The institutional review board of Liaocheng People's Hospital approved (No. 2016057) this retrospective trial, which was registered at chictr.org (ChiCTR-IPR-16008494). Patients who underwent burr-hole surgery for CSDH with MAC between January 2014 and December 2015 and provided written informed consent were enrolled in this study. The inclusion criteria were: age between 45 and 65 years and ASA grade I to III. Exclusion criteria included hypertension (diastolic blood pressure > 160 mmHg); bradycardia (<50 bpm); ischemic heart disease (<6 months); second- or third-degree heart block; long-term abuse of or addiction to alcohol, tobacco, opioids, and sedative–hypnotic drugs (>6 months); allergy to DEX and/or sufentanil; neuropsychiatric diseases; operation time shorter than 30 min; emergent surgery.

Patients were divided into the following three groups: Group D1 (n = 67, initial DEX infusion at 0.5 μg·kg⁻¹ for 10 min, then adjusted to 0.2–0.7 μg·kg⁻¹·h⁻¹); Group D2 (n = 75, initial DEX infusion at 1 μg·kg⁻¹ for 10 min, then adjusted to 0.2–0.7 μg·kg⁻¹·h⁻¹); and Group S (n = 73, initial sufentanil infusion at 0.3 μg·kg⁻¹ for 10 min, then adjusted to 0.1–0.3 μg·kg⁻¹·h⁻¹). Electronic charting and DoCare Clinic electronic anesthesia recording system data were utilized during this trial. All patients received an explanation about the operative procedure, Ramsay sedation score (RSS), and pain score prior to the surgery. The burr-hole surgery for CSDH was performed by a neurosurgeon with ≥5 years of residency experience.

No premedication was administered before surgery. Prior to starting the surgery, ASA standard monitoring five-lead electrocardiography, non-invasive arterial blood pressure, peripheral pulse-oximetry (SpO₂), respiratory rate (RR), and temperature were continuously monitored using an automated system (Philips IntelliVue MP50). Oxygen supplementation at 3 L·min⁻¹ was achieved through an oxygen mask; then, an 18-gauge intravenous catheter was placed in a peripheral vein under local infiltration anesthesia. A forced-air warming device (EQUATOR Convective Warmer, EQ-5000) was used during the surgery to maintain normothermia.

Patients in the D1 and D2 groups received an initial loading dose of 0.5 μg·kg⁻¹ and 1 μg·kg⁻¹ DEX, respectively, over 10 min followed by a maintenance infusion of 0.2–0.7 μg·kg⁻¹·h⁻¹ to achieve adequate sedation. Patients in group S received an initial loading dose of 0.3 μg·kg⁻¹ sufentanil over 10 min followed by a maintenance infusion of 0.1–0.3 μg·kg⁻¹·h⁻¹ to achieve adequate sedation. The target RSS of the three groups was 3 (patient exhibits response to commands). A rescue bolus of midazolam 0.02 mg·kg⁻¹ was repeated every 5 min to a maximum dose of 2.5 mg if RSS > 3 or movement was noted during the procedure, while fentanyl 1 μg·kg⁻¹ was repeated every 5 min to a maximum dose of 0.2 mg if the pain score (visual analog scale, VAS) >4. If the patient did not reach the ideal status after the maximum dose of midazolam and fentanyl was reached, propofol or GA was administered. Once patients showed the ideal state of sedation, the scalp was infiltrated with 5 mL of a local anesthetic solution containing 2.5 mL of 0.5% hydrochloride ropivacaine and 2.5 mL of 2% lidocaine with adrenaline at each burr hole site. DEX and sufentanil infusion were stopped when the drainage tube was fixed. All patients received 5 mg of tropisetron and were transferred to the post-anesthesia care unit (PACU) after surgery.

