ProSeal Laryngeal Mask Airway Attenuates Systemic and Cerebral Hemodynamic Response During Awakening of Neurosurgical Patients: A Randomized Clinical Trial.

CLINICAL INVESTIGATION

ProSeal Laryngeal Mask Airway Attenuates Systemic and Cerebral Hemodynamic Response During Awakening of Neurosurgical Patients: A Randomized Clinical Trial Laura Perello´-Cerda`, MD,* Neus Fa`bregas, MD, PhD,* Ana M. Lo´pez, MD, PhD,* Jose´ Rios, MSc,w Javier Tercero, MD,* Enrique Carrero, MD, PhD,* Paola Hurtado, MD,* Adriana Hervıás, MD,* Isabel Gracia, MD,* Luis Caral, MD,z Nicola´s de Riva, MD,* and Ricard Valero, MD, PhD*

Background: Extubation and emergence from anesthesia may lead to systemic and cerebral hemodynamic changes that endanger neurosurgical patients. We aimed to compare systemic and cerebral hemodynamic variables and cough incidence in neurosurgery patients emerging from general anesthesia with the standard procedure (endotracheal tube [ETT] extubation) or after replacement of the ETT with a laryngeal mask airway (LMA). Materials and Methods: Forty-two patients undergoing supratentorial craniotomy under general anesthesia were included in a randomized open-label parallel trial. Patients were randomized (sealed envelopes labeled with software-generated randomized numbers) to awaken with the ETT in place or after its replacement with a ProSeal LMA. We recorded mean arterial pressure as the primary endpoint and heart rate, middle cerebral artery flow velocity, regional cerebral oxygen saturation, norepinephrine plasma concentrations, and coughing. Results: No differences were found between groups at baseline. All hemodynamic variables increased significantly from baseline in both groups during emergence. The ETT group had significantly higher mean arterial pressure (11.9 mm Hg; 95% confidence interval [CI], 2.1-21.8 mm Hg) (P = 0.017), heart rate (7.2 beats/min; 95% CI, 0.7-13.7 beats/min) (P = 0.03), and

Received for publication March 26, 2014; accepted July 16, 2014. From the Departments of *Anesthesiology; zNeurosurgery; and wBiostatistics and Data Management Platform, IDIBAPS, Hospital Clinic, de Barcelona, University of Barcelona, Barcelona, Spain. Presented to the Euroanaesthesia Meeting in Barcelona, June, 2013. A.M.L. has received honoraria from the Company LMA, Bioser and Ambu. R.V. and L.P.-C. have received some travel expenses from the company Bioser. No conflicts of interest related to this study have been declared by the remaining authors. This study was performed with grant support from the Hospital Clı´ nic de Barcelona through the Emili Letang end-of-residency fund (awarded to L.P.-C. for 2011). This study was awarded the Archie Brain Award from the Spanish Anesthesiology Society (Sociedad Espanõla de Anestesiologı´ a y Reanimacioń)—Bioser in June 2012. Reprints: Ricard Valero, MD, PhD, Servicio de Anestesiologı´ a, Reanimacioń y Terapeútica del Dolor, Hospital Clı´ nic de Barcelona, Villarroel 170, Barcelona 08036, Spain (e-mail: [email protected]). Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved.

194 | www.jnsa.com

rate-pressure product (1045.4; 95% CI, 440.8-1650) (P = 0.001). Antihypertensive medication was administered to more ETTgroup patients than LMA-group patients (9 [42.9%] vs. 3 [14.3%] patients, respectively; P = 0.04). The percent increase in regional cerebral oxygen saturation was greater in the ETT group by 26.1% (95% CI, 9.1%-43.2%) (P = 0.002), but no between-group differences were found in MCA flow velocity. Norepinephrine plasma concentrations rose in both groups between baseline and the end of emergence: LMA: from 87.5 ± 7.1 to 125.6 ± 17.3 pg/mL; and ETT: from 118.1 ± 14.1 to 158.1 ± 24.7 pg/mL (P = 0.007). The differences between groups were not significant. The incidence of cough was higher in the ETT group (87.5%) than in the LMA group (9.5%) (P < 0.001). Conclusions: Replacing the ETT with the LMA before neurosurgical patients emerge from anesthesia results in a more favorable hemodynamic profile, less cerebral hyperemia, and a lower incidence of cough. Key Words: laryngeal mask airway, neurosurgery, neuroanesthesia, awakening, cerebral hemodynamic response, systemic hemodynamic response, coughing (J Neurosurg Anesthesiol 2015;27:194–202)

H

eart rate (HR) and arterial blood pressure increase during emergence from anesthesia and extubation after neurosurgery, leading to increases in cerebral blood flow, intracranial pressure, and regional brain oxygen saturation (rSO2).1,2 Venous pressure also rises if the patient coughs or gags. Recent studies have reported such hemodynamic effects in up to 50% of patients after supratentorial craniotomy.3,4 Hypertension is multifactorial in origin,3 arising secondary to a sympathetic response that increases circulating catecholamines, particularly norepinephrine.5 Such changes interfere with cerebral autoregulation, which may already be compromised in neurosurgery patients, such that cerebral blood flow comes to depend on systemic blood pressure; these adverse events increase the risk of postoperative intracranial hemorrhage and cerebral edema.2,6 Hemodynamic control and the avoidance of coughing or J Neurosurg Anesthesiol

