Summary of findings
Both laryngoscopy and tracheal intubation are associated with a sympathetically mediated increases in blood pressure and heart rate, which may be deleterious in patients with underlying cardiovascular and cerebrovascular diseases (Shribman 1987). These changes are of little consequence in healthy patients (Kovac 1996) but have lead to myocardial ischaemia (reduced blood supply to the muscles of the heart) (Edwards 1994), malignant ventricular arrhythmias (abnormal rhythm), cardiac failure (Fox 1977), intracranial haemorrhage and raised intracranial pressure (Robinson 2001) in vulnerable patients.
Several drugs belonging to different pharmacological classes have been used in either attenuating or obliterating this response. These include narcotics (Crawford 1987; Kay 1985; McAtamney 1998), local anaesthetics (Sun 2009), calcium channel blockers (Atlee 2000) and other peripheral vasodilators, sympathetic blockers (Sharma 1995), and centrally acting adrenergic agonists (Scheinin 1992). Each class has undesirable side effects. The majority of studies do not provide any information regarding outcome in terms of morbidity. Few studies have evaluated the risk of ischaemia and arrhythmias induced by laryngoscopy and intubation (Ali 2009; Mahajan 1993; Puri 1998; Zargar 2002), but no clear cause effect relationship has been established. It still needs to be shown that suppression of this response is of benefit to the patients in terms of outcome.
The primary objective of this review was to determine the effectiveness of pharmacological agents in preventing the morbidity and mortality resulting from the haemodynamic changes in response to laryngoscopy and tracheal intubation in adult patients aged 18 years and above who were undergoing elective surgery in the operating room setting.
Criteria for considering studies for this review
Types of studies
We included prospective randomized controlled trials (RCTs) which studied the effect of different pharmacological agents on the haemodynamic response to laryngoscopy and tracheal intubation in the presence of a control group. We included both blinded and unblinded trials.
Types of participants
We included trials involving adult patients aged 18 years and above, of any race and of either sex, undergoing elective surgery in the operating room setting. We excluded trials of patients undergoing tracheal intubation at other sites such as the intensive care unit (ICU) and accident and emergency departments.
Types of interventions
We included any pharmacological agent given for the specific purpose of attenuating the tracheal intubation associated haemodynamic response. We excluded studies relating to anaesthetic agents which form part of routine anaesthesia for example sedatives, induction agents, inhalational anaesthetics, and muscle relaxants. We grouped the drugs as follows: local anaesthetics; calcium channel blockers; sympathetic blockers (beta blockers, alpha blockers, alpha and beta blockers, and ganglion blockers); peripheral vasodilators; centrally acting adrenergic agonists; angiotensin converting enzyme (ACE) inhibitors; narcotics; and a miscellaneous group.
We included drugs given by different routes that is oral, sublingual, intravenous, tracheal, topical, inhalational. The intravenous route included drugs given by either bolus injection or infusion.
We attempted to answer the following questions.
- Does the administration of either narcotics, local anaesthetics, calcium channel blockers and other peripheral vasodilators, sympathetic blockers, centrally acting adrenergic agonists, and other miscellaneous drugs result in a significant reduction in cardiac, neurological, any other morbidity, or mortality associated with laryngoscopy and tracheal intubation when compared with placebo?
- Is the administration of any of the above pharmacological agent more effective than the placebo in respect to question 1?
- Is the route of administration of the above agents associated with significant differences in respect of the outcomes in question 1?
- Is the administration of above agents associated with adverse haemodynamic respiratory or other unwanted effects?
Types of outcome measures
We included RCTs which mentioned the following outcomes (either primary or secondary) in the methodology, results, or discussion sections.
We considered as the primary outcome major morbidity or mortality related to haemodynamic changes which may follow tracheal intubation. As the majority of these changes settled within five to 15 minutes, only morbidity and mortality occurring during this period or as a direct consequence of these changes were included.
Examples of major morbidity were as follows.
- Electrocardiographic (ECG) evidence of intraoperative ischaemia following intubation subsequently leading to myocardial infarction (evidence of raised creatinine phosphokinase MB fraction or troponin).
- Rupture of an aneurysm.
- Acute cerebral haemorrhage.
- Sudden deterioration in cardiovascular status or neurological status related to the haemodynamic response leading to prolongation of operating room or recovery room stay or admission to a critical care area.
We collected the data as number of patients who had mortality or morbidity.
We considered the following as secondary outcomes.
- Arrhythmias: studies recording any cardiac arrhythmia during or following intubation were included. We collected the data as the number of patients who had arrhythmias during the above period.
- Studies observing ECG evidence of ischaemia. We collected the data as the number of patients who had ECG evidence of ischaemia, ST segment depression on the ECG during or as a consequence of tracheal intubation.
Search methods for identification of studies
We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (2011, Issue 6), MEDLINE (1950 to June 2011), and EMBASE (1980 to June 2011) using the search strategies listed in the appendices (Appendix 1 (CENTRAL); Appendix 2 (MEDLINE); Appendix 3 (EMBASE)).
We reran the searches in December 2012 to identify papers published in 2011 to 2012 and the additional studies identified in this second search are listed in the Characteristics of studies awaiting classification. We will deal with these studies when we update our review.
Both authors independently searched the literature. We did not impose any language restriction. We did not seek unpublished studies.
Searching other resources
We handsearched the reference lists of the RCTs for additional studies.
Data collection and analysis
Selection of studies
We (FK, HU) independently reviewed the titles and the abstracts of the studies and reached a consensus on the abstracts that met the inclusion criteria. In the case of a disagreement regarding the inclusion of an abstract, we obtained the full article. We independently reviewed the full texts of all identified abstracts. We resolved any disagreements through discussion. As no major conflict arose we did not require a third review author. We were not blinded to the names of the authors, the institution, or the journal of publication.
We included RCTs on adult patients undergoing elective surgery, of either sex and any race, aged 18 years and above, who received a pharmacological intervention before or at induction of anaesthesia to prevent the haemodynamic response related to laryngoscopy and tracheal intubation.
We excluded RCTs involving patients in whom tracheal intubation was performed in other settings, for example the intensive care unit (ICU) or accident and emergency unit, or were undergoing emergency surgery. We also excluded trials without control groups. We excluded studies of tracheal intubation without the use of muscle relaxants or which involved the administration of induction agent with a target controlled infusion pump. We also excluded studies that simply mentioned looking for an 'abnormality in ECG' or 'changes in ECG'. This is because these authors did not differentiate between arrhythmias and the ECG changes of ischaemia.
Data extraction and management
We (FK, HU) completed a data extraction form for each study reviewed (Appendix 4). We recorded the following information: the type and number of patients; pharmacological intervention (route, dose, and timing); outcome (primary and secondary, duration of observation, and any adverse effects of the given drug).
One author (HU) entered the data for the meta-analysis into Revman 5.1, which were then checked by the second author (FK).
Assessment of risk of bias in included studies
We (FK, HU) independently evaluated the validity and design characteristics of each selected study. This included: random sequence generation; allocation concealment; blinding of participants and personnel; blinding of outcome assessment; incomplete outcome data; and selective reporting, as recommended in the Cochrane Collaboration's recommended tool for assessing risk of bias (Higgins 2011).