On arrival at the PACU, vital signs (HR, non-invasive blood pressure, RR, SpO₂, temperature), RSS, and VAS were monitored every 5 min for the first 20 min, then every 10 min for the rest of the time until the patients were discharged to the wards (Aldrete Score ≥ 9) (Mason et al., 2013). Surgeon satisfaction was assessed as follows 24 h after burr-hole surgery: 1, extremely dissatisfied; 2, not satisfied but able to manage; 3, satisfied; 4, extremely satisfied); patients were visited to assess satisfaction on a 7-point Likert verbal rating scale (Bagchi et al., 2014). Fentanyl, 1 μg·kg⁻¹, was given repeated every 5 min to a maximum dose of 0.2 mg if VAS > 4; however, in patients with a poor response to fentanyl or in the event of an obvious fentanyl-associated adverse effect, 30 mg of ketorolac was administered.

During the surgery, bradycardia and tachycardia were defined as HR <45 bpm decrease or >120 bpm increase from baseline and treated with intravenous atropine, 0.2 mg, or esmolol, 0.4 mg·kg⁻¹, respectively. Hypertension and hypotension were defined as a >20% increase or decrease from baseline and treated using urapidil (10–15 mg) or phenylephrine (20–80 μg), respectively. All patients were closely continuously monitored with the aid of five-lead electrocardiography and non-invasive arterial blood pressure, SpO2, and RR readings using an automated system (Philips IntelliVue MP50) for 48 h after surgery.

The intraoperative hemodynamic data (HR, non-invasive blood pressure, RR, SpO₂, temperature) were obtained from a Philips IntelliVue monitor at the following time points: arrival at the operating room (T1); after bolus administration of drug (T2); before administration of local anesthetic (T3); before skin incision (T4); at 5 min (T5) and 10 min (T6) after skin incision; and at arrival (T7), 5 min (T8), and 10 min (T9) at the PACU. Anesthesia onset time (from the initiation of anesthesia induction to the onset of the surgical procedure), recovery time (from stopping the DEX or sufentanil infusion to achieve RSS = 2), total number of intraoperative patient movements, amount of rescue midazolam or fentanyl, time to first dose of rescue midazolam or fentanyl, percentage of patients converted to alternative sedative or anesthetic therapy, adverse events, and patient and surgeon satisfaction scores were recorded.

The Kolmogorov–Smirnov test was used to assess the distribution of variables. Homogeneity of variance was determined using Levene's tests. Quantitative data were expressed as mean and standard deviation or median and inter-quartile range (IQR). Inter-group comparisons were performed using repeated-measures analysis of variance (ANOVA). The Bonferroni's correction was applied for post-hoc multiple comparisons. The non-parametric Kruskal–Wallis test was used for variables that were not normally distributed. Categorical data were expressed as frequency and percentage and analyzed using chi-squared tests or Fisher's exact tests when appropriate. Probability (P) values <0.05 were considered statistically significant. Statistical analysis was performed with SPSS for Windows Version 18.0 (SPSS Inc. Chicago, IL, USA).

Figure 1 shows the CONSORT diagram of patient recruitment. Initially, 1832 patients who underwent burr-Hole surgery for CSDH were screened between January 2014 and December 2015. Of these, 1617 patients were excluded for not meeting the inclusion criteria: 1023 patients required emergency surgery; 156 patients did not fall into the specified age range of 45–65 years; the ASA grade of 124 patients was more than III; 149 patients had cardiovascular and neuropsychiatric diseases; 23 patients had a long history of addiction to alcohol, opioids, and sedative–hypnotic drugs; the operation time of 15 patients was shorter than 30 min; and 127 patients were excluded after surgery due to incomplete clinical data. Finally, 215 patients were included in the primary analysis. Demographic and baseline clinical parameters were not significantly different among the three groups (P > 0.05, Table 1).

Baseline vital signs were not statistically different among the three groups (P > 0.05, Figure 2). Compared with S group, both D1 and D2 groups showed significantly decreased HR and MAP at T2, T3, T4, T7, T8, and T9 (P < 0.05, Figure 2). The lowest levels of HR and MAP among the three groups were recorded at T2.

