Flexible bronchoscopy (FB) is a routine procedure performed by pulmonologists. This procedure induces different degrees of nerve stimulation, particularly the cough reflex, which may complicate the performance of the examination. Repeated stimulation of cough reflex is not tolerated by patients, who frequently try to expel the bronchoscope from the airways. Therefore, patients often need to repeat the examination, which should also be taken into consideration when planning an endoscopic procedure that does not cause significant discomfort to the patient.
In addition to its low risk, bronchoscopy has a very favorable risk/benefit ratio. The incidence of complications ranges from 0.08% to 9%, and mortality, from 0.01% to 0.1%.1–3 Premedication and the use of topical anesthesia does contribute to these complications. In a study by Credle et al,4 these drugs were responsible for 11 of the 22 significant complications. Although infrequent, complications do exist, and physicians performing the procedure should be able to identify and rectify them immediately.
This procedure is usually performed under topical administration of anesthesia to the airways, with or without sedation. The sedative drugs of choice are opioids and benzodiazepines, either alone or in combination with other drugs. Studies have reported conflicting results related to the beneficial effects of benzodiazepines on patient's tolerance to undergo FB.5–8 Midazolam is the sedative of choice for most physicians performing FB.9 Lidocaine is the anesthetic most frequently used during FB.10
Although a large number of anesthetics have been used in airway surgeries, short-acting anesthetics have played an increasingly important role in the safe performance of procedures and rapid patient recovery.11,12 Such drugs have enabled the performance of outpatient bronchoscopy.10 Rapid regain of alertness, as well as recovery of the airway protective reflexes and muscular contractility are fundamental, especially in patients with respiratory disorders.
The objective of this study was to establish which anesthesia protocol used during FB has the lowest index of complications.
PATIENTS AND METHODS
This is a prospective double-blinded clinical trial comparing 4 different anesthetic regimens for sedation during FB.
The patients included in this study were referred for FB for suspected bronchogenic carcinoma. The diagnostic procedures included endobronchial biopsy and bronchoalveolar lavage. Bronchoscopy time was limited to 15 minutes. The time limit of 15 minutes was established for bronchoscopic procedure to avoid multiple doses administration of anesthetic drugs. The objective of this study was not determining the accuracy of FB in lung cancer diagnosis. It is very important to point out that the time limit of 15 minutes was the only inclusion criteria used for study. Procedures where anticipated duration was more than 15 minutes were excluded from the study (ie, therapeutic procedures, transbronchial biopsy). All procedures were performed by a single bronchoscopist.
Patients included in the study were classified according to the anesthesia risk classification of the American Society of Anesthesiologists (ASA) as class I, II, and III.
Patients were randomly assigned to 4 groups, described below:
- Lidocaine (LID) group: only topical anesthetic (200 mg lidocaine).13
- Propofol (PPF) group: topical anesthetic (200 mg lidocaine) and intravenous (IV) administration of propofol (2.0 mg/kg),12,14 1 mg/kg at induction, and 0.5 mg/kg at 5 and 10 minutes after induction.
- Alfentanil (ALF) group: topical anesthetic (200 mg lidocaine) and IV bolus administration of alfentanil (20 μg/kg).10
- Midazolam (MID) group: topical anesthetic (200 mg lidocaine) and IV administration of midazolam (0.05 mg/kg).10
All patients received 100% oxygen supplementation via mask immediately before the bronchoscopic procedure for 2 minutes, followed by oxygen at 3 L/min throughout the procedure via nasopharyngeal tube previously inserted through the patient's nostril.
Lidocaine gel was applied to the nostril used for the procedure; lidocaine solution was also administered through the cricothyroid membrane puncture.15 Total initial topical anesthetic volume was 20 mL of 1% lidocaine (200 mg) per patient. Additional doses of topical anesthetic were administered when requested by the bronchoscopist to inhibit cough reflex. Total dose did not exceed 5 mg/kg.10 The additional lidocaine dose was included in the evaluation of the performance of each anesthetic modality.
