- Email: [email protected]
Stress response to tracheal intubation in patients undergoing coronary artery surgery: direct laryngoscopy versus an intubating laryngeal mask airway
Stress Response to Tracheal Intubation in Patients Undergoing Coronary Artery Surgery: Direct Laryngoscopy Versus an Intubating Laryngeal Mask Airway Martin Kahl, MD,* Leopold H.J. Eberhart, MD,* Hagen Behnke, MD,* Stefan Sa¨nger, MD,* Udo Schwarz, MD,* Sebastian Vogt, MD,† Rainer Moosdorf, MD,† Hinnerk Wulf, MD,* and Go¨tz Geldner, MD* Objectives: Stress response caused by endotracheal intubation may be harmful for the coronary or cerebral circulation of high-risk patients. This study evaluated the hypothesis that tracheal intubation via an intubating laryngeal mask airway is associated with less cardiovascular and endocrine stress response than the conventional technique using direct laryngoscopy. Design: Randomized, patient-blinded trial. Setting: University department (single center). Participants: Eighty-six patients undergoing elective coronary artery surgery. Intervention: Tracheal intubation was performed via an intubating laryngeal mask or by conventional direct laryngoscopy after standardized induction of anesthesia. Measurements and Main Results: Electrocardiogram with automatic ST-segment analysis and invasive measured blood pressure were recorded continuously and blood samples to analyze norepinephrine plasma levels were taken at 4 times. Catecholamine concentrations and the pressure-
rate product were analyzed by using an analysis of variance for repeated measures. In both groups, the pressure-rate product (p ⴝ 0.003) and norepinephrine concentrations (p < 0.0001) significantly decreased after induction of anesthesia. However, the fall was more marked in the patients intubated via the laryngeal mask (p ⴝ 0.031) than in patients receiving direct laryngoscopy. There were neither signs of cardiac ischemia nor major adverse events during induction of anesthesia. Conclusions: Reduction of cardiovascular and endocrine stress response associated with endotracheal intubation is more pronounced when performed via the intubating laryngeal mask. Thus, this technique can be helpful in high-risk cardiac patients. © 2004 Elsevier Inc. All rights reserved.
D
All patients received oral premedication with clorazepate, 20 mg, the evening before surgery. On the day of surgery, patients were premedicated with clorazepate, 20 mg, and clonidine, 150 g, about 2 hours before arriving in the operating room. Antihypertensive agents and beta-blockers were continued. After arrival in the operating room, intravenous and radial artery catheters were inserted under topical anesthesia. Five hundred milliliters of hydroxyethyl starch (10%, 200,000 Dalton) and 500 mL of lactated Ringer’s solution were rapidly infused followed by an additional infusion of 500 mL as clinically necessary, mainly to treat hypotension. A 12-lead electrocardiogram (ECG) with automatic ST-segment analysis, oxygen saturation, invasive blood pressure monitoring, and a bispectral index (BIS)-monitoring device were connected to the patients. These variables were continuously recorded online by using a laptop computer connected to a Solar 9500 monitor (General Electric, Munich, Germany). After a 3-minute period of preoxygenation, anesthesia was induced using midazolam (0.05 mg/kg) and a continuous infusion of propofol (10 mg/kg/h) and sufentanil (10 g/kg/h) until loss of consciousness. When the BIS level reached a value of 50, the doses of the anesthetics were reduced (propofol: 3 mg/kg/h, sufentanil: 1.5 g/kg/h) to maintain this level of anesthesia and rocuronium bromide, 0.9 mg/kg, was administered for muscle relaxation after ventilation via a face mask with 100% oxygen was ensured. Two minutes later, conventional laryngoscopy was performed in the CL group using a Macintosh 3 laryngoscope and the trachea was intubated with a 7.5-mm low-pressure high-volume cuffed endotracheal tube (Super Safety; Ru¨sch, Kernen, Germany).
IRECT LARYNGOSCOPY TO facilitate tracheal intubation produces a marked stress response.1 Although these alterations are short lived, they might provoke detrimental effects on the coronary or cerebral circulation of high-risk patients. Direct laryngoscopy resulting in distention of supraglottic tissue is the major cause of stress response caused by conventional endotracheal intubation2 (eg, blind intubation has been shown to produce significantly fewer hemodynamic alterations than intubation after direct laryngoscopy).3 The intubating laryngeal mask (ILM) airway is a new device that facilitates tracheal intubation without laryngoscopy. Because of the high success rate, it is a suitable technique compared with other techniques of endotracheal intubation without direct laryngoscopy.4 The cardiovascular effects of inserting a laryngeal mask airway have been shown to be minimal and are comparable to that of an oropharyngeal device.5 Studies that have previously compared the hemodynamic and endocrine stress responses of endotracheal intubation via an ILM or after direct laryngoscopy showed conflicting results.6,7 Thus, this trial was undertaken to evaluate the stress response of the 2 techniques in patients undergoing coronary artery surgery because these patients are most likely to benefit from a gentle insertion of the endotracheal tube.