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

Volume 27, Number 3, July 2015

J Neurosurg Anesthesiol


any Valsalva maneuver are key elements of optimal management during emergence, especially in these high-risk patients.7 Delaying emergence from anesthesia does not lower the risk1,3 and impedes early neurological assessment.1,7,8 Orotracheal intubation is the standard technique for intraoperative control of the airway during neurosurgery. Intubation ensures adequate ventilation for long periods while the head is held in a position that interferes with rapid access to the airway, which is covered and hidden under the surgical field. However, several studies in nonneurosurgical populations have shown that tracheal intubation induces more intense hemodynamic effects (on blood pressure, HR, and rate-pressure product), and circulating catecholamines (especially norepinephrine) than those caused by the use of a laryngeal mask airway (LMA).5,9–14 Those differences are even more intense during emergence and extubation.14–16 The recently published extubation guidelines of the Difficult Airway Society, which are currently based largely on expert opinion, recommend replacing the endotracheal tube (ETT) with a LMA before awakening patients who are at risk because they have highly irritable airways or because the type of surgery they have undergone makes the cardiovascular stimulus of extubation inadvisable.17 The potential protective benefit of this approach has not yet been demonstrated for awakening neurosurgery patients, however. To our knowledge, there is no randomized clinical trial that has provided highquality clinical evidence for this practice in neurosurgical anesthesia even though patients remain highly vulnerable during emergence after craniotomy. Therefore, we wished to test the hypothesis that we would see improved systemic and cerebral hemodynamics and a reduced incidence of coughing in neurosurgery patients emerging from anesthesia after replacement of the ETT with a LMA. The aim of this randomized clinical trial was to compare systemic and cerebral hemodynamic variables and cough incidence during the emergence of neurosurgery patients from general anesthesia either according to standard procedure (ETT extubation) or after replacement of the ETT with a LMA.

MATERIALS AND METHODS This single-site randomized open-label parallel trial was approved by the research ethics committee of Hospital Clı´ nic de Barcelona, a university hospital in Catalonia, Spain, (file number 2011/6294) and registered at clinicaltrials.org (NCT01718470); informed consent was obtained from all patients. Consecutively recruited patients from the neurosurgery department of our center from July 2011 until March 2012 were undergoing elective supratentorial craniotomy for nonvascular procedures. We applied the following exclusion criteria: predicted difficult airway or Cormack-Lehane grade III-IV detected during laryngoscopy, risk of bronchial aspiration (eg, gastroesophageal reflux disease or lower cranial nerve palsy), and uncontrolled arterial hypertension detected Copyright

r

Neurosurgical Patients Awakening With Laryngeal Mask

during preoperative assessment. Patients with contraindications for early emergence based on anesthetic or surgical criteria or as a result of complications developing during surgery were withdrawn from the study. Patients took omeprazole (20 mg) and diazepam (5 to 10 mg) on the ward the night before surgery; in the morning compression sleeves were placed on the patients’ legs. In the operating room, patients were premedicated with intravenous midazolam (1 to 2 mg). Standard monitoring consisted of electrocardiography, pulse oximetry, arterial and central venous pressures, mean arterial pressure (MAP), depth of anesthesia (BIS; Covidien Medical, Boulder, CO), neuromuscular blockade (response to train-of-four stimulation), temperature (S/5; Datex Ohmeda, Helsinki, Finland); rSO2 (INVOS Cerebral/Somatic Oximeter 5100C; Somanetics, Troy, MI), and urine output. An arterial line was inserted before the start of anesthesia induction with the transducer probe located at the level of the foramen of Monro; this line remained in place during the entire study procedure. A transcranial Doppler ultrasound monitor (Intraview; Rimed, Singen, Germany) was fixed at the temporal window to monitor middle cerebral artery (MCA) flow velocity during induction and emergence. General anesthesia was provided with intravenous propofol and remifentanil delivered through a targetcontrolled infusion system (Orchestra Infusion Workstation, Base Primea, Fresenius Vial, Bad Homburg v.d.H., Germany) and intravenous perfusion of rocuronium. After intravenous injection of 1.5 mg/kg of lidocaine, we performed direct Mackintosh laryngoscopy and orotracheal intubation with a reinforced ETT (Lo-Contour Oral/Nasal cuffed tracheal tube; Mallinckrodt, Covidien, Tullamore, Ireland) in all patients (using 7.5 or 8 G, for women or men, respectively). We recorded the Cormack-Lehane grade, requirement for additional instrumentation during intubation (introducers, videolaryngoscope), and number of attempts to achieve insertion. A ventilator (Primus, Dra¨ger Medical Hispania, Madrid, Spain) was connected immediately afterwards and set to maintain a PaCO2 of 35 to 40 mm Hg and a PaO2 of 150 to 200 mm Hg. The target-controlled anesthesia system was set to administer propofol and remifentanil to maintain the bispectral index between 40 and 60 and rocuronium doses to maintain a response between T1 and T2 of a train-of-four. A nasogastric tube was inserted for the duration of surgery (until extubation) and a pharyngeal probe was placed to measure the temperature, which was kept between 351C and 361C by applying warm air to the patient if necessary. A 3-point Mayfield-Kees skeletal fixation headrest held the patient’s head in position during surgery. With general anesthesia and muscle relaxation still in effect after the procedure, we injected paracetamol (1 g) and ondansetron (4 mg) through the intravenous line, removed the Mayfield device, and shifted the patient to supine position if another position had been used during the procedure. The transcranial Doppler ultrasound monitor used during induction was once again placed to