We classified each trial quality component as: 'low risk of bias', 'high risk of bias', or 'unclear risk of bias'. We assessed attrition bias as low if the number of participants recruited in the trial was the same as the number of participants in whom the results were reported, or if an intention-to-treat (ITT) analysis had been performed after attrition. We assessed selective reporting as at low risk of bias if the expected outcome was pre-specified in the methodology section of the study. We resolved disagreements by discussion.
We completed a risk of bias summary table for each study.
Measures of treatment effect
Our outcome data were dichotomous. We used Peto odds ratios (OR) to establish the statistical difference between the intervention and control groups. We interpreted an OR of less than one as showing that the experimental intervention decreased the occurrence of an adverse outcome.
Unit of analysis issues
For any trials using cluster randomization or enrolling more than two arms we used the methods prescribed by the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) and identified and included data from only the intervention groups relevant to our systematic review. Detailed descriptions of only the groups relevant to our review are mentioned in the table Characteristics of included studies. We have mentioned any groups that were not included in our systematic review in the notes section of each study.
Dealing with missing data
Intention-to-treat analysis (ITT) analysis is recommended in order to minimize bias. No consensus exists on how to handle missing data in an ITT analysis in systematic reviews (Higgins 2011). We excluded all participants from the analysis where the outcomes were missing.
Assessment of heterogeneity
Our review was designed to include studies with several different pharmacological interventions for reducing morbidity associated with the haemodynamic response to tracheal intubation. We had specified in the protocol (Khan 2003) that we would combine the results from trials of similar pharmacological class of drugs. For assessing heterogeneity amongst the trials of each pharmacological class, by using the I
- 0% to 30% not important;
- 31% to 60% moderate heterogeneity;
- 61% to 80% substantial heterogeneity;
- 81 % to 100% considerable heterogeneity.
Assessment of reporting biases
We looked at the outcomes listed in the methods section of the trials and compared them with the outcomes reported in the results section. When the outcome was not mentioned in the methods but only reported in the results we listed it as at high risk of reporting bias.
We pooled the data from various trials, where appropriate, using Review Manager 5.1. We chose a fixed-effect model. The outcomes were rare events, therefore we used Peto's OR to estimate the effect size. We expressed the treatment effect as a pooled OR with 95% confidence interval (CI). We plotted the results of each study as point estimates with corresponding 95% CIs. We did not generate plots where only a single study was available for an intervention; these are mentioned in the text description and tallied in Appendix 5.
Subgroup analysis and investigation of heterogeneity
We planned the following subgroup analyses.
- Based on the pharmacological class of drugs that were used for suppressing the haemodynamic response related to tracheal intubation. These groups were: local anaesthetics, calcium channel blockers, sympathetic blockers, peripheral vasodilators, centrally acting alpha agonists, ACE inhibitors, narcotics, and a miscellaneous drugs group.
- Based on trials using the same drug within a single pharmacological group.
- Based on enrolment into the trials of high risk patients who were likely to have a higher risk of outcome events versus low risk patients.
For trials dealing with more than one pharmacological group, we broke them into the review intervention groups (one study group receiving one drug = one intervention). We then combined data from these individual intervention groups with the other studies of the same experimental pharmacological group to make a single group for subgroup analysis.
We assessed heterogeneity among trials by looking at the I
Due to the variability in the patient population, drug doses, and time points of measurements, sensitivity analysis was not applicable to this review.
Description of studies
Results of the search
(See Figure 1)
|Figure 1. Flow diagram.|
Our initial search (up to June 2011) identified 3854 citations across all sources, of which 290 were duplicate citations and 3188 were not relevant randomized trials. We were unable to obtain abstracts for 14 titles and full papers for an additional six titles identified in our initial search. These papers have been included in our Characteristics of studies awaiting classification.
We reran our search in 2012. A further seven studies were identified. These papers have also been placed in the Characteristics of studies awaiting classification and will be dealt with in our next update.
One publication (Fujii 1995) was initially selected but has now been removed to the Studies awaiting classification as the author is being investigated by his institution. We retrieved the full text of 376 trials. We excluded 275 trials because our primary or secondary outcomes were not mentioned in either the methodology, results, or discussion sections (Characteristics of excluded studies). One publication was a duplicate publication of Abdel-Razek 1991 (see Included studies) . We included 72 studies in the review.
The total numbers of patients in the control group for arrhythmias was 993, and the total number for ECG evidence of myocardial ischaemia was 604. The control group in all but four studies received a placebo (normal saline). In the four studies (Kale 1988; Lindgren 1987; Stone 1988; Wang 2003) no drug was administered in the control group.
There were 1939 patients in the intervention group of studies looking at arrhythmias.
There were 1012 patients in the intervention group studying ECG changes of myocardial ischaemia.
Of the included studies, 54 assessed a single drug, 15 assessed two different drugs, and three studies assessed three different drugs.
Route of administration
The intravenous route (IV) was used in 49 studies, the oral route in 13. Of the remaining 10 studies, two used the sublingual route (calcium channel blockers), one used the transdermal route (nitroglycerine), five administered local anaesthetics either by inhalation or transtracheally, and two used both the oral and IV route in different arms of the same study.
Type of patients
In 50 studies, the patients investigated were American Society of Anesthesiologists classification of anaesthesia risk (ASA) 1 and 2 only. Apart from the high risk studies, four studies additionally included ASA 3 and ASA 4 patients in addition to ASA 1 and 2 patients (Inada 1989; Mallon 1990; Miller 1991; Splinter 1989). Further details on status were only available from one study (Miller 1991), where these patients had several additional risk factors for example ischaemic heart disease, diabetes, obesity, etc. In one study the ASA status of the patients was not mentioned (Mikawa 1992a).
We identified 17 studies that investigated only high risk patients. Data from two studies were not included (Davies 1981; Stone 1988). The number of patients enrolled in the studies looking at arrhythmias was 337 in the high risk studies in comparison with 28,36l enrolled in the lower risk studies. There were 395 patients enrolled in the high risk studies that looked at myocardial ischaemia compared to 1306 in the lower risk studies.
We included 72 publications with an adult population aged above 18 years. The sample size varied from 12 (Denlinger 1974) to 548 participants (Miller 1991). Nine studies investigated local anaesthetics, 10 studies calcium channel blockers, 19 studies sympathetic blockers, 10 studies narcotics, five studies peripheral vasodilators, two studies centrally acting alpha agonists, and six studies miscellaneous drugs. The remaining 11 studies investigated drugs belonging to more than one pharmacological class in the same trial. For more details see Characteristics of included studies.
No study was excluded.
Studies awaiting classification
One study (Fujii 1995) with hypertensive patients was available but the lead author is being investigated by his institution for intellectual dishonesty (see Characteristics of studies awaiting classification).