KEY WORDS: tracheal intubation, stress response, intubating laryngeal mask airway, cardiac surgery
METHODS After ethics committee approval was obtained, patients gave their written and informed consent before entering the study. Eighty-six male American Society of Anesthesiologists status 3 patients undergoing coronary artery bypass surgery were enrolled into the study, which was performed according to the recommendations of the CONSORT Statement (http://www.consort-statement.org). The patients were randomized immediately before induction of anesthesia for tracheal intubation via the ILM airway (ILM group) or by conventional laryngoscopy (CL group). For this purpose, sealed numbered envelopes were used.
From the *Department of Anesthesiology and Intensive Care Medicine and †Department of Cardiothoracic Surgery, Philipps-University Marburg, Marburg, Germany. Address reprint requests to Leopold H.J. Eberhart, MD, Department of Anesthesiology and Intensive Care, Philipps-University Marburg, Baldingerstr. 1, D–35033 Marburg, Germany. E-mail: eberhart@ mailer.uni-marburg.de © 2004 Elsevier Inc. All rights reserved. 1053-0770/04/1803-0005$30.00/0 doi:10.1053/j.jvca.2004.03.005
Journal of Cardiothoracic and Vascular Anesthesia, Vol 18, No 3 (June), 2004: pp 275-280
275
276
KAHL ET AL
Table 1. Biometric Variables of the Patients of Both Groups
Age Weight Height Body mass index Ejection fraction Number of affected vessels (n ⫽ 1) (n ⫽ 2) (n ⫽ 3) Patients with IDDM NIDDM Patients pretreated with ACE inhibitors Beta-blockers Calcium antagonists Hydroxyethyl starch (mL) Lactated Ringer’s solution (mL) Propofol until tracheal intubation (mg) Sufentanil until tracheal intubation (mg)
ILM Group (n ⫽ 43)
Conventional Laryngoscopy Group (n ⫽ 43)
66 ⫾ 9 84 ⫾ 10 173 ⫾ 5 28.1 ⫾ 3.1 63 ⫾ 15
66 ⫾ 8 81 ⫾ 10 173 ⫾ 6 26.9 ⫾ 2.8 63 ⫾ 14
1 5 37
2 9 32
4 12
5 15
11 22 7 500 650 ⫾ 290 75 ⫾ 18
13 15 5 500 725 ⫾ 320 71 ⫾ 15
72 ⫾ 15
68 ⫾ 14
NOTE. All 86 patients intended to treat are shown. There were no statistically significant differences between the 2 groups. Abbreviations: ACE, angiotensin-converting enzyme; IDDM, insulin-dependent diabetes mellitus; NIDDM, noninsulin-dependent diabetes mellitus.
Alternatively, in patients of the ILM group, an ILM airway (LMAFastrach; LMA-Vertriebs GmbH, Windhagen, Germany) size 4 (for patients below 180 cm body height) or size 5 (patients over 180 cm) was inserted using a 1-handed rotational movement in the sagittal plane with the patient’s head supported by a pillow to achieve a neutral position. The cuff was inflated with air up to the maximum recommended volume (size 4: 30 mL, size 5: 40 mL) or until an effective seal was made. After adequate ventilation was judged by capnography, a lubricated 7.5-mm internal diameter straight, cuffed silicone tube was inserted and passed about 1.5 cm beyond the transverse marker indicating the point where the tip was about to lift the epiglottic elevator bar. If no resistance was felt, the downward movement of the tracheal tube was continued while observing for signs of tracheal placement.8 The correct position of the endotracheal tube was verified using auscultation and capnography and the numbers of attempts (up to 3) and the time until successful intubation were recorded. To determine stress response to these manipulations, arterial blood samples for the measurement of norepinephrine plasma levels were taken immediately before laryngoscopy or insertion of the ILM airway, respectively, and 5, 15, and 30 minutes after successful tracheal intubation. Ten-milliliter plastic lithium-heparin tubes were used for this purpose. Specimens were spun in a centrifuge for 20 minutes, and plasma was separated and stored at ⫺70°C pending analysis. The average heart rate and the mean arterial blood pressure were recorded every 15 seconds and multiplied to determine the pressurerate product (PRP) that was used for further analysis of the clinical stress response induced by intubation. Plasma norepinephrine levels were determined by high-performance liquid chromatography with electrochemical detection (Millipore, Billerica, MA). The lower limit of detection for norepinephrine was 0.018 to 0.024 nmol/L, and the
same-day coefficient of variation for norepinephrine measurements determined by repeated measures of a standardized probe was 3%. A prospective power analysis using analysis of variance for repeated measures (ANOVA with 2 groups, each with 4 measures of norepinephrine levels) with an ␣ of 5% revealed that 80 patients with an estimated between-group standard deviation of 150% of the mean and a within-group standard deviation of 120% of the mean provides a power of ⬎95% to detect an effect size of 0.6 between groups and 0.9 within groups. To detect the group-time interaction (relevant in this analysis) with an estimated effect size of 0.2, a power of 85% was achieved. Plasma norepinephrine levels and the PRP were analyzed by using a ANOVA procedure. Results were called “significant” when the p value was below the critical threshold of 0.05. All other measures (eg, the numbers of attempts to achieve successful intubation) were analyzed descriptively using 95% confidence limits. RESULTS
Eighty-six patients were recruited to participate in the trial. Of these, 1 patient was lost because of cancellation of surgery. The data on hemodynamic stress response of 4 other patients were not analyzed because catecholamines had to be administered during induction of anesthesia (n ⫽ 3) and difficult intubation required more than 3 attempts (n ⫽ 1). Another patient was withdrawn from analysis of norepinephrine levels because of failure of blood probe storage (n ⫽ 1). Finally, 80 patients, 40 in each group, completed the study. There were no statistically significant differences between the 2 groups with regard to any of the biometric variables listed in Table 1. In all patients, intubation via the ILM was performed on the first (n ⫽ 39) or the second attempt (n ⫽ 4). The times until successful tracheal intubation were slightly longer in the ILM group (70 ⫾ 45 seconds) than in the CL group (20 ⫾ 10 seconds). In 5 patients of the CL group and 2 patients of the ILM group, there were signs of cardiac ischemia (defined as ST-T
Fig 1. Plasma norepinephrine levels in the 2 groups. There was a decrease in both groups over time (p < 0.0001) that was more pronounced in the ILM group than in the CL group (p ⴝ 0.031). (Color version of figure is available online.)
