Asleep-awake-asleep craniotomy: A comparison with general anesthesia for resection of supratentorial tumors

Journal of Clinical Neuroscience 20 (2013) 1068–1073

Contents lists available at SciVerse ScienceDirect

Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn

Clinical Study

Asleep-awake-asleep craniotomy: A comparison with general anesthesia for resection of supratentorial tumors Shobana Rajan a, Juan P. Cata b, Eman Nada a, Robert Weil c, Rakhi Pal a, Rafi Avitsian d,⇑ a

Department of Anesthesiology, Cleveland Clinic, Cleveland, OH, USA Department of Anesthesiology, M.D. Anderson Medical Center, Houston, TX, USA c Department of Neurosurgery, Cleveland Clinic, Cleveland, OH, USA d Department of General Anesthesia, Cleveland Clinic, 9500 Euclid Avenue, E-31, Cleveland, OH 44195, USA b

a r t i c l e

i n f o

Article history: Received 25 June 2012 Accepted 15 September 2012

Keywords: Anesthetic methods Intra-arterial therapy Ischemic stroke

a b s t r a c t The anesthetic plan for patients undergoing awake craniotomy, when compared to craniotomy under general anesthesia, is different, in that it requires changes in states of consciousness during the procedure. This retrospective review compares patients undergoing an asleep-awake-asleep technique for craniotomy (group AW: n = 101) to patients undergoing craniotomy under general anesthesia (group AS: n = 77). Episodes of desaturation (AW = 31% versus AS = 1%, p < 0.0001), although temporary, and hypercarbia (AW = 43.75 mmHg versus AS = 32.75 mmHg, p < 0.001) were more common in the AW group. The mean arterial pressure during application of head clamp pins and emergence was significantly lower in AW patients compared to AS patients (pinning 91.47 mmHg versus 102.9 mmHg, p < 0.05 and emergence 84.85 mmHg versus 105 mmHg, p < 0.05). Patients in the AW group required less vasopressors intraoperatively (AW = 43% versus AS = 69%, p < 0.01). Intraoperative fluids were comparable between the two groups. The post anesthesia care unit (PACU) administered significantly fewer intravenous opioids in the AW group. The length of stay in the PACU and hospital was comparable in both groups. Thus, asleep-awake-asleep craniotomies with propofol-dexmedetomidine infusion had less hemodynamic response to pinning and emergence, and less overall narcotic use compared to general anesthesia. Despite a higher incidence of temporary episodes of desaturation and hypoventilation, no adverse clinical consequences were seen. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction Asleep-awake-asleep is an anesthesia technique for awake craniotomies (AWC) where management is targeted to allow resection or mapping with maximal functional protection of brain areas that control motor function and speech.1 Originally used for epilepsy surgery,2 the indication for AWC has expanded to brain tumor resection when eloquent areas are at risk, requiring intraoperative monitoring, or when general anesthesia (GA) should be avoided (for example, insertion of deep brain stimulators).3 The hallmark of AWC is to provide adequate analgesia and sedation in patients who are required to be conscious and cooperative during the monitoring phase. The procedure is made tolerable for the patient, and still allows neurological monitoring. During the first stage the patient is either under GA or deep sedation while the craniotomy is performed. Subsequently the level of sedation is reduced to achieve a consciousness level that allows motor and/or speech mapping.4 ⇑ Corresponding author. Tel.: +1 216 444 9735; fax: +1 216 444 2294. E-mail address: [email protected] (R. Avitsian). 0967-5868/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jocn.2012.09.031

The anesthesiologist’s challenge is to select a technique that provides sedation, anxiolysis and optimal analgesia during brain exposure as well as immobility and comfort during mapping and resection of the tumor while minimizing hypoxemia, hypercarbia, nausea, vomiting, seizures and hemodynamic instability. Several sedation, analgesia, and anesthetic techniques have been described in the medical literature for AWC.5,6 A retrospective review of 332 patients7 undergoing AWC for epilepsy surgery using a propofolbased sedation technique without a definitive airway highlighted intraoperative problems. A prospective study of 25 patients used propofol and remifentanil infusion and a laryngeal mask airway (LMA) during the asleep phase. The LMA was removed after sedation discontinuation during the awake phase.8 Some anesthesiologists use endotracheal intubation, nasotracheal tubes or Guedel oro-pharyngeal airways. Commonly used medications include propofol, fentanyl, remifentanil, midazolam and dexmedetomidine. In our study, we retrospectively reviewed two cohorts of patients who underwent surgery for a supratentorial brain tumor. We highlight issues faced during AWC and evaluate the safety of this procedure. A comparison with GA is also presented, as it is the standard technique in most craniotomies. We tested the