There were 20 additional titles identified in our initial search which may be relevant. Abstracts were available for only six studies and titles only for the remaining 14. Full texts of these publications were not available through our library, inter-library loans, or through the Cochrane Anaesthesia Review Group (CARG). Of these 20 studies, 17 are from non-English journals (see Characteristics of studies awaiting classification).
We reran our search from 2011 to 2012. Seven studies identified in our search rerun have also been added to this section. We will deal with them when we update the review.
Risk of bias in included studies
We evaluated the overall quality of the trials based on the domains described in the following subsections. The various bias domains are presented in the Risk of bias in included studies (see Figure 2; Figure 3).
|Figure 2. Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.|
|Figure 3. Risk of bias summary: review authors' judgements about each risk of bias item for each included study.|
Random sequence generation
All included studies were randomized and mentioned randomization in the methodology, but the actual method used for randomization was stated by only 10 trials (Ali 2009; Kale 1988; Koç 2007; Lee 2011; Memis 2006; Montazeri 2011; Singh 1995; Song 1997; Sun 2009; Victory 1995).
In 63 studies, the authors did not state the method of concealment of allocation (whether the investigators were unable to predict the prospective group to which a participant would be allocated). We categorized these studies as 'unclear'. Concealment was adequate in nine studies (Ali 2009; Kaya 2008; Koç 2007; Lee 2011; Memis 2006; Montazeri 2011; Oxorn 1990; Puri 1998; Sun 2009).
Both participants and outcome assessors were blinded in 53 studies; 16 studies were not blinded. Three studies (Mallon 1990; Oxorn 1990; Stone 1988) used subject experts (cardiologists) for analysis of the ECG changes.
Incomplete outcome data
Withdrawals were specifically mentioned in seven trials (Cucchiara 1986; Ko 1998; Lee 2011; Montazeri 2011; Puri 1998; Song 1997; Wang 2003). In the rest of the studies, the number of participants entering the trials and the number subjected to analysis, as mentioned in the results, were the same. Two studies (Puri 1998; Song 1997) reported six withdrawals each and one study (Wang 2003) reported 10 withdrawals. Two studies (Cucchiara 1986; Montazeri 2011) reported 12 dropouts but no ITT analysis was performed. Lee 2011 reported four patients who did not complete the study and ITT analysis was performed. The reasons for withdrawal were protocol violation in one study (Cucchiara 1986); difficulty in tracheal intubation in four studies (Ko 1998; Lee 2011; Song 1997; Wang 2003); the use of active pharmacological intervention to keep blood pressure stable in two studies (Montazeri 2011; Puri 1998); and refusal to participate in one study (Montazeri 2011).
In order to detect apparent reporting bias, we observed if the authors mentioned an outcome in their objective or methodology section but then failed to report it, or they reported a primary or secondary outcome without indicating that this was intended. We also noted when the authors failed to report on hypertension and tachycardia. Of the included studies, 13 mentioned a relevant outcome in the aims or methodology (Ali 2009; Davies 1981; Jakobsen 1992; Ko 1998; Lindgren 1987; Mallon 1990; Montazeri 2011; Oxorn 1990; Ryhanen 1977; Scheinin 1989; Stone 1988; Wang 2003; Zargar 2002); 59 trials mentioned outcomes only in the results. All of these articles also mentioned the associated changes in blood pressure and heart rate.
Effects of interventions
In total, 60 trials looked for arrhythmias (data from two trials were not included (Jakobsen 1992; Ryhanen 1977)) and enrolled 2932 participants, with 1939 participants allocated to receive an intervention. Of the included studies, 31 trials looked at myocardial ischaemia (data from two trials were not included (Davies 1981; Stone 1988)) and enrolled 1616 participants, with 1012 allocated to receive an intervention.
Only one study reported mortality as an outcome (Miller 1991). The author stated that there were no deaths reported in a multicentre trial of 548 patients who received either placebo or esmolol in a dose of 100 or 200 mg for controlling the haemodynamic response. This trial did not mention either arrhythmias or myocardial ischaemia as an outcome.
2. Major morbidity
Only one study reported morbidity as an outcome (Mikawa 1996). The authors stated in the discussion section that 60 patients, who either received placebo, diltiazem, nicardipine, or verapamil, were followed up for morbidity. No specific morbidity or period of follow-up was identified by the author. This trial additionally mentioned arrhythmias as an outcome.
Arrhthymias only were observed as an outcome in 40 trials, 11 trials observed only myocardial ischaemia as an outcome, and 20 trials observed both arrhythmias and myocardial ischaemia.
Outcome 1. Arrhythmias
1.1 All studies reporting arrhythmias, with subgroup analysis by drug type
This outcome was reported by 60 studies. The analysis from two studies was not included (Jakobsen 1992; Ryhanen 1977). The reason for not including the analysis is given under the respective subheadings.
Of the 993 participants, 134 (13.4%) in the control group experienced an arrhythmia compared to 80 out of 1939 in the combined active group (4.1%). This difference of 9.2% was statistically significant (the odds of an arrhythmia with an active drug compared to control was: OR 0.19, 95% CI 0.14 to 0.26, P < 0.00001, I
The test for a subgroup difference showed a P value of 0.15 (I
A list of all drugs used in these studies reporting arrhythmias, and their dosages, is given in Appendix 6 .
Arrhythmia was reported as an outcome with the use of local anaesthetics in 15 studies (Artru 1985; Asfar 1990; Denlinger 1974; Durrani 2000; Inada 1989; Jakobsen 1992; Kindler 1996; Pouttu 1988; Ryhanen 1977; Singh 1995; Sklar 1992; Splinter 1989; Sun 2009; Victory 1995; Wang 2003). An analysis of 13 of the 15 studies is included in this review. Different doses and different routes of administration of drugs were used. The largest group was lignocaine. Of these studies, 10 investigated lignocaine in different doses with different routes (spray 4%, spray 10%, 100 mg IV, 1 mg kg
Of 362 participants, 29 (8%) in the local anaesthetic group experienced an arrhythmia compared to 32 out of 188 (17.0%) in the control group. This difference of 9.0% was statistically significant (the odds of an arrhythmia with local anaesthetic pretreatment compared to control therapy: OR 0.34, 95% CI 0.18 to 0.62, P = 0.0006, I
We considered subgroup analysis by the individual agents within this group but there were insufficient data to make such a comparison meaningful since different routes and different dosages of drugs were used.
Calcium channel blockers
Arrhythmia was reported as an outcome by 11 studies using calcium channel blockers (Abdel-Razek 1991; Atlee 2000; Kale 1988; Maekawa 1992; Maekawa 1993; Mikawa 1990a; Mikawa 1990b; Mikawa 1992a; Mikawa 1994; Mikawa 1996; Puri 1986). Two of the studies administered verapamil 0.1 mg kg
Of 128 participants, 11 (8.5%) in the control group experienced an arrhythmia compared to three out of 253 (1.18%) in the intervention group. This difference of 7.3% was statistically significant (the odds of an arrhythmia with calcium channel blocker pretreatment compared to control: OR 0.13, 95% CI 0.04 to 0.41, P = 0.0004, I
Arrhythmia as an outcome was reported in 18 studies with sympathetic blockers. The largest group was beta blockers (14 studies). Three studies (Amar 1991; Bernstein 1989; Inada 1989) reported on arrhythmias with combined alpha and beta blockers and one reported on arrhythmias with alpha blockers (Quere 1990).
