.05 (the alpha error was corrected by six pairwise comparisons). Allowing for dropouts, we aimed to recruit 35 patients in each group, with a total of 140 patients.

A total of 168 patients were first assessed for eligibility in this study. However, 32 patients met the exclusion criteria. Thus, 136 patients were randomized and allocated to the study groups. Later, five participants in group C, three in group L-Sk, four in group M-Sk, and four in group H-Sk were excluded due to loss to follow-up and refusing postoperative assessment or postoperative blood draw. Consequently, 120 patients were analyzed, with 30 patients in each group (Figure 1).

The baseline characteristics and perioperative data are presented in Table 1. There were no significant differences with respect to patient demographic data, the duration of anesthesia and surgery, PACU recovery time, and awakening time among the four groups (p > 0.05). Because of the analgesic effects of S-ketamine, three different doses of S-ketamine (0.25, 0.5, and 0.75 mg/kg) effectively reduced the consumption of opioids compared with the C group (F = 62.45, p < 0.001). Further pairwise comparison indicated that the consumption of opioids was significantly decreased in the M-Sk and H-Sk groups compared to the C and L-Sk groups (Bonferroni correction, p < 0.001). However, a non-significant reduction was observed between the H-Sk and M-Sk groups. Additionally, according to the experimental design principles, the different doses of the S-ketamine group did not use sufentanil (p < 0.001). These results suggested that the use of S-ketamine can effectively reduce opioid consumption, especially in 0.5 or 0.75 mg/kg (Table 1).

The percentage and absolute counts of the lymphocyte subsets detected by FCM among the four groups are shown in Figure 2. Existing studies have verified that the low counts of CD3⁺, CD4⁺, and NK cells are associated with impaired antitumor immunity. When the ratio of CD4+/CD8+ decreases, the immune balance is damaged.

In this study, the preoperative cellular immune function was comparable among the four groups. However, at the end of the surgery, the percentage of CD3⁺ cells in the C group was significantly lower than that in the S-ketamine groups. Moreover, a pairwise comparison indicated that the percentage of CD3⁺ cells in group H-Sk was higher than the percentage in groups L-Sk and M-Sk (p < 0.05). Similarly, 24 h after the surgery, the percentage of CD3⁺ cells in groups M-Sk and H-Sk were higher than those in groups C and L-Sk (p < 0.05) (Figure 2A). The percentage of CD4⁺ cells at T1 in group C was lower than the percentage in groups M-Sk and H-Sk, and the percentages were higher in group H-Sk than that in group L-Sk (p < 0.05). However, at T2, the percentage of CD4⁺ in group C was lower than that in group H-Sk. Groups H-Sk and M-Sk had a higher percentage of CD4⁺ than group L-Sk (p < 0.05) (Figure 2B). There was a lower percentage of CD8⁺ cells at T1 in group H-Sk than in group M-Sk (p < 0.05) (Figure 2C). As for the lymphocyte subset absolute counts, the absolute counts of CD3⁺, CD4⁺, and CD8⁺ cells in the C group were significantly lower than those in the S-ketamine groups (p < 0.05). A pairwise comparison indicated that the absolute counts of CD3⁺ and CD4⁺ cells in group H-Sk were higher than those in group L-Sk (p < 0.05). The ratios of CD4⁺/CD8⁺ at T1 and T2 in group C were lower than those in groups H-Sk and M-Sk. Furthermore, the ratio in group H-Sk was higher than that in group L-Sk (p < 0.05) (Figure 2F). Additionally, there was no significant difference in the percentage and absolute counts of NK cells and B lymphocytes among the four groups (Figures 2D, E). These results indicated that S-ketamine is closely related to perioperative immune protection, and the protective effect in the M-Sk and H-Sk groups may be more obvious (Figure 2).

The systemic inflammatory responses among the four groups are depicted in Figure 3. Studies have indicated that inflammation plays a vital role in tumorigenesis and progression. Mounting evidence suggests that high levels of WBC, NEUT, hs-CRP, NLR, PLR, SII, and SIRI are positively associated with poor overall survival and disease-free survival and negatively associated with LMR in patients with cancer.