1069

S. Rajan et al. / Journal of Clinical Neuroscience 20 (2013) 1068–1073

hypothesis that AWC with dexmedetomidine/propofol infusion has acceptable perioperative outcomes compared to GA technique. 2. Materials and methods We obtained Cleveland Clinic Institutional Review Board approval for a retrospective medical records review of patients who had undergone a craniotomy with neurophysiological monitoring for supratentorial brain tumor resection between 2007 and 2010 with a waiver of informed consent. Patients were divided into two groups based on the type of anesthesia: asleep-awake-asleep technique group (AW) and asleep craniotomy group (AS) with endotracheal intubation under GA. The AW group underwent speech, motor and/or somatosensory evoked potential (SSEP) monitoring, while the AS group had only SSEP and/or motor evoked potential (MEP) monitoring. A total of 213 medical records were reviewed retrospectively, 116 from the AW group and 97 from the AS group. Hard copy, electronic chart and Cleveland Clinic electronic anesthesia recording system data were utilized. The American Society of Anesthesiologists standard intraoperative monitors (electrocardiogram, pulse oximeter, capnogram, and temperature probe) plus a radial arterial cannula for invasive blood pressure (BP) monitoring and two peripheral intravenous (IV) catheters were placed in all patients during induction. Patients in the AW group had spontaneous ventilation with supplemental oxygen (2 to 8 L/min) through a nasal airway and a nasal cannula with the ability to monitor end tidal CO2. Continuous propofol infusion (50– 250 mcg/kg/min) and dexmedetomidine (1 mcg/kg loading dose in 10–15 min followed by 0.4–0.7 mcg/kg/hr maintenance) was used in the AW group for the initial sleep phase. Local anesthetic (50:50 mixture of 1% lidocaine with 1:100,000 epinephrine; and 0.25% bupivacaine) infiltration was performed by a surgeon only in the AW group before application of the head frame at the pin and incision sites. During the ‘‘awake’’ phase, the propofol infusion was stopped while dexmedetomidine was reduced or stopped to achieve a patient responsive to verbal stimulation without signs of pain or cardio-respiratory derangements. Speech and/or motor monitoring were performed by a neurophysiologist. In patients who complained of pain, intermittent IV boluses of fentanyl (25– 50 lg) were administered. The AS group received a GA with IV induction with fentanyl (1– 2 lg/kg), propofol (1–2 mg/kg), and rocuronium (0.6–1.2 mg/kg). Positive pressure ventilation and oxygenation was maintained with endotracheal intubation to achieve an arterial partial pressure of carbon-dioxide (PaCO2) of 30 to 35 mmHg. General anesthesia was maintained with a 0.5 to 1 minimum alveolar concentration of sevoflurane or isoflurane in air/oxygen mixture, remifentanil (0.05–0.2 lg/kg/min) infusion and intermittent boluses of rocuronium for muscle relaxation. All anesthetics were targeted to emerge and extubate patients at the procedure end. A forced-air warming device was used in both groups to maintain normothermia. An esophageal temperature probe for the AS group patients and axillary temperature probe for the AW group patients was placed. Hemodynamic derangements were typically treated according to the anesthesiologist’s decision. Collected data included perioperative systolic, diastolic and mean arterial pressure before induction, during head pinning, during surgical incision, during the case and emergence. The frequency and dose of vasopressors or antihypertensives given intraoperatively and in the post anesthesia care unit (PACU) were recorded. Episodes of hypoxemia (saturation level of oxygen lower than 90%) were also noted. Since the CO2 collection was through the nasal cannula and this quantitative measure is not reliable, arterial blood gases were drawn and the PaCO2 was recorded twice, during the first hour after incision and between 1 and 2 h into the procedure. Fluid requirements and estimated blood loss were also