One study had two interventional arms, one with beta blockers and another with combined alpha and beta blockers (Maharaj 1983).
Of 470 participants, 19 (4%) in the sympathetic blocker group experienced an arrhythmia compared to 34 out of 272 (12.5%) in the control group. This difference of 8.5% was statistically significant (the odds of an arrhythmia with sympathetic blocker pretreatment compared to control therapy: OR 0.22 , 95% CI 0.12 to 0.41, P = 0.00001, I
In addition to the above mentioned studies , Jakobson (Jakobsen 1992) looked at the effect of 100 mg oral metoprolol administered preoperatively, and Ryhanen (Ryhanen 1977) studied the effect of 0.02 mg kg
1.2 Subgroup analysis for sympathetic blockers
Arrhythmias were reported as an outcome by 14 studies (Atlee 2000; Cucchiara 1986; Kindler 1996; Korpinen 1995a; Korpinen 1995b; Magnusson 1986; Maharaj 1983; Mallon 1990; Mikawa 1991a; Oxorn 1990; Sharma 1995; Singh 1995; White 2003; Zargar 2002). Ten studies investigated esmolol in different doses and with different regimens (25 mg IV, 50 mg IV, 100 mg IV, 200 mg IV, 2 µg kg
Of 218 participants in the control group, 31 (14.2%) experienced an arrhythmia compared to 18 out of 371 (4.8%) in the experimental group. This difference of 9.4% was statistically significant (the odds of an arrhythmia with beta blocker pretreatment compared to the control therapy: OR 0.23, 95% CI 0.12 to 0.44, P = 0.0001, I
Combined alpha and beta blockers
Four studies reported arrhythmias as an outcome with labetalol in the following doses: 0.15, 0.25, 0.5, 0.75 mgkg
No study was identified in this subgroup.
Centrally acting alpha agonists
Three studies reported arrhythmia as an outcome using clonidine (0.3 mg oral, 1.25 and 0.625 µg kg
All these studies were blinded and in low risk patients.
Three studies reported arrhythmias (Mahajan 1993; Mikawa 1992b; Singh 1995). Nitroglycerine was used in the following doses: 1.5 µg kg
Arrhythmia was reported as an outcome by 12 studies (Abdel-Razek 1991; Black 1984; Crawford 1987; Dahlgren 1981; Iyer 1988; Kay1987; Ko 1998; Korpinen 1995a; Korpinen 1995b; Lindgren 1987; Scheinin 1989; Splinter 1989). Four studies investigated alfentanil in different doses: 75 µg IV bolus, 0.03 mg kg
Of the 166 participants in the control group, 56 (33.7%) experienced an arrhythmia compared to 29 out of 454 (6.3%) in the intervention group. This difference of 27.4% was statistically significant (the odds of an arrhythmia with pretreatment with a narcotic compared to control therapy: OR 0.12, 95% CI 0.07 to 0.21, P < 0.00001, I
No study was identified for inclusion in this group.
Eight studies reported arrhythmias as an outcome (Ali 2009; Bafna 2011; Kaya 2008; Koç 2007; Memis 2006; Mikawa 1991b; Montazeri 2011; Puri 1998). None of the patients in the control or the treatment groups experienced an arrhythmia.
Six of these studies used oral gabapentin (400, 600, 800, 1000 and 1200 mg) (Ali 2009; Bafna 2011; Kaya 2008; Koç 2007; Memis 2006; Montazeri 2011). Dexamethasone 8 mg IV was used in the second arm of one study (Koç 2007).
1.3 All studies reporting arrhythmias, studies including high risk patients versus low risk patients
To investigate the possible influence of baseline risk in these patients we conducted a subgroup analysis to compare those trials enrolling high risk patients only to trials enrolling low risk patients. In the nine studies where high risk patients were enrolled (Cucchiara 1986; Dahlgren 1981; Denlinger 1974; Inada 1989; Iyer 1988; Kale 1988; Magnusson 1986; Mahajan 1993; Puri 1986), there was a statistically significant difference observed between placebo and the treatment groups for the risk of developing arrhythmias (OR 0.18, 95% CI 0.05 to 0.59, P < 0.005, I
The test for a subgroup difference showed a P value of 0.86 (I
Outcome 2. ECG evidence of myocardial ischaemia
2.1 All studies reporting ECG evidence of myocardial ischaemia, by drug type
This outcome was reported by 31 studies, but we analysed 29 studies only. The reason for not including data from two studies (Davies 1981; Stone 1988) is given under the subgroup 'Sympathetic blockers'. Of 604 participants in the control group, 21 (3.4%) showed evidence of the ECG changes of myocardial ischaemia compared to 10 out of 1012 (0.98%) in the treatment group. This difference of 2.4% was significant (the odds of ECG changes of myocardial ischaemia with an active drug compared to control: OR 0.45, 95% CI 0.22 to 0.92, P = 0.03, I
The test for a subgroup difference showed a P value of 0.17 (I
A list of all the drugs used in these studies and their dosages is given in Appendix 6.
Four studies used lignocaine (1 mg kg
Calcium channel blockers
Two studies reported ischaemic changes on the ECG as an outcome (Kale 1988; Song 1997). None of the 35 participants in the control group or the 95 in the treatment with calcium channel blocker (nifedipine, nicardipine, diltiazem) reported ischaemic changes.
Ischaemic changes in the ECG were reported by 11 studies as an outcome (Cucchiara 1986; De Brujin 1987; Ebert 1989; Girard 1986; Magnusson 1986; Mallon 1990; Newsome 1986; Oxorn 1990; Sharma 1995; Singh 1995; Zargar 2002). Of these studies, 10 had investigated esmolol in different doses. Eight out of 190 (4.2%) of the participants in the control group had ECG evidence of myocardial ischaemia compared to five out of 269 (1.8%) in the treatment group. This difference of 2.4% was not statistically significant (the odds of ECG evidence of myocardial ischaemia with a beta blocker pretreatment compared to control therapy: OR 0.61, 95% CI 0.19 to 1.94, P = 0.41, I
Additionally, one study (Stone 1988) observed the risk of myocardial ischaemia in untreated hypertensive patients who received a single oral dose of labetalol 100 mg, atenolol 50 mg, or oxprenolol 20 mg two hours before induction. We did not include the results of this study in our analysis as the authors reported the overall risk of myocardial ischaemia (seven patients had ischaemia during induction) but did not specify the percentage of patients who had myocardial ischaemia related to tracheal intubation. The anaesthesia technique was also not standardized.
Six of the included studies were with high risk patients (Cucchiara 1986; De Brujin 1987; Ebert 1989; Girard 1986; Magnusson 1986; Newsome 1986) and four of the included studies were not blinded (De Brujin 1987; Girard 1986; Magnusson 1986; Sharma 1995).
Combined alpha and beta blockers
No study was identified in this group.