2014 Wolters Kluwer Health, Inc. All rights reserved. Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

www.jnsa.com |

195


Perello´-Cerda` et al

measure MCA flow in the temporal window, and the nasogastric tube was removed. At the end of surgery, the anesthesiologist opened a sealed envelope labeled with software-generated randomized numbers to assign the patients, who were allocated in a 1:1 ratio from nonstratified blocks of 4 patients, to one of 2 groups to emerge from anesthesia with the ETT still in place (ETT group) or after it had been replaced by a LMA (LMA group). In the LMA group, after aspirating pharyngeal secretions and with the patient still under general anesthesia and before modification of any anesthetic drug administration, we extubated the trachea and inserted a ProSeal LMA (Laryngeal Mask Co. Ltd., Le Rocher, Victoria, Mahe Seychelles) using a guided technique. Specifically, the LMA was inserted after first advancing a suction catheter along the drain tube 8 to 10 cm beyond the distal end; the mask was then inserted using a digital technique, allowing the suction catheter to enter the esophagus first, and guiding the tip of the cuff.14,18 A number 4 or 5 mask was chosen for patients weighing 50 to 70 kg or 70 to 100 kg, respectively. The cuff was inflated to a pressure of 60 cm H2O measured with a manometer (VBM Medizintechnik GMBH, Sulz, Germany). We recorded the duration of apnea after disconnection of the ETT from the respirator until the capnography wave had recovered after insertion of the LMA. Ventilation then continued with the same parameters as had been used earlier. Target-controlled infusion of anesthetics and administration of neuromuscular relaxants were then stopped so the patient could emerge from anesthesia. Sugammadex (2 mg/kg) was given to reverse the neuromuscular blockade. Gentle manual ventilatory assistance was then provided until the patient resumed spontaneous breathing and responded to simple commands; the LMA or the ETT, depending on the group, was then removed. Patient variables (age, weight, height, sex) and relevant aspects of past medical history, such as controlled hypertension were also registered. Hemodynamic variables (blood pressure, HR, rSO2, MCA flow velocity), were recorded at 8 moments: baseline, in the operating room before anesthetic induction; end of surgery, before awakening (ETT group) or before ETT replacement (LMA group); and throughout emergence from anesthesia at 1, 5, 10, 15, 30, and 60 minutes after extubation or LMA removal (according to group assignment). The last blood pressure and HR measurements were taken in the postoperative recovery room. Respiratory variables (including end-tidal carbon dioxide concentration) were recorded during mechanical ventilation; arterial blood gases were also recorded. If systolic arterial pressure (SAP) exceeded 160 mm Hg and had risen >20% over baseline, an antihypertensive agent (urapidil) was administered at a dose the anesthesiologist considered appropriate and drug doses were recorded. We determined norepinephrine plasma concentration with a radioimmunoassay kit (Noradrenalin RIA, IBL, Hamburg, Germany) at 3 moments: before induction and 1 and 30 minutes after extubation. The normal norepinephrine concentration range in our laboratory is 136 to 364 pg/mL.

196 | www.jnsa.com


To facilitate analysis, rSO2 and norepinephrine plasma concentrations were expressed as percent increases from each patient’s baseline measurement. We also calculated the rate-pressure product (HR MAP). MAP was the primary endpoint. The other systemic and cerebral hemodynamic measures (SAP, HR, rSO2, MCA flow velocity), norepinephrine plasma concentration, and cough incidence were the secondary endpoints. Any coughing episode during the study was noted. We used specialized software (Rugloop, Demed BVBA, Temse, Belgium) approved for collecting medical research data from the monitor and respirator to record all data on a hard disk for storage in real time. Patient data were coded consecutively, ensuring that only the investigator would be able to identify individual patients to safeguard privacy.

Statistical Analysis We calculated that we would need 21 patients in each group to detect between-group differences of 10 mm Hg in MAP, the primary endpoint, assuming a SD of the differences of 11 mm Hg, with a type I error of 5% and power of 80% (2-tailed testing). Mean (± SD) results and 95% confidence intervals (CI) were tabulated for each group. The global estimated group effect and 95% CI, adjusted to baseline and followup, was also calculated for each variable for inferential purposes. The study variables were analyzed with a longitudinal linear model using the general estimating equation methodology to account for within-subject correlations over time with the assumption of an unstructured correlation matrix. The models included time and study group as the main factors. To evaluate the statistical significance of differences at each time, the same model including the interaction time by group was performed again for the evaluation of each dependent variable. Bonferroni correction of P values was used to adjust for multiplicity in time-by-time analyses. Homogeneity of groups at baseline was tested. All analyses were made with SPSS version 20 (IBM, Armonk, NY), assuming the superiority of the intervention and a 2-tailed type I error of 5%.

RESULTS A total of 48 patients were randomized (Fig. 1). Patient characteristics are shown in Table 1. No differences were found between groups at baseline. The Cormack-Lehane grade was II or lower in all cases; the reinforced ETTs were easily inserted in all cases. The LMA was inserted on first attempt in all LMA-group patients. The mean duration of apnea between removal of the ETT and effective ventilation through the LMA was 43.6 ± 27.4 seconds. Decreases in PaO2 and increases in end-tidal carbon dioxide concentration in expired air were never detected. SAP, HR, and rate-pressure product on emergence from anesthesia were significantly higher than at baseline in both groups (highest of the significant P < 0.02) (Figs. 2A–D). Copyright

r

2014 Wolters Kluwer Health, Inc. All rights reserved.




Patients randomized n = 48

Group ETT n = 24

Group PLMA n = 24 Lost for prolonged surgery n=3

Lost for prolonged surgery n=2 Lost for intraoperative complications n=1

Included patients n = 21

Included patients n = 21

FIGURE 1. Flow diagram of patient enrollment and data analysis.