recorded. Postoperative pain scores on the visual analogue scale (VAS) and postoperative opioid requirements of a fentanyl equivalent dose (0.1 mg of fentanyl equivalent to 10 mg of morphine)9 were recorded during the PACU stay. We also considered the length of the PACU stay based on the fulfillment of the modified Aldretes10 criteria and length of hospital stay. The incidence of intraoperative and PACU complications were noted. 2.1. Statistical analysis Results are shown as mean and standard deviation or median and quartiles when applicable. The chi-square test was used for 2  2 tables and Fisher’s exact test for 2  2 tables when one or more of the expected frequencies was less than five. Paired T-test or Mann-Whitney tests were used comparing preoperative and postoperative variables when applicable. An analysis of variance was used for multiple comparisons with Bonferroni corrections. A p-value < 0.05 was considered to be significant for all tests. Based on historical data of an anticipated effect size of 0.5 (Cohen’s d) and a desired statistical power level of 90%, we estimated that a minimum total sample of 140 patients (minimum sample size per group of 70 patients) would be necessary to detect a significant difference in pain scores after PACU arrival. Statistical analysis was performed with Statistica 6.1 (Statsoft Incorporated, Tulsa, OK, USA) software. 3. Results After reviewing medical records and excluding those with incomplete and missing data 101 AW and 77 AS patients were included for statistical analysis. 3.1. Intraoperative variables Baseline characteristics and demographics were comparable in both groups except for location and histology of the tumors (Table 1). Analysis of intraoperative hemodynamic variables shows that BP parameters and heart rate during pinning and emergence were significantly lower in the AW group than in the AS group (Table 2, p < 0.0001). More patients in the AS group (69%) than the AW group (53%) required intraoperative vasopressors with higher doses of ephedrine (Table 2, p < 0.003) but intraoperative dose of phenylephrine (Table 2, p < 0.066) was comparable. Use of antihypertensive medications was comparable between both groups (Table 2, p < 0.146). Respiratory parameter analysis (Table 3) demonstrates a significantly larger number of patients in the AW group with episodes of Table 1 Demographics and preoperative data comparing patients who underwent awake craniotomoy (AWAKE) and craniotomy under general anesthetic (ASLEEP). Variable

AWAKE (n = 101)

ASLEEP (n = 77)

P value

Age Gender, n (%) M/F Weight Height ASA status, n (%) I–II III–IV Tumor location, n (%) Frontal Parietal Temporal Occipital Histology, n (%) Malignant Benign

52 [40–65] 64 (63)/37 (37) 80 [71.5–92.15] 167.7 [162–178]

51 [37–63] 42 (55)/35 (45) 79 [65.5–99.9] 170 [160–180]

0.34 0.28 0.997 0.924 0.171

38 (38) 63 (62)

37 (48) 40 (52)

50 (49) 26 (26) 24 (24) 1 (1)

31 (40) 9 (12) 28 (36) 9 (12)

99 (98) 2 (2)

67 (87) 10 (13)

0.0009

0.005

ASA = American Society of Anesthesiologists

1070

S. Rajan et al. / Journal of Clinical Neuroscience 20 (2013) 1068–1073

Table 2 Intraoperative hemodynamic data comparing patients who underwent awake craniotomoy (AWAKE) and craniotomy under general anesthetic (ASLEEP). Variable

AWAKE (n = 101) Mean (SD)

ASLEEP (n = 77) Mean (SD)

Mean difference (95% CI of difference)

P value

Systolic BP Baseline Pinning Incision Emergence

130.7 132.5 125.5 122.2

(16.34) (23.31) (20.41) (18.36)

130.9 146.7 128.5 149.8

(14.5) (25.21) (20.62) (22.67)

Diastolic BP Baseline Pinning Incision Emergence

77.12 70.76 67.39 66.49

(11.06) (16.95) (14.05) (12.84)