Patients were evaluated according to different variables. Each variable was analyzed independently and patients were assigned a score for each variable, and together they made up a composite score: the greater the score, the greater the magnitude of the event and, consequently, the greater the intensity of the complication. The scoring system was developed according to severity and intensity of side effects observed during the bronchoscopic procedure. The lower score was 0, corresponding to absence of side effects and the maximum score was 6, corresponding to change of initial anesthetic modality (failure of the method). The different grades of the variables were adjusted according to statistical analysis. The composite score and all its components were described according to mean and SD (Table 1). An intergroup comparison was made through analysis of variance 1-way test, under composite scores ranks, with statistically significant differences localization by Duncan post-hoc test. The Kolmogorov-Smirnov did not reject the Gauss curve as approximation model of scores distribution. The approach was adopted according to literature argumentation.16,17 This study was submitted to analysis and epidemiologic specialists of the Universidade de Sao Paulo and Universidade Federal do Rio Grande do Sul.
Demographic data were described as mean and SD for the age and as frequency and percentages for sex.
This study was approved by the Research Ethics Committee of our institution. All the patients included in the series were provided with preinformed consent.
Eighty-nine patients were selected for the study. Nine were excluded: 4 had an ASA IV classification of anesthesiology risk, 3 had a procedure that lasted longer than 15 minutes, 1 patient had an indication of therapeutic bronchoscopy (hemoptysis), and 1 patient had cardiac arrhythmia detected at admission. Eighty patients were randomly assigned to 4 groups: 20 patients in the PPF group, 20 in the ALF group, 20 in the LID group, and 20 in the MID group. The power was 85% for a difference between groups of the 0.5 and a SD of 0.5 in a 2-sided analysis.
Demographic data are shown in Table 2. All patients included in the study were white.
Duration of Bronchoscopic Procedure
Mean duration of bronchoscopic procedure was 12.8±2.8 minutes for the PPF group, 14.2±1.4 minutes for the ALF group, 13.2±2.1 minutes for the LID group, and 14.0±1.9 minutes for the MID group, excluding the duration of interruptions (P=0.128).
Need to Change the Anesthetic Modality Initially Planned
Only 6 cases required the change of the anesthetic modality initially planned: 3 patients in the LID group (2 due to psychomotor agitation and 1 due to cough) and 3 in the MID group (2 due to cough and 1 due to psychomotor agitation).
Table 3 shows the comparison of the variables that made up the composite score for the evaluation of events observed during FB, distributed according to PPF, ALF, LID, and MID groups.
Table 4 shows the results of the composite score in the evaluation of events observed during bronchoscopy according to the different groups.
The comparison of composite scores in the evaluation of events observed during FB indicated that the PPF group had the lowest scores. In comparison with the other groups (LID, MID, and ALF), the quantification and qualification of events were less intense according to the classification system used. The PPF group was followed by the ALF group, which did not show any statistically significant difference from the LID group. The fourth and last group was MID, which did not show any statistically significant difference from the LID group (Fig. 1).
Although we did not identify a statistically significant difference between elderly and younger adult patients, the recommendation of lower propofol dosage requirements in elderly patients18 could be asset to pharmacokinetic rather than pharmacodynamic differences. Therefore, lower induction doses should be kept in use for geriatric patients.18
The usual responses to FB are an increase in cardiac work and in blood pressure, together with episodes of hemoglobin desaturation.19,20
The responses are associated with changes in partial oxygen pressure (PaO2), partial carbon dioxide pressure, and cardiac output. Application of suction through the working channel of the bronchoscope should be restricted to a shorter time interval in these patients to avoid significant changes in the ventilation/perfusion ratio (V/Q). These changes may lead to a substantial increase in the risk of arrhythmia21 and myocardial ischemia during FB.22 Another factor in the genesis of arrhythmia in FB is the vagal response secondary to the insertion of the bronchoscope in the airway, which may determine the occurrence of bradycardia.
This study did not find statistically significant differences in the occurrence of arrhythmias between the study groups (P=0.376).
Bradycardia may occur immediately after the IV administration of alfentanil.23 Although the use of propofol has been associated with bradycardia, this study did not observe bradycardia in the group of patients that received the drug. The mechanism by which propofol causes the bradycardia has not been well established. Ozturk et al24 reported that the occurrence of arrhythmia was significantly greater in the midazolam group than in the propofol group in their study.