STRESS RESPONSE TO TRACHEAL INTUBATION
277
Fig 2. Course of the mean arterial blood pressure and the heart rate in both groups. Note that the time axis is not linear. (Color version of figure is available online.)
change ⬎0.1 mV in any ECG lead). There were no major adverse events during the entire induction period. The clinical outcome after surgery (major complications, time to discharge from intensive care unit, total stay in the hospital, or deaths) was not different between the 2 groups and could not be attributed to events occurring during induction of anesthesia. In both groups, the mean norepinephrine concentrations significantly decreased after induction of anesthesia and after tracheal intubation (p ⬍ 0.0001). This drop was more marked in the patients intubated via the laryngeal mask (p ⫽ 0.031) than in patients receiving direct laryngoscopy. Despite the decreased mean catecholamine concentrations, there were also 10 patients in the ILM group (25%) and 8 in the CL group (20%) who showed increased norepinephrine levels 5 minutes
after successful tracheal intubation compared with baseline values. Ten minutes later (15 after intubation), there were still 5 patients (4 in the CL group and 1 in the ILM group) with elevated norepinephrine levels compared with preinduction values. Results of the mean norepinephrine levels are shown in Figure 1. The values of the PRP showed a similar course over the time with the endocrine markers of stress response (Figs 2-4). There was a gap between the patients of the CL group and the ILM group that starts when the airway of the patients were manipulated. The trend of decreasing PRP remains unchanged in the patients receiving the ILM airway device. However, in patients who received direct laryngoscopy and tracheal intubation, the PRP increases. This different reaction over time is statistically
278
KAHL ET AL
Fig 3. PRP product calculated by multiplying mean arterial pressure with heart rate over time. The gap between the 2 curves that starts with begin of airway manipulation is statistically significant (p ⴝ 0.003). Note that the time axis is not linear. (Color version of figure is available online.)
significant (p ⫽ 0.003). There was no evidence that removal of the ILM airway device has led to a delayed response that has been described in a previous study.6 DISCUSSION
In otherwise healthy patients, the stress response caused by direct laryngoscopy can be attenuated by providing deep anesthesia. In cardiac surgery patients, this approach would lead to a high incidence of hypotension. Thus, in these patients, techniques that provide a reduced stress response can be helpful to avoid problems that may occur with too deep anesthesia as well as with potential detrimental effects of hypertension and tachycardia. The present results show that endotracheal intubation via the ILM is less stressful for patients undergoing coronary artery surgery than intubation using direct laryngoscopy. Furthermore, previous results that tracheal intubation via the intubating laryngeal mask is successful in the great majority of patients could be confirmed.4
An earlier study compared the hemodynamic effects of intubation via an ILM device with direct laryngoscopy in patients induced with 1 mg/kg of propofol, sevoflurane 8% for 3 minutes, and topical anesthesia with 100 mg of aerolized lidocaine.7 The authors found an attenuated pressure response in the ILM group. However, the methodology used in this study has been criticized because the authors performed intubation 5 minutes after insertion of the ILM and this was not representative of how the device is used in clinical practice. A recent trial by Choyce et al6 investigated 61 American Society of Anesthesiologists status 1 to 2 patients undergoing elective surgery. Anesthesia was induced with a sleep dose of propofol, and, after neuromuscular blockade with 0.6 mg 䡠 kg⫺1 of rocuronium, anesthesia was maintained using 2% isoflurane in nitrous oxide/oxygen. In 20 patients, endotracheal intubation was performed by using direct laryngoscopy, and in 41 patients an ILM device was inserted immediately followed by blind tracheal intubation with an 8-mm silicone tube. In half of these
STRESS RESPONSE TO TRACHEAL INTUBATION
Fig 4. online.)