MAP was higher in the ETT group, with a global estimated mean between-group difference of 11.9 mm Hg (95% CI, 2.1-21.8 mm Hg) (P = 0.017) (Fig. 2B). Similarly, SAP rose higher in the ETT group than in the LMA group. The global estimated mean between-group difference was 35.6 mm Hg (95% CI, 8.9-62.3 mm Hg) (P = 0.009) (Fig. 2A). Mean differences of MAP and SAP between groups are shown in Table 2. HR was also higher in the ETT group, with a global estimated mean betweengroup difference of 7.2 beats/min (95% CI, 0.7-13.7 beats/ min) (P = 0.03) (Fig. 2C), as well as the rate-pressure product, with a global estimated mean difference of 1045.4 (95% CI, 440.8-1650) (P = 0.001) (Fig. 2D). Most differences were reduced after 30 minutes. Antihypertensive medication was administered at least once to more ETT-group patients than LMA-group patients (9 [42.9%] vs. 3 [14.3%] patients, respectively; P = 0.04). In addition, more ETT-group patients (6 [28.6%]) than LMA-group patients (2 [9.5%]) required repeated doses of antihypertensive agents to control pressure. All ETT-group patients known to have controlled chronic hypertension before surgery required administration of an antihypertensive during extubation, whereas only one of the 4 known chronically hypertensive patients in the LMA group required treatment. The MCA flow velocity was significantly higher in both groups on emergence from anesthesia (P < 0.04), but no between-group differences were observed (Fig. 3A).

TABLE 1. Patient Characteristics* Age (y) Sex (M/F) Height (cm) Weight (kg) BMI (kg/m2) Controlled hypertension (n)

ETT Group

LMA Group

52 ± 14 11/10 167 ± 10 66 ± 14 24 ± 4 4

51 ± 16 11/10 169 ± 12 72 ± 14 26 ± 6 4

*Values are expressed as mean ± SD or number of patients. BMI indicates body mass index; ETT, endotracheal tube; LMA, laryngeal mask airway.

Copyright

r


rSO2 was also significantly higher on emergence in both the groups (P < 0.001) (Fig. 3B). The percent increase in rSO2 between baseline and emergence was greater in the ETT group than in the LMA group, with a mean difference of 26.1% (95% CI, 9.1-43.2%) (P = 0.002) (Fig. 3B). Plasma norepinephrine concentrations varied greatly. For 21 determinations (16.7% of the total number of tests), the values did not reach the test kit’s detection limit. Mean concentrations were higher on emergence from anesthesia than at baseline in both groups (in the LMA group at baseline: 87.5 ± 7.1 pg/mL; at 1 minute: 103.4 ± 12.9 pg/mL; at 30 minutes: 125.6 ± 17.3 pg/mL; and in the ETT group at baseline: 118.1 ± 14.1 pg/mL; at 1 minute: 113.6 ± 13.4 pg/ mL; at 30 minutes: 158.1 ± 24.7 pg/mL) (P = 0.007). The percent increase in norepinephrine concentration between baseline and 30 minutes after emergence tended to be higher in the ETT group (67% ± 37% over baseline) than in the LMA group (43% ± 21% over baseline), but the differences were not significant (Fig. 4). Cough episodes were registered in 18 of 21 patients (87.5%) in the ETT group and in only 2 patients (9.5%) in the LMA group (P < 0.001). No patient had hemorrhagic complications in the postoperative period.

DISCUSSION Replacing an ETT with a ProSeal LMA before awakening at the end of neurosurgery reduced the hemodynamic impact of emergence from anesthesia in our study in terms of MAP reduction, the main outcome measure, as well as SAP, HR, and rate-pressure product, attenuating the effect on cerebral hyperemia. The incidence of coughing was also lower with LMA use. To our knowledge, this is the first trial to compare the effects of LMA use in neurosurgical patients, who are particularly susceptible to hemodynamic changes. Even though a statistically significant global mean difference of 11.9 mm Hg in MAP between the groups may not seem numerically impressive, in these neurosurgery patients it would be clinically significant. It must be remembered that all our patients were continuously evaluated for hypertension, which when found was treated following a standard protocol. Even so, most patients in the ETT group developed SAP over 150 mm Hg and MAP over 97 mm Hg, and between-group differences were statistically significant even though 42.9% of ETT patients had received antihypertensive medication that had probably blunted a more severe rise in blood pressure. In contrast, only 14.3% in the LMA group had to be treated for hypertension. The 3-fold greater need for antihypertensive drugs in the ETT group reflected a higher incidence of blood pressure surges, which are events that raise the risk of postoperative intracranial hemorrhage. We are therefore confident that the differences were also clinically relevant in this setting. The statistically significant changes all our awakening neurosurgery patients experienced in systemic hemodynamic


www.jnsa.com |

197



A 150 140 130 120 110

ETT LMA

Mean Blood Pressure (mmHg)

∗

90 85 80 75 70

ETT LMA

60

m in

m in

m in

30

10

15

m in

m in 5

ex t

B ub efo at re io n 1 m in

Ba se lin e

m in 60

m in

m in

30

10

15

m in

m in 5

m in

65

1

Ba se lin e ex B tu ef ba or tio e n

95

90 85

∗

80 75 70 65 60

ETT LMA

55 50

Heart Rate x Mean Blood pressure

D

95

9,500 9,000 8,500 8,000 7,500

∗

∗

∗

7,000

∗

6,500 6,000 5,500 ETT LMA

5,000

60

m

in

in m 30

in m 15

10

m

in

in m

ex

5

in m

se Ba

1

lin

e

in m 60

30

m

in

in m 15

in m 10

in m

ex

5

e lin se Ba

tu Bef ba or tio e n 1 m in

4,500

tu Bef ba or tio e n

Systolic Blood Pressure (mmHg)