Four studies reported ECG evidence of ischaemia (Hart 1989; Mahajan 1993; Singh 1995; Thomson 1984). Five participants in the control group out of 39 (12.8%) compared to five out of 44 (11.3%) patients in the treatment group experienced ischaemic changes on the ECG. This difference of 1.5% was not significant (the odds of ischaemic changes on ECG with nitroglycerine pretreatment: OR 1.47, 95% CI 0.26 to 8.18, P = 0.66). Heterogeneity was not applicable.
In addition, one study (Davies 1981) looked at haemodynamic responses related to intubation in 21 patients undergoing neurosurgical procedures. Of these patients 10 were pretreated with hydralazine and the others with saline. The authors mentioned the CM5 lead configuration to detect myocardial ischaemia but did not report these results in their study.
Centrally acting alpha agonists
Only one study reported ischaemic changes on the ECG as an outcome in this group (Montazeri 2011). There were no positive outcomes, therefore we did not perform an analysis.
Six studies (Chraemmer-Jorgensen 1992; Crawford 1987; Dahlgren 1981; Ebert 1989; Lee 2011; Splinter 1989) were identified in this group. Two of the studies were in high risk patients (Dahlgren 1981; Ebert 1989). None of the patients in the treatment group experienced ECG evidence of ischaemia. One patient in the control group experienced ECG evidence of ischaemia. This difference of 0.7% was not significant (the odds of ECG changes of myocardial ischaemia with an active drug compared to control: OR 0.14, 95% CI 0.00 to 6.82, P = 0.32, I
No study was identified in this group.
Six studies were identified in this group (Ali 2009; Kaya 2008; Koç 2007; Memis 2006; Montazeri 2011; Puri 1998). Only one study reported a positive outcome on ischaemic ECG changes (Puri 1998). In this study three participants out of 14 (21.4%) compared to none out of 16 in the magnesium sulphate group showed ECG evidence of ischaemia. Combining all six studies, three of the participants in the control group out of 135 (2.2%) showed evidence of ECG changes of myocardial ischaemia compared to none out of 203 in the treatment group. This difference of 2.2% was significant (the odds of ECG changes of myocardial ischaemia with an active drug compared to control: OR 0.10, 95% CI 0.01 to 1.05, P = 0.05) ( Analysis 2.1).
2.2 All studies reporting ECG evidence of myocardial ischaemia, analysis of studies with high risk patients versus low risk patients
Of the studies, 14 (Cucchiara 1986; Dahlgren 1981; Davies 1981; De Brujin 1987; Ebert 1989; Girard 1986; Hart 1989; Kale 1988; Magnusson 1986; Mahajan 1993; Newsome 1986; Puri 1998; Stone 1988; Thomson 1984) were in high risk patients. To investigate the possible influence of baseline risk we conducted a subgroup analysis of studies in high risk patients compared to low risk patients, with the exclusion of two studies (Davies 1981; Stone 1988). In the 12 studies where the high risk patients were enrolled 15 out of 193 participants in the control group (7.7%) and 10 out of 202 in the treatment group (4.9%) had ECG changes of myocardial ischaemia. This difference was not significant (OR 0.67, 95% CI 0.27 to 1.67, P = 0.39, I
In the 17 studies enrolling lower risk patients none of the 810 patients in the treatment group had myocardial ischaemia compared to six out of 411 (1.4%) in the control group. This difference was significant (OR 0.03, 95% CI 0.01 to 0.18, P = 0.0001, I
The test for a subgroup difference showed a P value of 0.003 (I
Outcome 3. Adverse effects associated with pharmacological interventions, analysis by drug type
Side effects were reported following lignocaine administration in three trials (Pouttu 1988; Splinter 1989; Wang 2003). One patient in the lignocaine group had bradycardia (heart rate (HR) < 37 beats min
A statistically significant difference was observed between the placebo and treatment groups with 16 patients out of 153 showing adverse effects compared to none in the control group (the odds of side effects with the active drug compared to control therapy: OR 12.27, 95 % CI 4.03 to 37.36, P < 0.0001, I
Calcium channel blockers
Hypotension requiring treatment was not observed in any of the studies using calcium channel blockers. In one study (Maekawa 1992) one patient had flushing following a 5 mg nitrendipine oral dose.
Five trials of sympathetic blockers reported side effects. Mallon 1990 reported bradycardia (HR < 50 beats min
A statistically significant difference was observed between the placebo and treatment groups with 37 patients out of 156 showing adverse effects in the treatment group compared to one out of 103 in the control group (the odds of side effects with the active drug compared to control therapy: OR 24.17, 95 % CI 8.88 to 65.79, P < 0.00001, I
Combined alpha and beta blockers
Amar 1991 reported a transient decrease of the mean arterial pressure (MAP) during surgery in two patients who received labetalol 0.75 µg kg
Inada 1989 reported one patient who developed bronchospasm following labetalol 5 mg.
Centrally acting alpha agonists
Intravenous clonidine 1.25 µg kg
Iyer 1988 reported a significant fall in blood pressure (BP) after induction in all patients in the treatment group who received fentanyl (2, 5, 10, and 15 µg kg
Crawford 1987 reported a 40% fall in MAP and bradycardia following a 40 µg kg
In Scheinin 1989 a dose of 75 µg kg
A statistically significant difference was observed between the placebo and treatment groups with 79 patients out of 155 showing adverse effects in the treatment group compared to none out of 63 in the control group (the odds of side effects with the active drug compared to control therapy: OR 149.5, 95 % CI 35.07 to 637.6, P < 0.00001, I
Puri 1998 reported three patients out of 16 who required pharmacological management of hypotension in the group who received magnesium sulphate 50 mg kg
A summary of these results on side effects is presented in Appendix 8.
Route of administration and morbidity
The intravenous route was used in 48 trials. Lignocaine and nitroglycerine were the two drugs where a variety of routes were used. As the dosage of these drugs was not the same in the different trials, we did not perform any further analysis based on route as no meaningful conclusion could be drawn.
High risk patient studies
Seventeen trials were conducted in high risk patients. Different drugs and different dosages were used.
Twelve studies investigated patients with coronary artery disease. The following drugs were tested: lignocaine spray via the tracheal route (Denlinger 1974), IV lignocaine (Inada 1989), sublingual nifedipine (Kale 1988), topical nitroglycerine (Mahajan 1993), IV nitroglycerine (Hart 1989; Thomson 1984), IV esmolol (De Brujin 1987; Girard 1986; Newsome 1986), IV labetalol (Denlinger 1974; Inada 1989), magnesium sulphate (Puri 1998), and IV fentanyl (Iyer 1988). Ebert 1989 investigated the effects of esmolol and fentanyl in ASA 3 and 4 patients. This was a mixed group with hypertension and Ischaemic heart disease.
Two studies investigated hypertensive patients (Magnusson 1986; Stone 1988) and looked at the effects of metoprolol, labetalol, atenolol, and oxprenolol. Dahlgren 1981 included neurosurgical patients in the trial and looked at the effects of IV fentanyl. Davies 1981 recruited patients undergoing intracranial surgery. Cucchiara 1986 studied the effects of IV esmolol in patients undergoing carotid endarterectomy.