279
Quotient of mean arterial pressure and heart rate (MAP/HR). Note that the time axis is not linear. (Color version of figure is available
patients, the ILM was removed immediately after intubation, whereas in the others removal was performed with a planned delay of 5 minutes. Very similar to the authors’ protocol, blood samples for analysis of norepinephrine were drawn before, 5, and 10 minutes after intubation. In Choyce et al’s trial,6 there were almost identical increases of heart rate (mean rate was about 105 in all groups) and systolic arterial blood pressure. These alterations reflect the increase of plasma levels of norepinephrine because of tracheal intubation that was statistically significant in 2 of the 3 groups. In a recent trial by Kihara et al,9 the hemodynamic response to direct laryngoscopy was compared with an ILM and the Trachlight lightwand device (Laerdal Medical Corp, Wappingers Falls, NY) in a group of normotensive and a group of hypertensive patients. Anesthesia was induced with 2.5 mg/kg of propofol and maintained with 2% sevoflurane in
nitrous oxide. In all groups, heart rate increased compared with preoperative baseline values but without a significant difference between the groups. However, both systolic and diastolic pressure increased after intubation for 2 minutes with highest values in the hypertensive group receiving direct laryngoscopy, and the authors concluded that stress response was more pronounced in this group. Searching for possible reasons for the contradictory results of previous studies6,7,9 regarding the stress response to conventional laryngoscopy and intubation via an ILM airway device, it was interesting to note that in none of the 3 identified trials were opioids used during induction of anesthesia. The study using topical anesthesia before airway manipulation7 found a significantly reduced stress response to blind and fiberopticguided intubation via the ILM compared with conventional endotracheal intubation. By using a comparable induction tech-
280
KAHL ET AL
nique but without using topical anesthesia, Choyce et al6 did not observe differences regarding clinical and humoral stress response. In the authors’ opinion, omission of opioids in patients undergoing tracheal intubation does not reflect everyday clinical practice, especially with high-risk patients undergoing cardiac surgery. The use of norepinephrine rather than other catecholamines as a measure of humoral stress response has been advocated in previous studies10 and was recently confirmed.6 No other trials using catecholamine measurements could be identified. The decision to measure norepinephrine levels immediately before and then 5 minutes after the stressful event of tracheal intubation can be criticized because maximum levels of catecholamines might have been undetected; they should have occurred shortly after tracheal intubation. There was a fairly good correlation between the plasma norepinephrine levels and the PRP ( between 0.41 and 0.55 according to Spearman’s rank correlation procedure) only for the first 2 measurements of norepinephrine. The PRP may be unreliable to determine the risk for myocardial ischemia in high-risk patients because the same value can be obtained with a high arterial pressure and a low heart rate (potential beneficial) and with a low arterial pressure and a high heart rate (potentially deleterious). For this reason, the original values as well as the quotient of mean arterial blood pressure and the heart rate were displayed. Looking at these
figures, it becomes obvious that the differences seen in the PRP are mainly caused by a higher arterial pressure in the patients intubated by the conventional technique. In contrast to this, heart rate was not affected by the technique of intubation. It can be speculated whether preoperative treatment with beta-blockers and premedication with clonidine, 150 g, are responsible for the absence of a parallel raise in heart rate. Induction of anesthesia using midazolam, propofol, and sufentanil administered as a continuous infusion and parallel restoration of intravenous fluid deficits lead to a smooth onset of anesthesia without relevant adverse effects. Conventional laryngoscopy and tracheal intubation were not associated with a clinically relevant stress response in most of these high-risk patients undergoing coronary artery surgery. The use of an ILM airway with subsequent blind tracheal intubation led to significantly decreased cardiovascular (by means of the PRP) and endocrine (by means of plasma norepinephrine levels) response. However, there were still patients in both groups who showed a humoral response to the tracheal intubation, indicating that the intubating laryngeal mask may alleviate stress response but not necessarily prevent it in every patient. It is concluded that the use of the ILM airway device is a useful tool in high-risk patients. The potential benefit might be optimal in patients with highly impaired myocardial reserve because it offers the opportunity to facilitate endotracheal intubation at a lighter level of anesthesia.