100

100

Heart Rate (bpm)


B 160

C

FIGURE 2. Changes in mean hemodynamic variables in patients with an endotracheal tube (ETT) or a laryngeal mask airway (LMA) during the study; bars indicate the SEM. Systolic blood pressure, mean arterial pressure (MAP), heart rate (HR), and the rate-pressure product (HRMAP) were significantly higher on extubation and emergence from general anesthesia compared with baseline (P < 0.001 for all variables, [A–D]) and the between-group differences overall were also significant. A, Systolic blood pressure, significant between-group differences overall (P = 0.009). B, MAP, significant between-group differences overall (P = 0.017) and at specific moments after emergence (*) (Bonferroni test; P < 0.00625). C, HR, significant between-group differences overall (P = 0.03) and at specific moments after emergence (*) (Bonferroni test; P < 0.00625). D, rate-pressure product, significant between-group differences overall (P < 0.01) and at specific moments after emergence (*) (Bonferroni test; P < 0.00625).

variables as well as rSO2 and MCA flow velocity are consistent with previous reports.1,2,5–8 Bruder et al2 reported that extubation induced a 60% rise from baseline in MCA flow velocity measured by transcranial Doppler ultrasound and a 75% increase in jugular venous oxygen saturation. This effect was confirmed in our study by similar rises in rSO2 on emergence from anesthesia. The higher values of rSO2 during extubation in the ETT group are probably attributable to cerebral hyperemia secondary to systemic hypertension and represent a sign of impaired cerebral hemodynamics produced during anesthesia emergence. This effect was independent of anesthetic technique in the study of Bruder et al,2 in which values did not return to normal until 60 minutes after extubation. In our study, all variables began to recover after 30 minutes, probably because of early use of ur-

198 | www.jnsa.com

apidil. Delaying awakening until later in the recovery room does not attenuate hemodynamic responses, and neurosurgery patients who are managed in this way have been shown to have higher oxygen consumption and norepinephrine concentrations than patients who emerge in the operating room, with deleterious impact on cerebral blood flow.1 Such hemodynamic changes have been linked to intracranial bleeding and cerebral edema. In a retrospective study, Basali et al6 detected an incidence of cerebral hemorrhage of 0.77% after craniotomy, finding that 62% of patients with this complication had developed hypertension in the immediate postoperative period. In that study, the presence of controlled hypertension before surgery did not seem to confer risk of postoperative cerebral hemorrhage. However, several reports have established a relationship between prior history of Copyright

r






TABLE 2. Mean Differences in MAP and SAP Between Groups During Emergence Variables

Time

SAP (mm Hg)

MAP (mm Hg)

P

Mean (95% CI) Differences Between Groups

Baseline Before extubation 1 min 5 min 10 min 15 min 30 min 60 min Baseline Before extubation 1 min 5 min 10 min 15 min 30 min 60 min

–3.8 –2.3 –13.9 –10.6 –5.3 –3.6 2 –1.5 –3.9 –4.1 –10.8 –7.6 –5.5 –5.7 –2 –0.8

(–17 to 9.4) (–12 to 7.3) (–25.1 to –2.7) (–20.2 to –1) (–14.8 to 4.2) (–12.4 to 5.1) (–8 to 12.1) (–11.4 to 8.4) (–12.2 to 4.3) (–10.4 to 2.2) (–17.3 to –4.3) (–13.4 to –1.9) (–11.2 to 0.2) (–11.1 to –0.4) (–7.9 to 3.9) (–6.9 to 5.2)

0.571 0.6353 0.0149 0.0308 0.2765 0.4173 0.6905 0.765 0.3504 0.2048 0.0011 0.0094 0.0588 0.0345 0.5016 0.7856

CI indicates confidence interval; MAP, mean arterial pressure; SAP, systolic arterial pressure.

hypertension and postoperative hematoma.19 Our results suggest that in such patients, the rise in blood pressure during emergence from anesthesia can be attenuated by inserting a LMA before emergence, as less antihypertensive medication was needed in this subgroup. All 4 patients in the ETT group with a history of chronic hypertension before surgery required antihypertensive medication during emergence, whereas only one of the 4 controlled-hypertensive patients in the LMA group required urapidil. Although our study was not designed to enroll enough patients to detect differences in the subgroup of patients with controlled chronic hypertension, we find that our results are consistent with those of previous studies of anesthesia induction.13 In non-neurosurgery patients with controlled chronic hypertension, Bhattacharya et al13 demonstrated that intubation increased HR

B

LMA

30

m

in

in m 15

m

lin se

m 30

in

90

in

in m 15

10

m

in

in 5

m

in m 1

tu Bef ba or tio e n ex

Ba

se

lin

e

30

ETT

95

10

LMA

in

ETT

35

100

m

40

105

5

45

110

in

50

m

55

115

1

60

120

tu Bef ba or tio e n

65

125

ex

70

130

e

75

Ba

Middle cerebral artery velocity (cm/s)

80

Percentage change SrO2 respect to Baseline (%)

A

and both systolic and diastolic blood pressure in patients intubated with an ETT rather than with a LMA; extubation was not studied. As a consequence, we think that our results allow us to suggest that the beneficial effects of patient awakening with a LMA in place would be of special interest in this subgroup of chronic hypertensive patients undergoing craniotomy. The high incidence of cough in the ETT group in our study contrast with the incidence of 3.6% reported in another recent study of patients emerging from anesthesia after craniotomy.20 The difference may be explained by our recording of any occurrence of cough during the awakening process. Our cough incidence rates are similar to the rates of 9% and 83%, for ProSeal LMA and ETT, respectively, reported by Carron et al5 for obese patients emerging from anesthesia after laparoscopy in a study