Four studies observed participants only for arrhythmias (Dahlgren 1981; Denlinger 1974; Inada 1989; Iyer 1988), eight observed participants only for ischaemia (Davies 1981; De Brujin 1987; Ebert 1989; Girard 1986; Hart 1989; Newsome 1986; Stone 1988; Thomson 1984), and five observed participants for both (Cucchiara 1986; Kale 1988; Magnusson 1986; Mahajan 1993; Puri 1998).
The details of the results of these studies have been summarized in Appendix 9 and Appendix 10. The analysis of studies with high risk patients has been presented in the text. Data from two studies has not been included in the appendices (Davies 1981; Stone 1988) but have been discussed in the text.
Pharmacological agents where only a single study was available
Appendix 5 lists the summary of results where only a single study was available for a particular drug. In addition, single studies were available for hydralazine (Davies 1981), atenolol, and oxprenolol (Stone 1988) but they have not been included in the Appendix.
This review concerns the randomized evidence for the use of pharmacological agents to modify the haemodynamic response to laryngoscopy and tracheal intubation. Specifically, we have collated the data on the clinically relevant outcomes of morbidity and mortality (primary outcomes) and the surrogate measures (secondary outcomes) arrhythmia and ECG evidence of myocardial ischaemia.
Summary of main results
Only two studies reported each of our primary outcomes of mortality and morbidity (Mikawa 1996; Miller 1991). Neither of these outcomes was reported in any arm of any other study, suggesting that these clinical outcomes are probably rare (as we anticipated).
To find out the risk of arrhythmias and myocardial ischaemia, we reviewed 72 studies involving eight different pharmacological groups (local anaesthetics, calcium channel blockers, sympathetic blockers, peripheral vasodilators, centrally acting alpha agonists, narcotics, ACE inhibitors, and a group of miscellaneous drugs). We found data on a total of 32 different drugs.Data from four trials was not included in the analysis. In total, 60 studies reported arrhythmias and 31 reported the ECG changes of myocardial ischaemia as an outcome measure. Our systematic review suggests that the pre-anaesthetic administration of local anaesthetics, calcium channel blockers, beta blockers, or narcotics is associated with a significant decrease in the risk of arrhythmias associated with laryngoscopy and intubation when compared to placebo. Arrhythmias were observed in 134 out of 993 patients in the control group compared to 80 out of 1939 in the intervention group (P < 0.00001).
In total, 31 studies included myocardial ischaemia as an outcome. None of the studies examined the association of intraoperative ECG changes to any clinical outcome such as postoperative myocardial infarction or postoperative persistent changes in ECG. Overall, the use of pharmacological agents resulted in a significant reduction in the risk of ischaemic episodes. Myocardial ischaemia was observed in 21 out of 604 patients in the control group compared to 10 out of 1012 in the treatment group with a statistically significant result (P = 0.01). This result should be treated with caution, however, because in 20 of the included studies no patient in either the control or interventional group suffered this outcome and thus only nine studies contributed to the analysis. Further, the local anaesthetic group was the only subgroup analysis that indicated a significant difference with pretreatment, and that analysis was dominated by a single study (Asfar 1990). The largest subgroup contributing to this data was studies with beta blockers (five studies with positive outcomes) and pretreatment did not show a statistically significant difference compared to placebo. These results could be explained on the basis that myocardial ischaemia was a relatively rare event and the studies were underpowered to look at this complication. Larger datasets may be required to reach meaningful conclusions.
In the subgroup analysis comparing the effects of pharmacological agents on the risk of myocardial ischaemia in high risk patients versus low risk patients, treatment was effective in low risk patients but no effect was observed in high risk patients. This could again be due to the rarity of the events or due to the fact that patients may have been on other medications that may influence the outcome. More studies are needed in the high risk population.
Overall completeness and applicability of evidence
The majority of studies (70%) in our review were conducted in relatively healthy ASA 1 and 2 patients, and therefore there are limitations in extrapolating these results to high risk, vulnerable patients. There were only two studies (Mikawa 1996; Miller 1991) that reported our primary outcome but provided no details. Most of the remaining studies were primarily designed to observe the blood pressure and heart rate changes, and only nine studies addressed the consequence of these changes as part of their objectives or methodology (Hart 1989; Ko 1998; Lee 2011; Lindgren 1987; Mallon 1990; Stone 1988; Sun 2009; Thomson 1984; Wang 2003). We were unable to usefully compare the effectiveness of individual drugs within each pharmacological class because of the variation in dose, timing, and route of administration. This review nevertheless suggests that even these low risk patients are susceptible to cardiac arrhythmia and ECG evidence of myocardial ischaemia as a consequence of haemodynamic changes related to tracheal intubation. Of these low risk patients, 13% had arrhythmias and 1.4 % had myocardial ischaemia when a placebo was given prior to laryngoscopy and intubation. Pre-induction use of pharmacological agents reduced this risk to 4.5% for arrhythmias and nil for myocardial ischaemia.
In clinical practice, the use of agents to reduce the response to laryngoscopy and intubation is likely to be of most importance in those patients at high risk of morbidity due to concurrent cardiovascular disease. In this review, 17 (23%) of the studies were conducted in high risk patients; two enrolled hypertensive patients, 12 enrolled patients with coronary artery disease, two enrolled neurosurgical patients, and one enrolled patients with carotid artery disease. Data from two of these studies were not subjected to analysis (Davies 1981; Stone 1988). It is in this vulnerable population of patients where it is most important to attenuate the haemodynamic response related to tracheal intubation in order to avoid important morbidity. When the high risk studies were pooled (15 studies), pharmacological treatment resulted in a reduction in the risk of arrhythmias but not in ECG evidence of ischaemia. It is difficult to comment on the negative results as, despite pooling these studies, our analysis may still be underpowered to reliably quantify any protective effect. There was also considerable heterogeneity with varying doses of individual agents used in these studies, therefore a particular dose of any agent cannot be recommended.
All the studies in our review only mentioned the nature and the risk of arrhythmia but did not comment on the percentage of participants requiring treatment or any further morbidity related to the rhythm disturbance. Patient anxiety was not taken into consideration and indeed, the premedication given to the participants was variable and may have influenced the outcome in some trials.
Different ECG leads were used for detection of myocardial ischaemia. Some studies used lead 2 for intraoperative ECG monitoring, whereas others did not mention the lead used. Lead 2 is not the best lead to be used for intraoperative ECG monitoring of ischaemia.
Substantial heterogeneity was seen in the results with two of the groups that is local anaesthetics (I² = 79%) and narcotics (I² = 58%) suggesting that drug effect may not be uniform across these groups. On the other hand, the calcium channel blocker group and beta blocker group were fairly homogenous across the studies indicating that different drugs within these classes may exhibit a similar magnitude of effect.
Pharmacological treatment was seen to be effective in both high risk and low risk patients, but substantial heterogeneity was seen in the studies of high risk patients for arrhythmia (I
Despite this clinical heterogeneity, the studies did show a benefit over placebo with some pharmacological agents in decreasing the risk of arrhythmia and future research should concentrate on further evaluation of these drugs.