75.42 79.83 66.32 78.89

(10.44) (15.13) (12.54) (13.26)

1.7 ( 4.64–8.05) 9.07 ( 15.43– 2.71) 1.06 ( 5.32–7.45) 12.4 ( 18.82– 5.99)

Mean BP Baseline Pinning Incision Emergence

90.99 91.47 86.31 84.85

(9.95) (18.86) (15.79) (13.52)

93.08 102.9 85.55 105.0

(10.73) (18.95) (12.54) (17.44)

2.08 ( 9.43–5.26) 11.45 ( 18.82– 4.08) 0.76 ( 6.63–8.16) 18.28 ( 35.04– 1.51)

Heart rate Baseline Pinning Incision Emergence

76.21 73.03 71.02 75.23

(14.37) (10.68) (10.25) (16.37)

74.69 79.53 69.17 92.00

(10.44) (16.91) (13.76) (17.64)

1.6 ( 8.07–4.68) 8.82 ( 15.21– 2.45) 1.47 ( 7.85–4.9) 20.11 ( 27.54– 12.69)

Intraop. fentanyl (mcg)

169.8 (80.32)

80.81 (80.32)

0.0001

Intraop. vasopressors Yes/No, n (%) Ephedrine (mg) Phenylephrine (mcg)

43(53)/58(57) 0 [0,0] 0 [0100]

53(69)/24(31) 5 [0,10] 0 [0200]

0.001 0.003 0.066

PACU vasopressors Yes/No, n (%)

0(0)/101(100)

0(0)/77(100)

1.00

Intraop. AHT Yes/No, n (%)

28(28)/73(72)

30(39)/47(61)

0.146

PACU AHT Yes/No, n (%)

11(11)/90(89)

20(26)/57(74)

0.01

<0.0001 0.19 ( 9.86–9.46) 14.12 ( 23.81– 4.43) 3.011 ( 12.74–6.71) 26.22 ( 35.14– 17.3)

  <0.0001   <0.0001  

 

SD = standard deviation, CI = confidence interval, BP = blood pressure, Intraop = intraoperative, PACU = post anesthesia care unit, AHT = antihypertensive medications

Table 3 Intraoperative respiratory and fluid data comparing patients who underwent awake craniotomoy (AWAKE) and craniotomy under general anesthetic (ASLEEP). Variable

AWAKE (n = 101) Mean (SD)

ASLEEP (n = 77) Mean (SD)

P value

SaO2 <90%, n (%) PaCO2 60 min (mmHg) PaCO2 120 min (mmHg)

26 (26) 43.75 [38,50] 42 [37,46.7]

1 (1) 32.75 [31,35.85] 32 [30,34]

0.00001 0.0001 0.0001

Fluids (mL) Crystalloids Colloids EBL PRBC, Y/N n (%)

2500 [2000,3400] 0 [0500] 200 [100,300] 1 (1)

2900 [2275,3750] 0 [0500] 200 [100,275] 5 (6)

0.009 0.089 0.63 0.083

Mannitol, n (%)

41 (40)

35 (45)

0.54

SD = standard deviation, EBL = estimated blood loss, PRBCs = packed red blood cells

desaturation (690% pulse oximetry) (p < 0.00001) which in all cases improved by placement of a nasopharyngeal airway, jaw thrust, increase in oxygen flow rate and/or reduction of the propofol infusion rate. One patient needed insertion of a LMA. No AS group patients showed intraoperative episodes of desaturation; however, two patients became hypoxemic immediately after emergence. One had to be reintubated while the other was treated with a simple jaw thrust maneuver. In the AW group the average PaCO2, recorded 1 and 2 h intraoperatively, was 43.75 and 42 respectively. In the AS group these recordings were determined by the ventilator rate set by the anesthesiologist and were 32.75 and 32 respectively. Statistical analysis of fluid administration demonstrates that crystalloids (p < 0.09), colloids (p < 0.089), packed red blood cells (p < 0.083) and mannitol (p < 0.54) were comparable in both groups.