FIGURE 3. Changes in mean cerebral hemodynamic variables in patients with an endotracheal tube (ETT) or a laryngeal mask airway (LMA) during the study; bars indicate the SEM. Both middle cerebral artery (MCA) flow velocity and regional cerebral oxygen saturation (rSO2) expressed as a mean percent change from baseline were significantly higher on emergence from anesthesia compared with baseline (P < 0.001). A, MCA flow velocity changes were not significant between groups, probably because of insufficient statistical power. B, Between-group changes in rSO2 were significant overall (P = 0.02). Copyright

r


www.jnsa.com |

199


Plasma norepinephrine, % of baseline


100 80 60 ETT LMA

40 20 0 -20 Baseline

1 min

30 min

FIGURE 4. Changes in mean norepinephrine plasma concentrations in patients with an endotracheal tube (ETT) or a laryngeal mask airway (LMA) on emergence from general anesthesia; data points indicate percent change in the group means from baseline and the bars indicate the SEM. Norepinephrine plasma concentrations rose in both groups at emergence from anesthesia (P = 0.007). The percent increase in concentration between baseline and 30 minutes after extubation tended to be higher in the ETT group than in the LMA group, but the differences were not significant.

that used the same criteria for recording this event that we used. The LMA has been used successfully in a variety of neurosurgical procedures, such as ventriculoperitoneal shunt,21 lumbar spine microsurgery,22,23 and awake craniotomy.24 Nevertheless, the gold standard for airway management during neurosurgery is ETT insertion, given the complexity of these procedures, their duration, and patient positioning, which makes intraoperative airway access difficult. However, these circumstances do not preclude replacing an ETT with a LMA at the end of the procedure, once the patient has been repositioned. If this step is taken, the advantages of the LMA become available during emergence, the moment when systemic and cerebral hemodynamic changes are most intense even when extubation procedures have been followed meticulously.1,2,7,8 We were able to insert the LMA on the first attempt in all cases in this study. None of our patients presented any complications due to ETT replacement with a LMA, and the short apnea time was well tolerated. Apnea could be shortened even further by inserting the LMA behind the ETT (Bailey technique) before extubating,4,25 but that maneuver could be seen to be slightly more complicated and we chose not to use it to increase the generalizability of this study. The potential risks involved in replacing an ETT with a LMA in this scenario include momentary loss of control of the airway and the aspiration of gastric content while the airway is unprotected, but LMA insertion by experienced anesthesiologists in selected patients under adequate depth of anesthesia is fast, effective, and safe. The Bailey technique (inserting the LMA behind the ETT), a fairly simple procedure in expert hands, would facilitate immediate ventilation, thus decreasing

200 | www.jnsa.com


the amount of apnea time and the possible increase in PaCO2, something which in and of itself can cause both systemic and cerebral hemodynamic effects that are deleterious. However, it must be remembered that such replacement techniques cannot be recommended in patients with anticipated or known difficult airway or at high risk of bronchial aspiration.17 In our study, all patients had fasted and none had added risk of bronchial aspiration (no lower cranial nerves affected and no history of gastric reflux). Our systematic use of the nasogastric tube in these patients during surgery served as an additional guarantee of reduced gastric content and lower risk of bronchial aspiration. Other important safety factors are the immediate availability of material to establish an emergency airway and the choice of the most appropriate supraglottic device. We used the ProSeal LMA because it seals the airway better than simpler devices, allows evacuation of gastroesophageal contents, and guided insertion is possible.26 Several studies have explored methods to reduce the effects of extubation on cerebral hemodynamics. Drugs such as lidocaine,27 esmolol,28,29 diltiazem/nicardipine,30 fentanyl,31 and dexmedetomidine32 have been used with this purpose in mind. Remifentanil administration, which is widely used in neurosurgery, has also been suggested as a strategy for smoothing emergence generally,17 although no studies have demonstrated its efficacy in the type of patients we studied. The use of remifentanil during this period improves analgesia and could reduce hemodynamic impairment but doses must be carefully titrated to avoid neurological depression (which could mask a neurological complication), as well as respiratory depression (which could cause hypercapnia and further hyperemia, thus abolishing the beneficial effect of hemodynamic control). Among the limitations of this study we mention that we did not record PaCO2 after extubation, although this variable was monitored strictly throughout surgery. As ETTs had been removed when ventilation was restored and spontaneous breathing was carefully monitored, we believe any possible increase in PaCO2 that might have occurred would not have influenced the results. We must also note the small effect urapidil has on increasing HR, because a certain degree of the observed increase in that variable in patients who required that drug may have been drug related. Furthermore, as we only evaluated the first postoperative hour of emergence, we do not know how long the beneficial effects of LMA use lasted or how much the risk of cerebral hemorrhage was potentially reduced. Regarding norepinephrine levels, it is interesting that we could not demonstrate statistically significant differences as have been reported in another study.5 Some norepinephrine concentrations were below the detection limit of the test we used. We double checked plasma measurements and ruled out the possibility of malfunctioning radioimmunoassay kits and now hypothesize that detection limit concerns may have been underreported in other studies. For example, one study analyzing the stress response to tracheal intubation in patients undergoing Copyright