The results of the present study showed that the preoperative inflammatory responses were comparable among the four groups. As shown in Figures 3A, B, at T1 and T2, compared with the C group, the concentrations of WBC and NEUT were significantly decreased in the M-Sk and H-Sk groups (p < 0.05). Furthermore, the pairwise comparison indicated that the concentrations of WBC and NEUT in group H-Sk were higher than that in group L-Sk (F = 7.589, p = 0.0001 < 0.05). Compared with group C, the concentration of LYM at T1 and T2 in group M-Sk was significantly increased (p < 0.05) (Figure 3C). The concentration of hs-CRP and the ratio of NLR at T1 and T2 in groups L-Sk, M-Sk, and H-Sk were significantly decreased compared to group C. The ratio of NLR in group M-Sk was lower than that in group L-Sk. A lower concentration of hs-CRP was observed in group H-Sk compared to group L-Sk. Furthermore, the concentration of hs-CRP at T2 in group L-Sk was higher than that in group M-Sk (p < 0.05) (Figures 3D, E). The ratios of SIRI and SII both at T1 and T2 in groups L-Sk, M-Sk, and H-Sk were significantly decreased compared to group C, and the ratio of SII at T2 in group M-Sk was lower than that in group L-Sk (p < 0.05) (Figures 3H, I). There were no significant differences in terms of the ratios of LMR and PLR at the corresponding time point among the four groups (p > 0.05) (Figures 3F, G). The aforementioned results indicate that S-ketamine might have anti-inflammatory effects. Moreover, the experimental data showed that the anti-inflammatory effects in groups M-Sk and H-Sk may be more significant (Figure 3).

The perioperative adverse reactions and patient satisfaction are presented in Table 2. The incidence of postoperative nausea and vomiting (PONV) in group C was 10 cases (33.3%), which was higher than that in the groups L-Sk, M-Sk, and H-Sk (p < 0.05), and non-difference was observed among the S-ketamine groups. The incidence of shivering in groups M-Sk and H-Sk was 0 cases (0%), which was lower than that in group C (16.7%) (p < 0.05) (Table 2). There were no significant differences in terms of dizziness, agitation, hypertension, and other adverse reactions among the four groups, and no patient exhibited psychotomimetic side effects related to S-ketamine. S-ketamine has a slight excitatory effect on the circulatory system, often causing a slight increase in blood pressure and heart rate. Therefore, our study has shown that the rates of hypotension and using vasoactive drugs in groups L-Sk, M-Sk, and H-Sk were lower than those in group C, and the rates were lower in group H-Sk than those in group L-Sk (p < 0.001). Furthermore, compared with group C, the anesthesia satisfaction score in groups M-Sk and H-Sk increased significantly and was higher in group M-Sk than in group L-Sk (p < 0.001). Similarly, the rate of remedial analgesia in group C was eight cases (26.7%), which was higher than that in groups M-Sk (3.3%) and H-Sk (6.7%) (p < 0.05). These results suggest that S-ketamine can be used safely and effectively in cancer patients without increasing the incidence of complications (Table 2).

Postoperative VAS scores for participants are illustrated in Figure 4. Briefly, no statistically significant differences in pain scores were detected among the four groups at the end of and 48 h after the surgery. However, 4 and 24 h after the surgery, the VAS scores in the L-Sk, M-Sk, and H-Sk groups decreased significantly compared to the C group. There were lower VAS scores at 6 h in the M-Sk and H-Sk groups than in the C and L-Sk groups (p < 0.05). However, no significant difference was detected between the M-Sk and H-Sk groups. These results suggest that the S-ketamine doses of 0.5 or 0.75 mg/kg showed a better analgesic effect; it can reduce the intensity of postoperative pain, which means S-ketamine is a promising option for perioperative analgesia (Figure 4).