The score found in our study did not show any significant differences in the occurrence of hypoxemia between the study groups (P=0.117). All FB determines some degree of deleterious effects on respiratory physiology.10 The usual and expected response in bronchoscopy is a decrease in hemoglobin oxygen saturation (SaO2).19,20 Schnapf25 reported a decrease in SaO2 to below 5% of baseline in 80% of the cases studied, particularly when the bronchoscope was placed in the middle of the trachea.
The causes of hypoxemia have been assigned to the procedure itself19,26 or to respiratory depression secondary to the use of sedation.27
Matsushima et al28 and Neuhaus et al29 showed that patients had a decrease in forced expiratory volume in 1 second and an increase in the residual functional capacity during the performance of FB.
Albertini et al30 reported that the decrease of PaO2 secondary to FB may be avoided in many patients by the use of a Venturi mask at 40% during and after the procedure. Different techniques may be used to ventilate the patient or to provide oxygen during the performance of rigid or FB: spontaneous respiration, intermittent positive pressure ventilation, use of the Sanders injection system, and ventilation with high-frequency jets.31 Antonelli et al32 studied the use of noninvasive positive pressure ventilation via facemask in immunosuppressed patients with abnormal gas exchange and recommended its use during FB.
The use of low or moderate doses of sedation with midazolam also did not change the probability of a decrease in SaO2 in the study conducted by Jones and O'Driscoll.33
Cartwright et al34 showed that alfentanil reduced acute ventilatory response to hypoxia.
The PPF and ALF groups in our study had significantly lower scores for the variable “cough” than the LID and MID groups (P=0.003).
Cough is a complex reflex that initiates with the stimulation of receptors in the respiratory system. These receptors may be sensible to touch and mobilization (mechanic receptors) or primarily sensible to toxic gases (chemical receptors).35 The occurrence of cough during bronchoscopy is the rule. The direct stimulation of mechanical receptors triggers the cough reflex.
Different techniques are used for the application of local anesthetics in the airways, and their purpose is to inhibit the laryngeal and cough reflexes.10
The analgesic and cough suppressant properties of opioids are well known.23
Greig et al9 demonstrated that alfentanil is a more effective antitussive agent than midazolam for outpatient FB.
The absorption of local anesthesia through the oropharyngeal mucosa is relatively low, but its absorption through the lower airways may be quite significant. The estimation of the amount of local anesthesia that is absorbed after its administration in the airway may be inaccurate. The maximal dose recommended for the administration of topical lidocaine is 5 to 8 mg/kg.10
Complications caused by the use of topical anesthesia are frequently the result of a large dose of the drug.33,36,37 Signs and symptoms of neuroexcitation, such as dizziness and audiovisual disturbances, are indications of intoxication by lidocaine and usually precede more serious complications, such as seizures, coma, and cardiovascular collapse.10
No significant differences were found in psychomotor agitation between the MID and ALF groups, but a significant difference was found between LED and PPF, LED and ALF, LED and MID, MID and PPF, and PPF and ALF (P<0.001).
Clarkson et al38 reported that propofol has a sedative effect similar to that of midazolam, but its action was more rapid both in the onset of action and in the return to initial consciousness levels. Although respiratory depression may be a frequent complication in bronchoscopy,10 it was not significant in our study groups. Although benzodiazepines cause less respiratory depression than opioids,38,39 no significant difference was found between the different groups in our study (P=0.138). Peacock et al40 found that local anesthesia may be responsible for prolonged respiratory depression in patients who undergo bronchoscopy. High doses of alfentanil or midazolam may produce respiratory depression and apnea, although the same may occur with lower doses.22
The combination of amnesia and antianxiety effect of benzodiazepines with the analgesic and cough suppressant effect of opioids may be beneficial, although their synergetic action increases the risk of cardiovascular and respiratory complications.2 Randell and Lindgren41 found a decrease in respiratory rate with the use of fentanyl (1 mcg/kg) and diazepam (0.05 mg/kg) in the sedation of patients who underwent FB.