The above limitations should be kept in mind when applying this evidence in everyday practice
Quality of the evidence
While we only included randomized studies in this review, many of them (particularly those done before year 2000) did not report details of randomization and treatment allocation. We included both blinded and non-blinded studies. Many of the blinded studies did not report the details of the blinding process. Therefore, in general, the methodological quality of the majority of the included studies was poor (Figure 2).
There were many other sources of heterogeneity present in this review. In total, 27 studies used only a small sample size of 10 or fewer participants in one of the study arms. This is likely to result in a lot of random variation. A sample size calculation was again not mentioned in the majority of studies.
Different routes and dosages were used in the same pharmacological group or for the same drug in the different studies. There were also variations in the use of induction agents and the muscle relaxant used for tracheal intubation. The timing of the administration of the same drugs in relation to induction was different in different studies. Factors like duration of laryngoscopy were controlled in some studies but not in all. This type of heterogeneity made it difficult to compare different trials of even the same drug. Because of this heterogeneity we were unable to get an answer to our third review question that is whether the route of administration of the agents was associated with a significant difference in respect of the studied outcomes.
Potential biases in the review process
We attempted to minimize bias by following the guidelines recommended in the Cochrane Handbook for Systematic Reviews of Interventions.The eligibility for inclusion and exclusion and assessment of risk of bias of the different studies was carried out independently by two authors.
Agreements and disagreements with other studies or reviews
We are unaware of other systematic reviews specifically addressing the complications related to the haemodynamic response to laryngoscopy and tracheal intubation. Two systematic reviews addressed the efficacy of esmolol (Figueredo 2001) and landilol ( Ionue 2009) on the haemodynamic changes following laryngoscopy and tracheal intubation but they did not include myocardial ischaemia or arrhythmia as an outcome. Another narrative review (Kovac 1996) reviewed the common causes and treatment of hypertension and tachycardia during laryngoscopy and tracheal intubation but did not include myocardial ischaemia or arrhythmia as an outcome. A French narrative review (Bruder 1992) looked at the consequences and prevention methods of haemodynamic changes during laryngoscopy and intratracheal intubation. We are awaiting translation of this review.
Implications for practice
This review suggests that pre-induction or co-induction administration of local anaesthetics, calcium channel blockers, beta blockers, or narcotics are effective in reducing the risk of arrhythmias associated with laryngoscopy and tracheal intubation. There is no evidence that these agents influence morbidity or mortality in this setting. The evidence does not allow us to determine a recommended dose or route of administration for these agents.
The results of a single study (Asfar 1990) suggest that the administration of lignocaine in this setting also reduces the risk of ECG changes suggestive of myocardial ischaemia. Overall, these studies have a low power to detect a reduction in the risk of such changes and we are unable to confirm any benefit from the use of other agents for this purpose.
Implications for research
There is a deficiency of literature that has looked at morbidity related to the haemodynamic response to laryngoscopy and tracheal intubation and there is need to focus on this aspect in future research. Researchers looking at the blood pressure and heart rate responses should report the consequences of these short lived changes and refer to them as part of their methodology rather than a chance comment in the results. They should also look at short term immediate effects as well as any postoperative effects of these changes. The identified trials were heterogeneous in terms of drug dosages, routes of administration, and timing of intervention of the same drug as well as different study drugs and the anaesthetic drugs used for induction. These need to be standardized to allow pooling of data.
Most literature on the haemodynamic response to tracheal intubation has focused on healthy ASA 1 and 2 patients. In view of the difficulty in undertaking placebo controlled trials in the high risk population, a combination of smaller doses of effective drugs can be compared with doses of single drugs that were found effective in isolated studies.
Randomized controlled trials may not be a good study design to assess the rate of complications for interventions like tracheal intubation. One approach may be to establish prospective registries or a multi-database for a large cohort. Recently a similar approach was taken for looking at complications related to paediatric regional blocks, where a consortium of 14 North American hospitals established the Paediatric Regional Anaesthesia Network (PRAN) to study paediatric epidural data. They recently published their report on the first three years of data (Polaner 2012). Another design approach could be a large scale survey, or well designed national or regional audits. An example of such a survey is NAP 3 in the UK, which was undertaken to study the major complications of central neuraxial blocks (Cook 2009).
We would like to thank: Karen Hovhannisyan (CARG Trials Search Co-ordinator) for designing the search strategy and retrieving articles that we were unable to access, and Jane Cracknell (Managing Editor CARG) for co-ordinating this review. We would also like to thank Mike Bennett (content editor), Nathan Pace (statistical editor), Maurizio Solca, Preethy J Mathew (external peer reviewers) and Janet Wale (Cochrane Consumer Network representative) for their help and editorial advice during the preparation of this systematic review.
We would like to acknowledge Gary Kantor's contribution to the protocol (Khan 2003). We would like to thank Drs Mike Bennett, Pramod Bapat and Filippo Bressan for their help and editorial advice during the preparation of the protocol for the systematic review.
We would also like to acknowledge the input of the following individuals in the preparation of this protocol: Dr Lee Fleisher, Baltimore, USA and Dr Andrew Smith, Lancaster, UK.
Data and analyses
- Top of page
- Summary of findings [Explanations]
- Authors' conclusions
- Data and analyses
- Contributions of authors
- Declarations of interest
- Sources of support
- Differences between protocol and review
- Index terms
Appendix 1. CENTRAL search strategy
Database: CENTRAL, The Cochrane Library, March 2010
#1 MeSH descriptor Laryngoscopy, this term only
#2 MeSH descriptor Intubation, Intratracheal, this term only
#3 laryngoscop*:ti or (intubat* near (tracheal or endotracheal or intratracheal)):ti,ab or (heart rate or blood pressure or arrhythmia* or stroke):ti
#4 (h?emodynamic* near change*):ti,ab or ((cardiovascular or h?emodynamic) near respon*):ti,ab or (hearth near (isch?emia or failure)):ti,ab
#5 (#1 OR #2 OR #3 OR #4)
#6 MeSH descriptor Anesthetics, Local, this term only
#7 MeSH descriptor Analgesics, this term only
#8 (ganglion blocker* or alpha agonist* or agiotensin converting enzyme inhibitor* or adrenergic agonist* or lignocaine or Verapamil or Diltiazem or nisoldipine or nivaldipine or nicardipine or nitrendipine or nifedipine or Esmolol or Pindolol or Labetalol or clonidine or dexmedatomidine or gunabenz or nitroglycerine or Fentanyl or Alfentanil or Sufentanil or Remifentanil or Morphine or Pethidine or Nalbuphine or Tramadol or Butorphenol or Buprenorphine or adenosine triphosphate or magnesium sulphate or gabapentin or dexamethasone):ti
#9 (#6 OR #7 OR #8)
#10 (#5 AND #9)
Appendix 2. MEDLINE search strategy
#1. (laryngoscop* or(intubat*adj6 (tracheal or endotracheal or intratracheal))).mp. or exp Laryngoscopy/ or exp Intubation, Intratracheal/
#2. ((h?emodynamic* adj3 change*) or((cardiovascular or h?emodynamic) adj3 respon*) or heart rate or blood pressure or arrhythmia* or stroke or (hearth adj3 (isch?emia or failure))).mp. or exp Blood Pressure/ or exp Heart Rate/ or Arrythmias, Cardiac/ or Myocardial Ischemia/ or Heart Failure/ or Stroke/
#3. #1 and #2
#4. Exp Anesthetics , Local/ or exp Calcium Channel Blockers/ or exp Adrenergic Agonists/ or exp Adrenergic alpha-Agonists/ or exp Vasodilator Agents/ or exp Analgesics Opioid/ or exp Analgesics/ or (local anesthetic* or calcium channel blocker* or sympathetic blocker* or ((alpha or beta) adj3 blocker*) or ganglion blocker* or alpha agonist* or vasodilator* or narcotic* or opioid* or analgesic* or agiotensin converting enzyme inhibitor* or adrenergic agonist* or lignocaine or Verapamil or Diltiazem or nisoldipine or nivaldipine or nicardipine or nitrendipine or nifedipine or Esmolol or Pindolol or Labetalol or clonidine or dexmedatomidine or gunabenz or nitroglycerine or Fentanyl or Alfentanil or Sufentanil or Remifentanil or Morphine or Pethidine or Nalbuphine or Tramadol or Butorphenol or Buprenorphine or adenosine triphosphate or magnesium sulphate or gabapentin or dexamethasone).mp.