r





coronary artery surgery found values very close to zero and SDs below the zero line, yet the authors did not discuss any concerns about the technique’s detection limits.33 We note the possibility that very low values may be attributable to the effect benzodiazepines and certain other anesthetics have in some patients. One study, for example, showed that the neuroendocrine response was lower in patients undergoing craniotomy under anesthesia with propofol and remifentanil (as in our present study) than it was in patients under inhaled anesthetics.20 We also cannot rule out that an increase in sympathetic outflow or circulating factors may play a role in patients’ responses. Also relevant is the fact that the sample size was calculated to detect hemodynamic variations, not changes in other variables such as norepinephrine plasma concentrations. Sample size, therefore, may account for the lack of statistical significance in the between-group comparisons of some variables. Given the relatively low incidence of postoperative cerebral hemorrhage and edema, we cannot draw conclusions regarding the ultimate protective benefit of LMA use during emergence. More and larger studies should be designed to confirm our findings before it can be firmly established that lowering hemodynamic risk in this way leads to improved outcomes overall. We conclude that patients who are ventilated through a LMA when emerging from anesthesia after supratentorial craniotomy display a more favorable hemodynamic profile and less cerebral hyperemia. They also have a lower incidence of cough than patients who emerge from anesthesia and are extubated in the traditional way. Replacement of the ETT with a LMA may therefore offer a way to reduce the risk of the serious potential complications of hemodynamic stability in vulnerable neurosurgical patients. Use of the LMA in this way may be of special interest in patients with a history of chronic hypertension.


5.

6.

7. 8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

ACKNOWLEDGMENTS The authors thank neurosurgery nurses Gemma Cabedo, Marisa Ferrer, Gloria Pastor, and Ana Quintana for their assistance with the biochemistry analyses; and Dr. Gregori Casals and Dr. Manuel Morales and their laboratory support staff for their measurements of norepinephrine plasma concentrations (Biochemistry and molecular genetics laboratory, Hospital Clıńic de Barcelona, Spain). They also thank Mary Ellen Kerans for advice on English usage.

18.

19. 20.

21.

REFERENCES 1. Bruder N, Stordeur JM, Ravaussin P, et al. Metabolic and hemodynamic changes during recovery and tracheal extubation in neurosurgical patients: immediate versus delayed recovery. Anesth Analg. 1999;89:674–678. 2. Bruder N, Pellissier D, Grillot P, et al. Cerebral hyperemia during recovery from general anesthesia in neurosurgical patients. Anesth Analg. 2002;94:650–654. 3. Olsen K, Pedersen C, Madsen J, et al. Vasoactive modulators during and after craniotomy: relation to postoperative hypertension. J Neurosurg Anesthesiol. 2002;14:171–179. 4. Wong AYC, O’Regan AM, Irwin MG. Total intravenous anaesthesia with propofol and remifentanil for elective neurosurgical

Copyright

r

22.

23.

24.

25.

procedures: an audit of early postoperative complications. Eur J Anesthesiol. 2006;23:586–590. Carron M, Veronese S, Gomiero W, et al. Hemodynamic and hormonal stress responses to endotracheal tube and ProSeal Laryngeal Mask Airway for laparoscopic gastric banding. Anesthesiology. 2012;117:309–320. Basali A, Mascha E, Kalfas I, et al. Relation between perioperative hypertension and intracranial hemorrhage after craniotomy. Anesthesiology. 2000;93:48–54. Bruder N. Awakening management after neurosurgery for intracranial tumours. Curr Opin Anaesthesiol. 2002;15:477–482. Bruder N, Ravaussin P. Modifications he´modynamiques ce´re´brales et syste´miques au moment du re´veil en neurochirurgie. Ann Fr Anesth Reanim. 2004;23:410–416. Imai M, Matsumura C, Hanaoka Y, et al. Comparison of cardiovascular response to airway management: fiberoptic intubation using a new adapter, laryngeal mask insertion or conventional laryngoscopic intubation. J Clin Anesth. 1995;7:14–18. Oczenski W, Krenn H, Dahaba AA, et al. Hemodynamic and catecholamine stress responses to insertion of the Combitube, laryngeal mask airway or tracheal intubation. Anesth Analg. 1999;88:1389–1394. Wood ML, Forrest ET. The haemodynamic response to the insertion of the laryngeal mask airway: a comparison with laryngoscopy and tracheal intubation. Acta Anaesthesiol Scand. 1994;38:510–513. Akbar AN, Muzi M, Lopatka CW, et al. Neurocirculatory responses to intubation with either an endotracheal tube or laryngeal mask airway in humans. J Clin Anesth. 1996;8:194–197. Bhattacharya D, Ghosh S, Chaudhuri T, et al. Pressor responses following insertion of laryngeal mask airway in patients with controlled hypertension: comparison with tracheal intubation. J Indian Med Assoc. 2008;106:787–788. Brimacombe J. Pathophysiology. In: Brimacombe J, ed. Laryngeal Mask Anesthesia Principles and Practice. 2nd ed. Philadelphia: Elsevier Ltd.; 2005:105–136. Russo S, Goetze B, Troche S, et al. LMA-ProSeal for elective postoperative care one the intensive care unit: a prospective randomized trial. Anesthesiology. 2009;111:116–121. Koga K, Asai T, Vaughan RS, et al. Respiratory complications associated with tracheal extubation and use of laryngeal mask during emergence from anaesthesia. Anaesthesia. 1998;53:540–544. Popat M, Mitchell V, Dravid R, et al. Difficult Airway Society Guidelines for the management of tracheal extubation. Anaesthesia. 2012;67:318–340. Garcı´ a Aguado R, Vinoles J, Brimacombe J, et al. Suction catheter guided insertion of the ProSeal laryngeal mask airway is superior to the digital technique. Can J Anaesth. 2006;53:398–403. Seifman MA, Lewis PM, Rosenfeld JV, et al. Postoperative intracranial haemorrhage: a review. Neurosurg Rev. 2011;34:393–407. Citerio G, Pesenti A, Latini R, et al. Neuromorfeo Study Group. A multicentre, randomised, open-label, controlled trial evaluating equivalence of inhalational and intravenous anaesthesia during elective craniotomy. Eur J Anaesthesiol. 2012;29:371–379. Hurtado P, Valero R, Tercero J, et al. Experiencia con el uso de la mascarilla ları´ ngea ProSeal en pacientes sometidos a cirugı´ a de derivacioń ventrı´ culo peritoneal. Rev Esp Anestesiol Reanim. 2011;58:362–364. Sharma V, Verghese C, McKenna PJ. Prospective audit on the use of LMA-Supreme for airway management of adult patients undergoing elective orthopedic surgery. Br J Anaesth. 2010;105:228–232. Lo´pez AM, Valero R, Hurtado P, et al. Comparison of the LMA Supreme TM with the LMA ProSeal TM for airway management in patients anaesthetized in prone position. Br J Anaesth. 2011;107: 265–267. Tongier WK, Joshi GP, Landers DF, et al. Use of the laryngeal mask airway during awake craniotomy for tumor resection. J Clin Anesth. 2000;12:592–594. Nair I, Bailey PM. Review of uses of the laryngeal mask in ENT anaesthesia. Anaesthesia. 1995;50:898–900.