No case of bronchospasms or laryngospasms was observed in any of the cases included in this study.
No significant differences in the occurrence of arterial hypotension were found between any of our groups (P=0.287). McRae10 demonstrated that anesthetic induction with propofol provides hemodynamic stability in patients who undergo bronchoscopy. Yamaguchi et al14 demonstrated that the association of propofol and fentanyl may prevent the occurrence of hemodynamic instability during bronchoscopy. According to Matot et al,42 the association of propofol and alfentanil reduces the occurrence of arterial hypotension but does not prevent myocardial ischemia.
Uetsuki et al43 did not find significant differences in arterial blood pressure, heart rate, respiratory rate, or levels of SaO2 in the comparison of patient groups of different ages who underwent anesthesia with propofol for bronchoscopy.
We also did not find any significant differences in the occurrence of nausea or vomiting between our study groups (P=0.106).
Although the role of propofol in the management of postoperative nausea and vomiting is not well established, Gan et al44 reported that propofol was effective in their management.
No statistically significant differences were found between the PPF and LID groups. These groups, however, had significantly lower scores for the variable “consciousness level” than the ALF and MID groups. The scores for the ALF group were significantly lower than those for the MID group (P<0.001). Clarkson et al38 conducted a comparative study of propofol (mean induction dose=104.7 mg; mean maintenance dose=121.9 mg) and midazolam (mean induction dose=9.3 mg; mean maintenance dose=3.7 mg). Consciousness level was more rapidly recovered with the use of propofol (2.3 vs.6.3 min; P<0.01). Crawford et al45 also showed that the recovery after sedation was faster in patients who received propofol during FB than in patients who received midazolam.
Hypnotic sedative drugs not only increase sedation, but also decrease the level of memory as their serum concentration increases.46
The administration of alfentanil in the prescribed dose does not provide, according to some authors,22,47 adequate levels of sedation.
Our results showed that in only 6 cases was it necessary to change the anesthetic method initially planned, and the causes were discussed above in the analysis of the variables cough and psychomotor agitation.
The low incidence of collateral effects and accidents associated with anesthesia, together with the reduced hemodynamic depression and a short recovery time, also made Mesiti et al48 recommend the use of propofol to anesthetize patients who undergo FB.
Randell49 reported that the use of propofol in addition to topical lidocaine provides better conditions for the performance of FB than other methods.
Despite the number of the collateral effects to have been small, it was possible to detect significant differences between the studied groups. This means that propofol demonstrated lower intensity of adverse effects than other tested drugs in accordance with the evaluation system adopted.
A comparison of composite scores was made for the evaluation of inconvenient events with usual intravenous anesthetic drugs for FB. Four groups of patients were individualized: with only topical lidocaine employed (I) and topical lidocaine plus propofol (II), fentanil (III) and midazolan (IV), each drug delivered as the sole intravenous anesthetic prescription. The group treated with intravenous propofol (II) had the lowest inconvenience scores followed by intravenous alfentanil (III). But alfentanil (III) and midazolan (IV) groups did not show statistically significant differences from patients taking only topical lidocaine (I) but did to the propofol one (II). This study is an attempt to qualify available intravenous anesthetic drugs toward its appropriate use in sedation for patient comfort and safety during FB procedures.
1. Pue CA, Pacht ER. Complications of fiberoptic bronchoscopy
at a university hospital. Chest. 1995;107:430–432.
2. Weiss SM, Hert RC, Gianola JG, et al. Complications of fiberoptic bronchoscopy
in thrombocytopenic patients. Chest. 1993;104:1025–1028.
3. Slonim AD, Ognibene FP. Amnestic agents in pediatric bronchoscopy
. Chest. 1999;116:1802–1808.
4. Credle WF, Smiddy JF, Elliott RC. Complications of fiberoptic bronchoscopy
. Am Rev Respir Dis. 1974;109:67–72.
5. Maltais F, Laberge F, Laviolette M. A randomized double-blind, placebo-controlled study of lorazepam as premedication for bronchoscopy
. Chest. 1996;109:1195–1198.