REFERENCES 1. Forbes AM, Dally FG: Acute hypertension during induction of anaesthesia and endotracheal intubation in normotensive man. Br J Anaesth 42:618-624, 1970 2. Shribman AJ, Smith G, Achola KJ: Cardiovascular and catecholamine responses to laryngoscopy with and without tracheal intubation. Br J Anaesth 59:295-299, 1987 3. Pernerstorfer T, Krafft P, Fitzgerald RD, et al: Stress response to tracheal intubation: Direct laryngoscopy compared with blind oral intubation. Anaesthesia 50:17-22, 1995 4. Baskett PJ, Parr MJ, Nolan JP: The intubating laryngeal mask. Results of a multicentre trial with experience of 500 cases. Anaesthesia 53:1174-1179, 1998 5. Hickey S, Cameron AE, Asbury AJ: Cardiovascular response to insertion of Brain’s laryngeal mask. Anaesthesia 45:629-633, 1990
6. Choyce A, Avidan MS, Harvey A, et al: The cardiovascular response to insertion of the intubating laryngeal mask airway. Anaesthesia 57:330-333, 2002 7. Joo HS, Rose DK: The intubating laryngeal mask airway with and without fiberoptic guidance. Anesth Analg 88:662-666, 1999 8. Agro F, Brimacombe J, Carassiti M, et al: The intubating laryngeal mask. Clinical appraisal of ventilation and blind tracheal intubation in 110 patients. Anaesthesia 53:1084-1090, 1998 9. Kihara S, Brimacombe J, Yaguchi Y, et al: Hemodynamic responses among three tracheal intubation devices in normotensive and hypertensive patients. Anesth Analg 96:890-895, 2003 10. Russell WJ, Morris RG, Frewin DB, et al: Changes in plasma catecholamines during endotracheal intubation. Br J Anaesth 53:837839, 1981
rate product were analyzed by using an analysis of variance for repeated measures. In both groups, the pressure-rate product (p ⴝ 0.003) and norepinephrine concentrations (p < 0.0001) significantly decreased after induction of anesthesia. However, the fall was more marked in the patients intubated via the laryngeal mask (p ⴝ 0.031) than in patients receiving direct laryngoscopy. There were neither signs of cardiac ischemia nor major adverse events during induction of anesthesia. Conclusions: Reduction of cardiovascular and endocrine stress response associated with endotracheal intubation is more pronounced when performed via the intubating laryngeal mask. Thus, this technique can be helpful in high-risk cardiac patients. © 2004 Elsevier Inc. All rights reserved.
D
All patients received oral premedication with clorazepate, 20 mg, the evening before surgery. On the day of surgery, patients were premedicated with clorazepate, 20 mg, and clonidine, 150 g, about 2 hours before arriving in the operating room. Antihypertensive agents and beta-blockers were continued. After arrival in the operating room, intravenous and radial artery catheters were inserted under topical anesthesia. Five hundred milliliters of hydroxyethyl starch (10%, 200,000 Dalton) and 500 mL of lactated Ringer’s solution were rapidly infused followed by an additional infusion of 500 mL as clinically necessary, mainly to treat hypotension. A 12-lead electrocardiogram (ECG) with automatic ST-segment analysis, oxygen saturation, invasive blood pressure monitoring, and a bispectral index (BIS)-monitoring device were connected to the patients. These variables were continuously recorded online by using a laptop computer connected to a Solar 9500 monitor (General Electric, Munich, Germany). After a 3-minute period of preoxygenation, anesthesia was induced using midazolam (0.05 mg/kg) and a continuous infusion of propofol (10 mg/kg/h) and sufentanil (10 g/kg/h) until loss of consciousness. When the BIS level reached a value of 50, the doses of the anesthetics were reduced (propofol: 3 mg/kg/h, sufentanil: 1.5 g/kg/h) to maintain this level of anesthesia and rocuronium bromide, 0.9 mg/kg, was administered for muscle relaxation after ventilation via a face mask with 100% oxygen was ensured. Two minutes later, conventional laryngoscopy was performed in the CL group using a Macintosh 3 laryngoscope and the trachea was intubated with a 7.5-mm low-pressure high-volume cuffed endotracheal tube (Super Safety; Ru¨sch, Kernen, Germany).
IRECT LARYNGOSCOPY TO facilitate tracheal intubation produces a marked stress response.1 Although these alterations are short lived, they might provoke detrimental effects on the coronary or cerebral circulation of high-risk patients. Direct laryngoscopy resulting in distention of supraglottic tissue is the major cause of stress response caused by conventional endotracheal intubation2 (eg, blind intubation has been shown to produce significantly fewer hemodynamic alterations than intubation after direct laryngoscopy).3 The intubating laryngeal mask (ILM) airway is a new device that facilitates tracheal intubation without laryngoscopy. Because of the high success rate, it is a suitable technique compared with other techniques of endotracheal intubation without direct laryngoscopy.4 The cardiovascular effects of inserting a laryngeal mask airway have been shown to be minimal and are comparable to that of an oropharyngeal device.5 Studies that have previously compared the hemodynamic and endocrine stress responses of endotracheal intubation via an ILM or after direct laryngoscopy showed conflicting results.6,7 Thus, this trial was undertaken to evaluate the stress response of the 2 techniques in patients undergoing coronary artery surgery because these patients are most likely to benefit from a gentle insertion of the endotracheal tube.
KEY WORDS: tracheal intubation, stress response, intubating laryngeal mask airway, cardiac surgery
METHODS After ethics committee approval was obtained, patients gave their written and informed consent before entering the study. Eighty-six male American Society of Anesthesiologists status 3 patients undergoing coronary artery bypass surgery were enrolled into the study, which was performed according to the recommendations of the CONSORT Statement (http://www.consort-statement.org). The patients were randomized immediately before induction of anesthesia for tracheal intubation via the ILM airway (ILM group) or by conventional laryngoscopy (CL group). For this purpose, sealed numbered envelopes were used.