3.2. Postoperative variables Analysis of postoperative hemodynamic variables in the PACU demonstrated the need for antihypertensives was significantly less in the AW group than the AS (Table 2, p < 0.01). Respiratory variables analysis demonstrated no drop in saturation in the AW group except for one patient who had a seizure that required treatment with IV midazolam, accompanied by hypoxemia which improved after an oro-pharyngeal airway insertion. Pain scores were significantly lower in the AW group at 15 and 60 min in the PACU (Table 4). Need for postoperative opioids was also lower in this group (Table 4, p < 0.0001) despite lower intraoperative fentanyl use compared to the AS group, and the analysis did not include

S. Rajan et al. / Journal of Clinical Neuroscience 20 (2013) 1068–1073

1071

Table 4 Postoperative data comparing patients who underwent awake craniotomoy (AWAKE) and craniotomy under general anesthetic (ASLEEP). Variable

AWAKE (n = 101) Mean (SD)

ASLEEP (n = 77) Mean (SD)

Fentanyl (mcg) Pain VAS (0/10) 15 min 30 min 60 min 120 min 240 min

50 [25,100]

150 [100,200]

2.71 3.16 3.09 2.95 3.28

4.32 (3.5) 4.47 (3.26) 4.44 (3.25) 4.01 (3.12) 3.7 (2.87)

Opioids, Y/N n (%) POV, Y/N n (%) LOS PACU (min) DVT Y/N n (%) LOS Hospital (days)

66/35 (65/35) 5/96 (5/95) 240 [154,300] 3/98 (3/97) 3 (3,4)

(3.2) (3.19) (3.06) (2.87) (2.9)

67/10 (87/13) 15/62 (19/81) 240 [157.5360] 3/74 (4/96) 3 (3,4)

Mean difference (95% CI of difference)

P value

1.7 (0.30–3.09) 1.37 ( 0.02–2.76) 1.43 (0.03–2.83) 1.02 ( 0.36–2.42) 0.31 ( 1.08–1.7)

0.0001 0.0001  ns  ns ns 0.0001 0.003 0.319 1.00 0.631

SD = standard deviation, CI = confidence interval, VAS = verbal analogue scale, POV = postoperative vomiting, LOS = length of stay, PACU = post anesthesia care unit, DVT = deep vein thrombosis

remifentanil used in the AS group (169.8 lg versus 80.81 lg), p < 0.0001, Table 2). Post-operative nausea and vomiting requiring IV ondansetron was only 4.1% in the AW group versus 21% in the AS group (p < 0.003). The PACU length of stay (LOS) was based on Aldrete’s score for discharge and there was no difference between the two groups (240 min, p < 0.319, Table 4). 3.3. Complications In the AW group, four patients developed complications including delayed arousal (n = 1), seizures (n = 2), and airway obstruction requiring laryngeal mask airway insertion (n = 1). In the AS group the number of complications were comparable and included seizures (n = 1) and post extubation hypoxia (n = 2), including one reintubation. The PACU and hospital LOS, and incidence of deep vein thrombosis (Table 4) were comparable in both groups.

4. Discussion Before designing the study, the authors had questioned the validity of a comparison between the so-called AWC and craniotomy under GA. The indications necessitating an AWC are specific and one can argue that differences in outcome may be the result of factors necessitating either form of the procedure. Histology may be different since many craniotomies for meningioma excision rarely require AWC although they may require SSEP or MEP monitoring. However, the investigators agreed that the current differences in studied outcomes could not be explained by a mere difference in location or histology of the tumor. The aim of this study was not to compare GA with monitored anesthesia care for AWC. Rather, we aimed to highlight issues faced during an AWC in evaluating the safety and efficacy of using a combination of propofol and dexmedetomidine infusions for the asleep-awake-asleep technique in supratentorial brain tumors, in comparison to GA which is the current accepted standard. Hemodynamic instability has been reported to be more common in AWC than in craniotomy under GA.6,7 In our study, hemodynamic changes that occurred with pinning and incision were significantly less in the AW group. One potential explanation is that local anesthetic block was applied in all AWC patients but not universally in the AS patients. Also the effect of dexmedetomidine used in the AW group could have prevented the usual BP surge seen during pinning and incision with sympathetic response. A study by Uyar et al.11 suggested a single bolus dose of dexmedetomidine before induction of anesthesia attenuated the hemodynamic and neuroendocrine responses to skull-pin insertion, thus