6. Williams TJ, Nicoulet J, Coleman E, et al. Safety and patient acceptability of intravenous midazolam
for fibreoptic bronchoscopy
. Respir Med. 1994;88:305–307.
7. Colt HG, Morris JF. Fiberoptic bronchoscopy
without premedication: a retrospective study. Chest. 1990;98:1327–1330.
8. Hatton MQF, Allen MB, Vathenen AS, et al. Does sedation help in fibreoptic bronchoscopy
? BMJ. 1994;309:1206–1207.
9. Greig JH, Cooper SM, Kasimbazi HJ, et al. Sedation for fibre optic bronchoscopy
. Respir Med. 1995;89:53–56.
10. McRae K. Anesthesia for airway surgery. Anesthesiol Clin North Am. 2001;19:497–541.
11. Barr J, Egan TD, Sandoval NF, et al. Propofol
dosing regimens for ICU sedation based upon an integrated pharmacokinetic-pharmacodynamic model. Anesthesiology. 2001;95:324–333.
12. Schaeuble J, Heidegger T, Gerig HJ, et al. Comparison of etomidate and propofol
for fibreoptic intubation as part of an airway management algorithm: a prospective, randomizes, double-blind study. Eur J Anaesthesiol. 2005;22:762–767.
13. Amitai Y, Zylber-Katz E, Avital A, et al. Serum lidocaine
concentrations in children during bronchoscopy
with topical anesthesia. Chest. 1990;98:1370–1373.
14. Yamaguchi S, Koguchi T, Midorikawa Y, et al. Comparative evaluation of TIVA with propofol
-fentanyl and thiopental-sevoflurane anesthesia using laryngeal mask airway for diagnostic bronchoscopy
. J Anesth. 1998;12:53–56.
15. Negrin MP, Martin JJB, Fernández AO, et al. Subjective tolerance to flexible bronchoscopy
. J Bronchol. 2001;8:166–169.
16. Campbell MJ, Machin D. Medical Statistics: A Commonsense Approach. 2nd ed. New York, Chichester: John Wiley & Sons; 1993.
17. Snedecor GW, Cochran WG. Statistical Methods. 7th ed. Ames: The Iowa State University Press; 1980.
18. Kanto J, Gepts E. Pharmacokinetic implications for the clinical use of propofol
. Clin Pharmacokinet. 1989;17:308–326.
19. Dubrawsky C, Awe RJ, Jenkins DE. The effect of bronchofiberscopic examination on oxygenation status. Chest. 1975;67:137–140.
20. Lundgren R, Haggmark S, Reiz S. Hemodynamic effects of flexible fiberoptic bronchoscopy
performed under topical anesthesia. Chest. 1982;82:295–299.
21. Katz AS, Michelson EL, Stawicki J, et al. Cardiac arrhythmias: frequency during fiberoptic bronchoscopy
and correlation with hypoxemia. Arch Intern Med. 1981;141:603–606.
22. Matot I, Kramer MR, Glantz L, et al. Myocardial ischemia in sedated patients undergoing fiberoptic bronchoscopy
. Chest. 1997;112:1454–1458.
23. Hardman JG, Limbird LE. Goodman & Gilman's The Pharmacological Basis of Therapeutics. New York: McGraw-Hill; 2001.
24. Ozturk T, Cakan A, Gulerce G, et al. Effect of haemodynamics and FEV1
on dysrhythmia in sedated patients undergoing fiberoptic bronchoscopy
. Eur J Anaesthesiol. 2001;18(suppl 21):39, A-135.
25. Schnapf BM. Oxygen desaturation during fiberoptic bronchoscopy
in pediatric patients. Chest. 1991;99:591–594.
26. Salisbury BG, Metzer LF, Altose MD, et al. Effect of fiberoptic bronchoscopy
on respiratory performance in patients with chronic airways obstruction. Thorax. 1975;30:441–446.
27. Webb AR, Doherty JF, Chesser MR, et al. Sedation for fibreoptic bronchoscopy
: comparison of alfentanil
, papaveretun, and diazepam. Respir Med. 1989;83:213–217.
28. Matsushima Y, Jones RL, King EG, et al. Alterations in pulmonary mechanics and gas exchange during routine fiberoptic bronchoscopy
. Chest. 1984;86:184.