From the *Department of Anesthesiology and Intensive Care Medicine and †Department of Cardiothoracic Surgery, Philipps-University Marburg, Marburg, Germany. Address reprint requests to Leopold H.J. Eberhart, MD, Department of Anesthesiology and Intensive Care, Philipps-University Marburg, Baldingerstr. 1, D–35033 Marburg, Germany. E-mail: eberhart@ mailer.uni-marburg.de © 2004 Elsevier Inc. All rights reserved. 1053-0770/04/1803-0005$30.00/0 doi:10.1053/j.jvca.2004.03.005
Journal of Cardiothoracic and Vascular Anesthesia, Vol 18, No 3 (June), 2004: pp 275-280
275
276
KAHL ET AL
Table 1. Biometric Variables of the Patients of Both Groups
Age Weight Height Body mass index Ejection fraction Number of affected vessels (n ⫽ 1) (n ⫽ 2) (n ⫽ 3) Patients with IDDM NIDDM Patients pretreated with ACE inhibitors Beta-blockers Calcium antagonists Hydroxyethyl starch (mL) Lactated Ringer’s solution (mL) Propofol until tracheal intubation (mg) Sufentanil until tracheal intubation (mg)
ILM Group (n ⫽ 43)
Conventional Laryngoscopy Group (n ⫽ 43)
66 ⫾ 9 84 ⫾ 10 173 ⫾ 5 28.1 ⫾ 3.1 63 ⫾ 15
66 ⫾ 8 81 ⫾ 10 173 ⫾ 6 26.9 ⫾ 2.8 63 ⫾ 14
1 5 37
2 9 32
4 12
5 15
11 22 7 500 650 ⫾ 290 75 ⫾ 18
13 15 5 500 725 ⫾ 320 71 ⫾ 15
72 ⫾ 15
68 ⫾ 14
NOTE. All 86 patients intended to treat are shown. There were no statistically significant differences between the 2 groups. Abbreviations: ACE, angiotensin-converting enzyme; IDDM, insulin-dependent diabetes mellitus; NIDDM, noninsulin-dependent diabetes mellitus.
Alternatively, in patients of the ILM group, an ILM airway (LMAFastrach; LMA-Vertriebs GmbH, Windhagen, Germany) size 4 (for patients below 180 cm body height) or size 5 (patients over 180 cm) was inserted using a 1-handed rotational movement in the sagittal plane with the patient’s head supported by a pillow to achieve a neutral position. The cuff was inflated with air up to the maximum recommended volume (size 4: 30 mL, size 5: 40 mL) or until an effective seal was made. After adequate ventilation was judged by capnography, a lubricated 7.5-mm internal diameter straight, cuffed silicone tube was inserted and passed about 1.5 cm beyond the transverse marker indicating the point where the tip was about to lift the epiglottic elevator bar. If no resistance was felt, the downward movement of the tracheal tube was continued while observing for signs of tracheal placement.8 The correct position of the endotracheal tube was verified using auscultation and capnography and the numbers of attempts (up to 3) and the time until successful intubation were recorded. To determine stress response to these manipulations, arterial blood samples for the measurement of norepinephrine plasma levels were taken immediately before laryngoscopy or insertion of the ILM airway, respectively, and 5, 15, and 30 minutes after successful tracheal intubation. Ten-milliliter plastic lithium-heparin tubes were used for this purpose. Specimens were spun in a centrifuge for 20 minutes, and plasma was separated and stored at ⫺70°C pending analysis. The average heart rate and the mean arterial blood pressure were recorded every 15 seconds and multiplied to determine the pressurerate product (PRP) that was used for further analysis of the clinical stress response induced by intubation. Plasma norepinephrine levels were determined by high-performance liquid chromatography with electrochemical detection (Millipore, Billerica, MA). The lower limit of detection for norepinephrine was 0.018 to 0.024 nmol/L, and the
same-day coefficient of variation for norepinephrine measurements determined by repeated measures of a standardized probe was 3%. A prospective power analysis using analysis of variance for repeated measures (ANOVA with 2 groups, each with 4 measures of norepinephrine levels) with an ␣ of 5% revealed that 80 patients with an estimated between-group standard deviation of 150% of the mean and a within-group standard deviation of 120% of the mean provides a power of ⬎95% to detect an effect size of 0.6 between groups and 0.9 within groups. To detect the group-time interaction (relevant in this analysis) with an estimated effect size of 0.2, a power of 85% was achieved. Plasma norepinephrine levels and the PRP were analyzed by using a ANOVA procedure. Results were called “significant” when the p value was below the critical threshold of 0.05. All other measures (eg, the numbers of attempts to achieve successful intubation) were analyzed descriptively using 95% confidence limits. RESULTS
Eighty-six patients were recruited to participate in the trial. Of these, 1 patient was lost because of cancellation of surgery. The data on hemodynamic stress response of 4 other patients were not analyzed because catecholamines had to be administered during induction of anesthesia (n ⫽ 3) and difficult intubation required more than 3 attempts (n ⫽ 1). Another patient was withdrawn from analysis of norepinephrine levels because of failure of blood probe storage (n ⫽ 1). Finally, 80 patients, 40 in each group, completed the study. There were no statistically significant differences between the 2 groups with regard to any of the biometric variables listed in Table 1. In all patients, intubation via the ILM was performed on the first (n ⫽ 39) or the second attempt (n ⫽ 4). The times until successful tracheal intubation were slightly longer in the ILM group (70 ⫾ 45 seconds) than in the CL group (20 ⫾ 10 seconds). In 5 patients of the CL group and 2 patients of the ILM group, there were signs of cardiac ischemia (defined as ST-T
Fig 1. Plasma norepinephrine levels in the 2 groups. There was a decrease in both groups over time (p < 0.0001) that was more pronounced in the ILM group than in the CL group (p ⴝ 0.031). (Color version of figure is available online.)