it is possible that the dexmedetomidine use as an adjunct can help lower the stress response to skull pin fixation. The higher dose of vasopressors needed in the AS group may be explained by a deeper level of anesthesia with greater decrease in the peripheral vascular resistance from the volatile anesthetics, plus use of positive pressure ventilation causing a preload drop.12,13 This may also be a reason for more crystalloids administered in AS group, although colloid administration was similar in the two groups. The difference in vasopressor use was only for ephedrine, not for total intraoperative phenylephrine. Since choice of the vasopressor is usually dependent on heart rate, a possible explanation may be a more predominant bradycardic effect of remifentanil14 during lower noxious stimulation, prompting the use of ephedrine. Since AW patients did not have controlled ventilation, oxygen and carbon dioxide partial pressure were dependent on their physiologic and respiratory drive. This is in contrast to GA where the end tidal CO2 is a result of the ventilator setting by the anesthesiologist. In a spontaneously ventilating patient, IV anesthetics and opioids decrease the ventilatory drive,15 however dexmedetomidine simulates normal sleep and does not hamper the ventilatory drive.16 Using a combination of propofol and dexmedetomidine rather than just propofol alone has the advantage of using lower propofol doses, thus lessening chance of respiratory depression. All desaturation episodes in the AW group happened at the beginning of surgery when relatively deeper anesthesia was needed for incision. All patients had a nasal airway but an additional chin lift maneuver was done if saturation decreased. Other maneuvers performed were increasing the oxygen flow and decreasing the propofol infusion rate. Rughani et al.17 described 25 patients who underwent AWC with a propofol infusion, of whom five (20%) developed respiratory problems. Four patients (20% of the total patients) were treated by cessation of propofol and one (4%) required endotracheal intubation due to coughing and inability to tolerate the surgery. In our experience, only one patient (1%) in the AW group needed placement of an LMA during the sleep phase. There are other reports of airway management with a variety of methods including non-invasive positive pressure ventilation18 with I-gel airway (Intersurgical Incorporated, Liverpool, NY, USA)19 and LMA Supreme (Teleflex Medical, Research Triangle Park, NC, USA).20 Manninen et al. described an 18% incidence of drop in saturation in AWC patients.21 Sarang et al.21 reported airway obstruction in three out of seven of their sedated and spontaneously breathing AWC patients. They reported no episodes of desaturation using a LMA with sedation using propofol and fentanyl or remifentanil. In our AS patients, there was no desaturation intraoperatively as expected due to controlled ventilation, but hypoxemia became