#5. #3 and #4
6. ((randomized controlled trial or controlled clinical trial).pt. or randomized.ab. or placebo.ab. or clinical trials as topic.sh. or randomly.ab. or trial.ti.) and humans.sh.
7. #5 and #6
Appendix 3. EMBASE search strategy
Database: EMBASE via Ovid SP <1980 to 2010 Week 12>
1 exp Laryngoscopy/ or exp Intubation, Intratracheal/ or laryngoscop*.mp. or (intubat* adj6 (tracheal or endotracheal or intratracheal)).mp. (26819)
2 ((h?emodynamic* adj3 change*) or ((cardiovascular or h?emodynamic) adj3 respon*) or heart rate or blood pressure or arrhythmia* or stroke or (hearth adj3 (isch?emia or failure))).mp. (459212)
3 1 and 2 (3227)
4 (local anesthetic* or calcium channel blocker* or sympathetic blocker* or ((alpha or beta) adj3 blocker*) or ganglion blocker* or alpha agonist* or vasodilator* or narcotic* or opioid* or analgesic* or agiotensin converting enzyme inhibitor* or adrenergic agonist* or lignocaine or Verapamil or Diltiazem or nisoldipine or nivaldipine or nicardipine or nitrendipine or nifedipine or Esmolol or Pindolol or Labetalol or clonidine or dexmedatomidine or gunabenz or nitroglycerine or Fentanyl or Alfentanil or Sufentanil or Remifentanil or Morphine or Pethidine or Nalbuphine or Tramadol or Butorphenol or Buprenorphine or adenosine triphosphate or magnesium sulphate or gabapentin or dexamethasone).mp. (493050)
5 exp Calcium Channel Blocking Agent/ or exp Local Anesthetic Agent/ or exp Adrenergic Receptor Stimulating Agent/ or exp Alpha Adrenergic Receptor Stimulating Agent/ or exp Vasodilator Agent/ or exp Narcotic Analgesic Agent/ or exp Analgesic Agent/ (903481)
6 4 or 5 (1075322)
7 6 and 3 (2134)
8 (placebo.sh. or controlled study.ab. or random*.ti,ab. or trial*.ti,ab.) not (animals not (humans and animals)).sh. (745957)
9 7 and 8 (843)
Appendix 4. Study data extraction form
DATA EXTRACTION FORM
QUALITY OF STUDY
Selection bias assessment
Performance bias assessment
Attrition bias assessment
Detection bias assessment
Risk of bias
INDUCTION OF ANAESTEHSIA
ALTERATION TO ANAESTHESIA
Appendix 5. Summary of results where only a single study was available for a particular drug
Appendix 6. Drugs and doses in studies observing arrhythmias as outcome
Appendix 7. Drugs and doses in studies observing myocardial ischaemia as outcome
Appendix 8. Adverse effect of individual drugs
Appendix 9. Studies with high risk patients observing arrhythmias as outcome
Appendix 10. Studies with high risk patients observing myocardial ischaemia as outcome
Contributions of authors
Conceiving the review: Fauzia Khan (FK)
Coordinating the review: FK
Undertaking manual searches: FK, Hameed Ullah (HU)
Screening search results: FK and HU
Organizing retrieval of papers: HU
Screening retrieved papers against inclusion and exclusion criteria: FK and HU
Appraising quality of papers: FK and HU
Abstracting data from papers: FK and HU
Writing to authors for additional information: FK
Data management for review: HU
Entering data into Review Manager: HU and FK
Review Manager statistical data: HU
Statistical analysis: HU
Interpretation of data: HU and FK
Writing the review: FK
Performing previous work that was foundation of present review: FK
Guarantor of the review: FK
Responsible for reading and checking review before submission: FK
Declarations of interest
Fauzia Khan: none known
Hameed Ullah: none known
Sources of support
- Resources from the Department of Anaesthesiology, Aga Khan University, Pakistan.
- No sources of support supplied
Differences between protocol and review
1. Methods. Types of interventions.
ACE inhibitors and calcium channel blockers were added to the pharmacological groups.
2. Methods. Search methods for identification of studies.
a. The search period was extended to June 2011 and rerun in December 2012.
b. LILACS was not searched as it was not accessible to us.
3. Data collection and analysis
a. Studies relating to haemodynamic responses and anaesthetic agents which form part of routine medications for example sedatives, induction agents, inhalational anaesthetics and muscle relaxants were not included
b. We excluded studies of haemodynamic response and tracheal intubation without use of muscle relaxants
c. We excluded studies of haemodynamic response and tracheal intubation which involved administration of induction agent with a controlled infusion pump
4. Criteria for assessing quality of studies
The recent Version 5.1.0. of the Cochrane Handbook for Systematic Reviews of Interventions (Chapter 8) was used for risk assessment, and 'Risk of bias' tables were generated (Higgins 2011).
5. Data collection
Interventions - codes were removed from the review as these were meant for data collection only
Gary Kantor withdrew from the review after publication of the protocol (Khan 2003) and before the actual review was started
Medical Subject Headings (MeSH)
Anesthetics, Local [therapeutic use]; Arrhythmias, Cardiac [etiology; *prevention & control]; Calcium Channel Blockers [therapeutic use]; Hemodynamics [*drug effects; physiology]; Intubation, Intratracheal [*adverse effects; mortality]; Laryngoscopy [*adverse effects; mortality]; Morbidity; Myocardial Ischemia [etiology; *prevention & control]; Narcotics [therapeutic use]; Randomized Controlled Trials as Topic
MeSH check words
* Indicates the major publication for the study