STRESS RESPONSE TO TRACHEAL INTUBATION
277
Fig 2. Course of the mean arterial blood pressure and the heart rate in both groups. Note that the time axis is not linear. (Color version of figure is available online.)
change ⬎0.1 mV in any ECG lead). There were no major adverse events during the entire induction period. The clinical outcome after surgery (major complications, time to discharge from intensive care unit, total stay in the hospital, or deaths) was not different between the 2 groups and could not be attributed to events occurring during induction of anesthesia. In both groups, the mean norepinephrine concentrations significantly decreased after induction of anesthesia and after tracheal intubation (p ⬍ 0.0001). This drop was more marked in the patients intubated via the laryngeal mask (p ⫽ 0.031) than in patients receiving direct laryngoscopy. Despite the decreased mean catecholamine concentrations, there were also 10 patients in the ILM group (25%) and 8 in the CL group (20%) who showed increased norepinephrine levels 5 minutes
after successful tracheal intubation compared with baseline values. Ten minutes later (15 after intubation), there were still 5 patients (4 in the CL group and 1 in the ILM group) with elevated norepinephrine levels compared with preinduction values. Results of the mean norepinephrine levels are shown in Figure 1. The values of the PRP showed a similar course over the time with the endocrine markers of stress response (Figs 2-4). There was a gap between the patients of the CL group and the ILM group that starts when the airway of the patients were manipulated. The trend of decreasing PRP remains unchanged in the patients receiving the ILM airway device. However, in patients who received direct laryngoscopy and tracheal intubation, the PRP increases. This different reaction over time is statistically
278
KAHL ET AL
Fig 3. PRP product calculated by multiplying mean arterial pressure with heart rate over time. The gap between the 2 curves that starts with begin of airway manipulation is statistically significant (p ⴝ 0.003). Note that the time axis is not linear. (Color version of figure is available online.)
significant (p ⫽ 0.003). There was no evidence that removal of the ILM airway device has led to a delayed response that has been described in a previous study.6 DISCUSSION
In otherwise healthy patients, the stress response caused by direct laryngoscopy can be attenuated by providing deep anesthesia. In cardiac surgery patients, this approach would lead to a high incidence of hypotension. Thus, in these patients, techniques that provide a reduced stress response can be helpful to avoid problems that may occur with too deep anesthesia as well as with potential detrimental effects of hypertension and tachycardia. The present results show that endotracheal intubation via the ILM is less stressful for patients undergoing coronary artery surgery than intubation using direct laryngoscopy. Furthermore, previous results that tracheal intubation via the intubating laryngeal mask is successful in the great majority of patients could be confirmed.4
An earlier study compared the hemodynamic effects of intubation via an ILM device with direct laryngoscopy in patients induced with 1 mg/kg of propofol, sevoflurane 8% for 3 minutes, and topical anesthesia with 100 mg of aerolized lidocaine.7 The authors found an attenuated pressure response in the ILM group. However, the methodology used in this study has been criticized because the authors performed intubation 5 minutes after insertion of the ILM and this was not representative of how the device is used in clinical practice. A recent trial by Choyce et al6 investigated 61 American Society of Anesthesiologists status 1 to 2 patients undergoing elective surgery. Anesthesia was induced with a sleep dose of propofol, and, after neuromuscular blockade with 0.6 mg 䡠 kg⫺1 of rocuronium, anesthesia was maintained using 2% isoflurane in nitrous oxide/oxygen. In 20 patients, endotracheal intubation was performed by using direct laryngoscopy, and in 41 patients an ILM device was inserted immediately followed by blind tracheal intubation with an 8-mm silicone tube. In half of these
STRESS RESPONSE TO TRACHEAL INTUBATION
Fig 4. online.)