1072

S. Rajan et al. / Journal of Clinical Neuroscience 20 (2013) 1068–1073

apparent after extubation leading to re-intubation in one patient (1.5%). Due to the depressant effect of the sedatives and analgesics, the AW patients tended to have slightly higher PaCO2 values as they were breathing spontaneously under sedation compared to GA where the ventilator setting determined the PaCO2. However, these PaCO2 levels did not seem to have any adverse impact on intracranial swelling since there was no significant difference in the administration of mannitol between the two groups, nor did any patient have to be converted to GA due to brain swelling. Seizure is a possible complication in any craniotomy. A focal or generalized seizure can occur intraoperatively during the stimulation of the brain surface. In clinical studies the incidence of seizure has been quoted as 3–18%.7,21 In our series, two AW and one AS patients had a generalized seizure. Seizure in the operating room is managed with immediate cessation of the cortical stimulation and gentle irrigation of the brain surface with a small amount of cold saline,22 as well as low dose propofol or midazolam.7 In the postoperative period supportive measures to maintain a patent airway, antiseizure medications, maintenance of cardiovascular stability and immediate CT scan to eliminate an acute intracranial process are typical treatments. Craniotomy patients frequently complain of nausea and vomiting in the peri-operative period. The incidence of postoperative nausea and vomiting has been reported as 8% in the Keifer et al.23 study. Manninen et al.1 reported the incidence of postoperative nausea and vomiting in AWC as less than GA. We found that the incidence of nausea and/or vomiting was significantly less in the AW group. The employment of propofol infusion and lower opioid usage likely account for this lower incidence.24 Current studies suggest that acute as well as chronic pain is common in craniotomy patients.25,26 A multimodal approach to acute post-craniotomy pain is recommended.27 Insufficient pain control can increase the intracranial pressure in patients with compromised cerebral autoregulation.28 Most neurosurgeons prefer to obtain neurologic evaluation as early as possible, hence it becomes essential to balance management of pain and early emergence with an awake patient responsive to neurologic evaluation. Remifentanil, an opioid regarded as the drug of choice in many centers, has an ultra-short half-life which facilitates rapid emergence and postoperative neurological examination, but falls short of being the ideal agent since its analgesic effect diminishes rapidly due to the short half-life.29 According to a study by Nair et al.,30 63% of patients complained of inadequate analgesia after craniotomy in the first 12 h postoperatively, with 12% experiencing severe pain. In our study, postoperative opioid requirements and pain intensity was significantly different between the two groups. Opioid requirement was significantly less in the AW group despite lower intraoperative fentanyl doses. This agrees with previous studies.30–32 The criterion for opioid (usually fentanyl) administration is a VAS pain score of greater than or equal to four. We postulate that pain control is better in patients who underwent AWC due to the addition of dexmedetomidine and/or a better local anesthetic blockade. Improved control of pain postoperatively likely correlates with the decreased PACU use of antihypertensives in the AW patients. Our study has limitations. The study is retrospective, and we included patients with both malignant and benign tumors in different supratentorial locations. Also, the two groups are fundamentally different with respect to the anesthesia type which leads to different sets of expectations and risk-benefit tradeoffs, yet the aim here was to highlight the issues faced in one group (AW) in comparison to general anesthesia (AS). Finally, we excluded patients with missing pain data (15 from the AW group and 20 from the AS group).

5. Conclusion This study shows that asleep-awake-asleep craniotomy using a combination of dexmedetomidine and propofol with intermittent boluses of fentanyl provides adequate sedation, analgesia and a smooth wake-up during the period of neurological monitoring with stable hemodynamics and acceptable respiratory parameters.