29. Neuhaus A, Markowitz D, Rotman HH, et al. The effects of fiberoptic bronchoscopy
with and without atropine premedication on pulmonary function in humans. Ann Thorac Surg. 1978;25:393–398.
30. Albertini RE, Harrell JH, Moser KM. Management of arterial hypoxemia induced by fiberoptic bronchoscopy
. Chest. 1975;67:134–136.
31. Hautmann H, Bauer M, Pfeifer KJ, et al. Flexible bronchoscopy
: a safe method for metal stent implantation in bronchial disease. Ann Thorac Surg. 2000;69:398–401.
32. Antonelli M, Conti G, Riccioni L, et al. Noninvasive positive-pressure ventilation via face mask during bronchoscopy
with BAL in high-risk hypoxemic patients. Chest. 1996;110:724–728.
33. Jones AM, O'Driscoll R. Do all patients require supplemental oxygen during flexible bronchoscopy
? Chest. 2000;119:1906–1909.
34. Cartwright CR, Henson LC, Ward DS. Effects of alfentanil
on the ventilatory response to sustained hypoxia. Anesthesiology. 1998;89:612–619.
35. Dicpinigaitis PV, Rauf K. The influence of gender on cough reflex sensitivity. Chest. 1998;113:1319–1321.
36. Pereira W, Kovnat DM, Smider GL. A prospective cooperative study of complications following flexible fiberoptic bronchoscopy
. Chest. 1978;73:813.
37. Suratt PM, Smiddy JF, Gruber B. Deaths and complications associated with fiberoptic bronchoscopy
. Chest. 1976;69:747.
38. Clarkson K, Power CK, O'Connell F, et al. A comparative evaluation of propofol
as sedative agents in fiberoptic bronchoscopy
. Chest. 1993;104:1029–1031.
39. Dawidowicz AL, Formal E, Mardarowicz M, et al. The role of human lungs in the biotransformation of propofol
. Anesthesiology. 2000;93:992–997.
40. Peacock AJ, Benson-Mitchell R, Godfrey R. Effect of fiberoptic bronchoscopy
on pulmonary function. Thorax. 1990;45:38–41.
41. Randell T, Lindgren L. Propofol
or combination of fentanyl with diazepam for sedation for fiberoptic bronchoscopy
. Acta Anaesthesiol Scand. 1991;35(suppl):188, O45.
42. Matot I, Drenger S, Kremer M, et al. Incidence of myocardial ischemia during sedation for fiberoptic bronchoscopy
. Anesthesiology. 1996;85(suppl):3A;A-52.
43. Uetsuki N, Murakawa M, Adachi T, et al. Propofol
sedation for fiberoptic bronchoscopy
in intensive care unit. Int Congr Ser. 1998;1164:171–172.
44. Gan TJ, Molem HE, Ray J, et al. Patient-controlled antiemesis. A randomized, double-blind comparison of two doses of propofol
versus placebo. Anesthesiology. 1999;90:1564–1570.
45. Crawford M, Pollock J, Anderson K, et al. Comparison of midazolam
for sedation in outpatient bronchoscopy
. Br J Anaesth. 1993;70:419–422.
46. Veselis RA, Reinsel RA, Feshchenko VA. Drug-induced amnesia is a separate phenomenon from sedation: electrophysiologic evidence. Anesthesiology. 2001;95:896–907.
47. Wood-Baker R, Burdon J, McGregor A, et al. Fibre-optic bronchoscopy
in adults: a position paper of the Thoracic Society of Australia and New Zealand. Intern Med J. 2001;31:479–487.
48. Mesiti T, Belli T, Colini BG, et al. Propofol
as sole anaesthetic agent in trachobronchial endoscopic laser therapy. Proceedings of the 43rd SIAARTI Congress; vol 1, 1989 June 8 to 12; Modena, Italy. Modena: SIAARTI; 1989:51–54.
49. Randell T. Sedation for bronchofiberoscopy: comparison between propofol
infusion and intravenous boluses of fentanyl and diazepam. Acta Anaesthesiol Scand. 1992;36:221–225.