279
Quotient of mean arterial pressure and heart rate (MAP/HR). Note that the time axis is not linear. (Color version of figure is available
patients, the ILM was removed immediately after intubation, whereas in the others removal was performed with a planned delay of 5 minutes. Very similar to the authors’ protocol, blood samples for analysis of norepinephrine were drawn before, 5, and 10 minutes after intubation. In Choyce et al’s trial,6 there were almost identical increases of heart rate (mean rate was about 105 in all groups) and systolic arterial blood pressure. These alterations reflect the increase of plasma levels of norepinephrine because of tracheal intubation that was statistically significant in 2 of the 3 groups. In a recent trial by Kihara et al,9 the hemodynamic response to direct laryngoscopy was compared with an ILM and the Trachlight lightwand device (Laerdal Medical Corp, Wappingers Falls, NY) in a group of normotensive and a group of hypertensive patients. Anesthesia was induced with 2.5 mg/kg of propofol and maintained with 2% sevoflurane in
nitrous oxide. In all groups, heart rate increased compared with preoperative baseline values but without a significant difference between the groups. However, both systolic and diastolic pressure increased after intubation for 2 minutes with highest values in the hypertensive group receiving direct laryngoscopy, and the authors concluded that stress response was more pronounced in this group. Searching for possible reasons for the contradictory results of previous studies6,7,9 regarding the stress response to conventional laryngoscopy and intubation via an ILM airway device, it was interesting to note that in none of the 3 identified trials were opioids used during induction of anesthesia. The study using topical anesthesia before airway manipulation7 found a significantly reduced stress response to blind and fiberopticguided intubation via the ILM compared with conventional endotracheal intubation. By using a comparable induction tech-
280
KAHL ET AL
nique but without using topical anesthesia, Choyce et al6 did not observe differences regarding clinical and humoral stress response. In the authors’ opinion, omission of opioids in patients undergoing tracheal intubation does not reflect everyday clinical practice, especially with high-risk patients undergoing cardiac surgery. The use of norepinephrine rather than other catecholamines as a measure of humoral stress response has been advocated in previous studies10 and was recently confirmed.6 No other trials using catecholamine measurements could be identified. The decision to measure norepinephrine levels immediately before and then 5 minutes after the stressful event of tracheal intubation can be criticized because maximum levels of catecholamines might have been undetected; they should have occurred shortly after tracheal intubation. There was a fairly good correlation between the plasma norepinephrine levels and the PRP ( between 0.41 and 0.55 according to Spearman’s rank correlation procedure) only for the first 2 measurements of norepinephrine. The PRP may be unreliable to determine the risk for myocardial ischemia in high-risk patients because the same value can be obtained with a high arterial pressure and a low heart rate (potential beneficial) and with a low arterial pressure and a high heart rate (potentially deleterious). For this reason, the original values as well as the quotient of mean arterial blood pressure and the heart rate were displayed. Looking at these
figures, it becomes obvious that the differences seen in the PRP are mainly caused by a higher arterial pressure in the patients intubated by the conventional technique. In contrast to this, heart rate was not affected by the technique of intubation. It can be speculated whether preoperative treatment with beta-blockers and premedication with clonidine, 150 g, are responsible for the absence of a parallel raise in heart rate. Induction of anesthesia using midazolam, propofol, and sufentanil administered as a continuous infusion and parallel restoration of intravenous fluid deficits lead to a smooth onset of anesthesia without relevant adverse effects. Conventional laryngoscopy and tracheal intubation were not associated with a clinically relevant stress response in most of these high-risk patients undergoing coronary artery surgery. The use of an ILM airway with subsequent blind tracheal intubation led to significantly decreased cardiovascular (by means of the PRP) and endocrine (by means of plasma norepinephrine levels) response. However, there were still patients in both groups who showed a humoral response to the tracheal intubation, indicating that the intubating laryngeal mask may alleviate stress response but not necessarily prevent it in every patient. It is concluded that the use of the ILM airway device is a useful tool in high-risk patients. The potential benefit might be optimal in patients with highly impaired myocardial reserve because it offers the opportunity to facilitate endotracheal intubation at a lighter level of anesthesia.
REFERENCES 1. Forbes AM, Dally FG: Acute hypertension during induction of anaesthesia and endotracheal intubation in normotensive man. Br J Anaesth 42:618-624, 1970 2. Shribman AJ, Smith G, Achola KJ: Cardiovascular and catecholamine responses to laryngoscopy with and without tracheal intubation. Br J Anaesth 59:295-299, 1987 3. Pernerstorfer T, Krafft P, Fitzgerald RD, et al: Stress response to tracheal intubation: Direct laryngoscopy compared with blind oral intubation. Anaesthesia 50:17-22, 1995 4. Baskett PJ, Parr MJ, Nolan JP: The intubating laryngeal mask. Results of a multicentre trial with experience of 500 cases. Anaesthesia 53:1174-1179, 1998 5. Hickey S, Cameron AE, Asbury AJ: Cardiovascular response to insertion of Brain’s laryngeal mask. Anaesthesia 45:629-633, 1990
6. Choyce A, Avidan MS, Harvey A, et al: The cardiovascular response to insertion of the intubating laryngeal mask airway. Anaesthesia 57:330-333, 2002 7. Joo HS, Rose DK: The intubating laryngeal mask airway with and without fiberoptic guidance. Anesth Analg 88:662-666, 1999 8. Agro F, Brimacombe J, Carassiti M, et al: The intubating laryngeal mask. Clinical appraisal of ventilation and blind tracheal intubation in 110 patients. Anaesthesia 53:1084-1090, 1998 9. Kihara S, Brimacombe J, Yaguchi Y, et al: Hemodynamic responses among three tracheal intubation devices in normotensive and hypertensive patients. Anesth Analg 96:890-895, 2003 10. Russell WJ, Morris RG, Frewin DB, et al: Changes in plasma catecholamines during endotracheal intubation. Br J Anaesth 53:837839, 1981