Conflicts of interest/disclosures The authors declare that they have no financial or other conflicts of interest in relation to this research and its publication. References 1. Manninen PH, Balki M, Lukitto K, et al. Patient satisfaction with awake craniotomy for tumor surgery: a comparison of remifentanil and fentanyl in conjunction with propofol. Anesth Analg 2006;102:237–42. 2. Sahjpaul RL. Awake craniotomy: controversies, indications and techniques in the surgical treatment of temporal lobe epilepsy. Can J Neurol Sci 2000;27:S55–63 discussion S92–6. 3. Deiner S, Hagen J. Parkinson’s disease and deep brain stimulator placement. Anesthesiol Clin 2009;27:391–415. 4. Sacko O, Lauwers-Cances V, Brauge D, et al. Awake craniotomy vs. surgery under general anesthesia for resection of supratentorial lesions. Neurosurgery 2011;68(5):1192–8. 5. Andersen JH, Olsen KS. Anaesthesia for awake craniotomy is safe and welltolerated. Dan Med Bull 2010;57:A4194. 6. Berkenstadt H, Perel A, Hadani M, et al. Monitored anesthesia care using remifentanil and propofol for awake craniotomy. J Neurosurg Anesthesiol 2001;13:246–9. 7. Skucas AP, Artru AA. Anesthetic complications of awake craniotomies for epilepsy surgery. Anesth Analg 2006;102:882–7. 8. Olsen KS. The asleep-awake technique using propofol-remifentanil anaesthesia for awake craniotomy for cerebral tumours. Eur J Anaesthesiol 2008;25:662–9. 9. Foley K. The treatment of cancer pain. N Eng J Med 1985;313:84–95. 10. Aldrete J. The post anaesthesia recovery score revisited. J Clin Anesth 1995;7:89–91. 11. Uyar AS, Yagmurdur H, Fidan Y, et al. Dexmedetomidine attenuates the hemodynamic and neuroendocrinal responses to skull-pin head-holder application during craniotomy. J Neurosurg Anesthesiol 2008;20:174–9. 12. Lowe D, Hettrick DA, Pagel PS, et al. Influence of volatile anesthetics on left ventricular afterload in vivo. Differences between desflurane and sevoflurane. Anesthesiology 1996;85:112–20. 13. Hettrick DA, Pagel PS, Warltier DC. Differential effects of isoflurane and halothane on aortic input impedance quantified using a three-element Windkessel model. Anesthesiology 1995;83:361–73. 14. Sato K, Kamii H, Kato M. Clinical use of remifentanil in neuroanesthesia. Masui 2008;57:704–7. 15. Gottschalk A, Yaster M. The perioperative management of pain from intracranial surgery. Neurocrit Care 2009;10:387–402. 16. Mantz J, Josserand J, Hamada S. Dexmedetomidine: new insights. Eur J Anaesthesiol 2011;28:3–6. 17. Rughani AI, Rintel T, Desai R, et al. Development of a safe and pragmatic awake craniotomy program at Maine Medical Center. J Neurosurg Anesthesiol 2011;23: 18–24. 18. Yamamoto F, Kato R, Sato J, et al. Anaesthesia for awake craniotomy with noninvasive positive pressure ventilation. Br J Anaesth 2003;90:382–5. 19. Tsuruta S, Yamada M, Shimizu T, et al. Airway management using i-gel in two patients for awake craniotomy. Masui 2010;59:1411–4. 20. Murata H, Nagaishi C, Tsuda A, et al. Laryngeal mask airway Supreme for asleep-awake-asleep craniotomy. Br J Anaesth. 2010;104:389–90. 21. Sarang A, Dinsmore J. Anaesthesia for awake craniotomy–evolution of a technique that facilitates awake neurological testing. Br J Anaesth 2003;90:161–5. 22. Sartorius CJ, Berger MS. Rapid termination of intraoperative stimulationevoked seizures with application of cold Ringer’s lactate to the cortex. Technical note. J Neurosurg 1998;88:349–51. 23. Keifer JC, Dentchev D, Little K, et al. A retrospective analysis of a remifentanil/ propofol general anesthetic for craniotomy before awake functional brain mapping. Anesth Analg 2005;101:502–8. 24. Mishra L, Pradhan S, Pradhan C. Comparison of propofol based anaesthesia to conventional inhalational general anaesthesia for spine surgery. J Anaesthesiol Clin Pharmacol 2011;27:59–61. 25. Batoz H, Verdonck O, Pellerin C, et al. The analgesic properties of scalp infiltrations with ropivacaine after intracranial tumoral resection. Anesth Analg 2009;109:240–4. 26. Gottschalk A, Berkow LC, Stevens RD, et al. Prospective evaluation of pain and analgesic use following major elective intracranial surgery. J Neurosurg 2007; 106:210–6.

S. Rajan et al. / Journal of Clinical Neuroscience 20 (2013) 1068–1073 27. Flexman AM, Ng JL, Gelb AW. Acute and chronic pain following craniotomy. Curr Opin Anaesthesiol 2010;23:551–7. 28. Quiney N, Cooper R, Stoneham M, et al. Pain after craniotomy. A time for reappraisal? Br J Neurosurg 1996;10:295–9. 29. Egan TD. Pharmacokinetics and pharmacodynamics of remifentanil: an update in the year 2000. Curr Opin Anaesthesiol 2000;13:449–55. 30. Nair S, Rajshekhar V. Evaluation of pain following supratentorial craniotomy. Br J Neurosurg 2011;25:100–3.

1073

31. Turgut N, Turkmen A, Ali A, et al. Remifentanil-propofol vs dexmedetomidinepropofol–anesthesia for supratentorial craniotomy. Middle East J Anesthesiol 2009;20:63–70. 32. Wu ZL, Zhou ZF, Xu LX, et al. Effect of dexmedetomidine on patient-controlled intravenous analgesia with fentanyl in elderly patients after total hip replacement. Nan Fang Yi Ke Da Xue Xue 2011;31:701–4.