Upon intergroup comparison, we found that anesthesia onset time was significantly shorter in group D2 (17.36 ± 4.23 vs. 13.42 ± 2.12 vs. 15.98 ± 4.58 min, respectively, for D1, D2, S; P < 0.001, Table 2), while durations of anesthesia and surgery and hematoma volume were not statistically different among the three groups (P > 0.05, Table 2). More patients in groups D1 and S required rescue midazolam to achieve RSS = 3 than in group D2 (74.63 vs. 42.67 vs. 71.23%, respectively, for D1, D2, S; P < 0.001, Figure 3A). However, the total dose of rescue midazolam was significantly higher in group D1 group than that in groups S and D2 (2.8 ± 0.3 vs. 1.9 ± 0.3 vs. 2.0 ± 0.4 mg, respectively, for D1, D2, S; P < 0.001, Figure 3A). The time to first dose of rescue midazolam was significantly longer in group D2 than in groups D1 and S (17.32 ± 4.47 vs. 23.56 ± 5.36 vs. 16.55 ± 4.91 min, respectively, for D1, D2, S; P < 0.001, Table 2). Significantly fewer patients in groups S and D2 required rescue fentanyl to relieve pain than in group D1 (62.69 vs. 21.33 vs. 27.40%, respectively, for D1, D2, S; P < 0.001, Figure 3B). Additionally, the total dose of rescue fentanyl in group D1 was significantly higher than that of groups S and D2 (212.5 ± 43.6 vs. 107.2 ± 35.9 vs. 98.6 ± 32.2 μg, respectively, for D1, D2, S; P < 0.001, Figure 3B). However, the time to first dose of rescue fentanyl was not statistically different among the three groups (18.47 ± 3.74 vs. 18.56 ± 2.92 vs. 18.24 ± 3.65 min, respectively, for D1, D2, S; P = 0.845, Table 2). Total cumulative dose of DEX was higher in group D2 (65.39 ± 18.53 vs. 92.15 ± 23.27 μg, respectively, for D1 and D2; P < 0.001, Table 2).

Total number of patient movements during the burr-hole surgery was higher in groups D1 and S (47.76 vs. 20.00 vs. 47.95%, respectively, for D1, D2, S; P = 0.000; Table 2). Although most of the patient movements could be controlled with reassurance, DEX or sufentanil infusion adjustment, and rescue drugs (midazolam or fentanyl), four patients from the D1 group and five from the S group still converted to an alternative sedative (propofol), while no patient required GA to complete the burr-hole surgery (Table 2).

The time to recovery for discharge from the PACU (i.e., time to Aldrete score ≥9) was significantly shorter in group D2 than in groups S and D1 (16.24 ± 4.15 vs. 12.48 ± 3.29 vs. 15.91 ± 3.66 min, respectively, for D1, D2, S; P = 0.000, Table 3). Results from the patient and surgeon satisfaction scores showed significant differences favoring the D2 group compared with D1 and S groups (P < 0.05, Table 3).

More patients in groups D1 and S than group D2 required higher doses of urapidil (41.79 vs. 20.00 vs. 34.25%, respectively, for D1, D2, S; P = 0.017, Table 4) and esmolol (32.84 vs. 16.00 vs. 31.51%, respectively, for D1, D2, S; P = 0.038, Table 4). The intraoperative dose of phenylephrine (11.94 vs. 12.00 vs. 13.70%, respectively, for D1, D2, S; P = 0.950, Table 4) and atropine (8.96 vs. 9.33 vs. 6.85%, respectively, for D1, D2, S; P = 0.824, Table 4) were comparable among the three groups during the surgery and in PACU.

The main adverse events are shown in Table 5. Patients in group D2 showed lower levels of the overall incidence of tachycardia (40.30 vs. 20.00 vs. 34.25% 25/73, respectively, D1, D2, S; P = 0.008, Table 5) and hypertension (44.78 vs. 22.67 vs. 36.99%, respectively, for D1, D2, S; P = 0.013, Table 5). The percentage of bradycardia, hypotension, nausea, and vomiting were comparable among the three groups during the surgery and in PACU (P > 0.05, Table 5). Six patients experienced respiratory depression (defined as a respiratory rate <8 times·min⁻¹ or an oxygen saturation <90%) in group S.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.