www.jnsa.com |

201



26. Brimacombe J, Keller C, Judd DV. Gum elastic bougie-guided insertion of the ProSeal laryngeal mask airway is superior to the digital and introducer tool techniques. Anesthesiology. 2004;100:25–29. 27. Andrzejowski J, Francis G. The efficacy of lidocaine administered via the LITA tracheal tube in attenuating the extubation response in beta-blocked patients following craniotomy. Anaesthesia. 2002;57: 399–401. 28. Bilotta F, Lam AM, Doronzio A, et al. Esmolol blunts postoperative hemodynamic changes after propofol-remifentanil total intravenous fast-track neuroanesthesia for intracranial surgery. J Clin Anesth. 2008;20:426–430. 29. Grillo P, Bruder N, Auquier P, et al. Esmolol blunts the cerebral blood flow velocity increase during emergence from anesthesia in neurosurgical patients. Anesth Analg. 2003;96:1145–1149.

202 | www.jnsa.com


30. Tsutsui T. Combined administration of diltiazem and nicardipine attenuates hypertensive responses to emergence and extubation. J Neurosurg Anesthesiol. 2002;14:89–95. 31. Bhagat H, Dash HH, Bithal PK, et al. Planning for early emergence in neurosurgical patients: a randomized prospective trial of low-dose anesthetics. Anesth Analg. 2008;107:1348–1355. 32. Soliman RN, Hassan AR, Rashwan AM, et al. Prospective, randomized study to assess the role of dexmedetomidine in patients with supratentorial tumors undergoing craniotomy under general anaesthesia. Middle East J Anesthesiol. 2011;21:325–334. 33. Kahl M, Eberhart LHJ, Behnke H, et al. Stress response to tracheal intubation in patients undergoing coronary artery surgery: direct laryngoscopy versus an intubating laryngeal mask airway. J Cardiothor Vasc Anesth. 2004;18:275–280.

Copyright

r



Comparison of Supreme Laryngeal Mask Airway and ProSeal Laryngeal Mask Airway during Cholecystectomy.

Comparisons of clinical performance of Guardian laryngeal mask with laryngeal mask airway ProSeal.

Supreme Laryngeal Mask Airway versus Face Mask during Neonatal Resuscitation: A Randomized Controlled Trial.

Ascaris: An unusual cause of airway obstruction during general anesthesia with ProSeal laryngeal mask airway.

Effect of Paralysis at the Time of ProSeal Laryngeal Mask Airway Insertion on Pharyngolaryngeal Morbidities. A Randomized Trial.

Head elevation by 3 vs. 6 cm in ProSeal laryngeal mask airway insertion: a randomized controlled trial.

Position of bite block in size 1 laryngeal mask airway ProSeal™ and stability during insertion.

Comparing performance of ProSeal laryngeal mask airway and I-gel in anesthetized adult patients.

Muscle relaxant effects on insertion efficacy of the laryngeal mask ProSeal(®) in anesthetized patients: a prospective randomized controlled trial.

The uses of laryngeal mask airway ProSeal™ and endobronchial blocker for one lung anesthesia.

Comparison of the i-gel and the laryngeal mask airway proseal during general anesthesia: a systematic review and meta-analysis.

The laryngeal mask airway: clinical appraisal in 250 patients.

Comparison of ProSeal laryngeal mask airway (PLMA) with cuffed and uncuffed endotracheal tubes in infants.

Desflurane versus sevoflurane in pediatric anesthesia with a laryngeal mask airway: A randomized controlled trial.

Comparison of the size 3 and size 4 ProSeal™ laryngeal mask airway in anesthetized, non-paralyzed women: a randomized controlled trial.

Suction catheter guided insertion of ProSeal laryngeal mask airway: Experience by untrained physicians.

Comparison of the Supreme™ and ProSeal™ laryngeal mask airways in infants: a prospective randomised clinical study.

The laryngeal mask airway.

Laryngeal mask airway.

Laryngeal mask airway.

The optimum dose of intranasal remifentanil for laryngeal mask airway insertion during sevoflurane induction in children: a randomized controlled trial.

Techniques for the insertion of the ProSeal laryngeal mask airway: comparison of the Foley airway stylet tool with the introducer tool in a prospective, randomized study.

Difficult Laryngeal Mask Airway Placement.

A problem with a laryngeal mask airway.