Anticholinergic agents have been widely utilized. The Role of Anticholinergics in Bronchoscopy* A Randomized Clinical Trial

Transcription

1 The Role of Anticholinergics in Bronchoscopy* A Randomized Clinical Trial Clayton T. Cowl, MD, MS; Udaya B. S. Prakash, MD, FCCP; and Becky R. Kruger, RN Background: Anticholinergic medications have been utilized frequently prior to bronchoscopy and are thought to facilitate the drying of secretions to limit the amount of required topical anesthetic on the airway mucosa, prevent cardiac arrhythmias during the procedure, and increase patient comfort. Objective: To determine if atropine or glycopyrrolate, two anticholinergic agents utilized most frequently in this setting, have any significant role for this purpose. Design: Double-blind, placebo-controlled study, in which patients were randomly selected to receive atropine (0.01 mg/kg body weight, IM injection), glycopyrrolate (0.005 mg/kg, IM injection), or saline solution placebo (approximately 2 ml, IM injection) 15 to 45 min prior to being sedated with midazolam until judged to be lightly sedated. Setting: A large academic teaching hospital in the midwestern United States. Participants: Two hundred seventeen outpatients referred for bronchoscopy who satisfied inclusion and exclusion criteria. Measurements and results: Using a modified visual analog scale (0 to 100 mm), the bronchoscopist and the nurse anesthetist estimated the antisialagogic effect, effectiveness in cough suppression, and overall patient comfort during the procedure. The patients completed a similar questionnaire after recovering from the procedure. Patients were also monitored for complications (cardiac arrhythmias, oxygen desaturation, hypertension, wheezing, or coughing severe enough to curtail the procedure). There was no significant difference found among atropine, glycopyrrolate, and placebo for the primary end point of secretion control. In addition, there was no difference found between either medication and placebo for effectiveness of cough suppression, amount of topical anesthetic used, complication rates, or overall patient comfort. Conclusion: The use of anticholinergic agents prior to bronchoscopy did not affect performance of bronchoscopy or complication rates, and there was no appreciable benefit from the resultant reduction in airway secretions in a population of patients receiving concurrent sedation with benzodiazepines. (CHEST 2000; 118: ) Key words: anticholinergics; antisialagogue; atropine; bronchoscopy; glycopyrrolate; midazolam; premedication Anticholinergic agents have been widely utilized as antisialagogues (to reduce airway secretions) prior to laryngoscopy and bronchoscopy in an effort to improve visibility during the procedures and maximize the efficacy of topical anesthetic agents. 1,2 In addition, anticholinergic agents have been reported to prevent vasovagal phenomena, reflex bronchoconstriction, and bradycardia. 3 7 Glycopyrrolate *From the Division of Pulmonary, Critical Care, and Internal Medicine, Mayo Medical School and Mayo Medical Center, Rochester, MN. Supported by a clinical trial grant from Glaxo Wellcome, Research Triangle Park, NC. Manuscript received August 31, 1999; revision accepted December 9, Correspondence to: Clayton T. Cowl, MD, MS, Pulmonary and Critical Care Medicine, Mayo Medical Center, 200 First St, SW, Rochester, MN 55905; cowl.clayton@mayo.edu and atropine are two such medications administered systemically prior to bronchoscopy. Glycopyrrolate has been claimed to cause less tachycardia in the awake patient 8 and, because of its slower rate of clinical onset, has been reported to maintain a more stable heart rate 9 and less frequent cardiac arrhythmias than atropine. 10 Reports regarding the antisialagogic effectiveness between these two agents have varied considerably. Some authors have noted significantly less secretions with glycopyrrolate administration, 11,12 while others noted no substantial benefit In addition, several other adverse side effects of these anticholinergic agents have been reported, including urinary retention and tachyarrhythmias. Grønnebech et al 16 reported a randomized clinical trial of 45 patients comparing the two agents with placebo and concluded no significant difference 188 Bronchoscopy

2 existed between the effectiveness or side effects of the two drugs, although there was an antisialagogic effect as compared to placebo. Unfortunately, no power calculations were reported in the study, which likely suffered from type I statistical error. More recently, atropine was shown to be of no benefit for drying secretions, producing bronchodilation, or preventing vasovagal reactions when compared to placebo in a randomized clinical trial. 17 Benefits of the routine use of these medications have been questioned, especially in high-risk individuals with prostatic and cardiac disease. 18,19 The efficacy and safety of atropine and glycopyrrolate as premedication prior to bronchoscopy have not been adequately investigated. Therefore, a clinical trial with adequate statistical power to analyze these questions was designed. Materials and Methods A total of 217 patients undergoing elective outpatient fiberoptic bronchoscopy participated in this prospective, placebo-controlled, double-blind study after providing informed written consent and after approval by the Institutional Review Board for ethical research. Patients were randomly assigned to receive atropine, 0.01 mg/kg body weight; glycopyrrolate, mg/kg; or saline solution placebo, approximately 2 ml. All agents were administered IM, 15 to 45 min prior to the procedure. Study subjects were randomized according to the schedule provided (using the SAS function RANUNI; SAS Institute; Cary, NC), 20 which generates random deviates from the uniform distribution on the interval (0, 1). This function implements the algorithm of Fishman and Moore 21 and designated one of three medication types in blocks of nine patients, with three patients assigned to each treatment in each block. Patients were excluded if they were 18 years of age, unable to provide informed consent, were intubated prior to initiation of the procedure, were scheduled to undergo tracheobronchial toilet for removal of copious or inspissated secretions, were agitated and/or confused, had received prior head and neck irradiation, were scheduled for general anesthetic during the procedure, or had previously participated in the study. The posterior oropharynx of each patient was then prepared with topical anesthetic aerosol, and they were then lightly sedated with incremental aliquots of midazolam. Approximately 5 ml of 2% lidocaine solution was instilled over the vocal cords prior to advancing the bronchoscope, and additional lidocaine was administered during the procedure with total amounts recorded. All patients were intubated during the procedure with an 8-mm uncuffed wire spiral endotracheal tube (model 15 W080; Bivonna Medical Technologists; Gary, IN). Supplemental oxygen was administered through an attachment to the endotracheal tube at 2 L/min. Flexible bronchoscopic examinations were performed using a bronchoscope (Olympus P20D; Olympus America; New York, NY). All patients were monitored using continuous pulse oximetry (model M1020A; Hewlett-Packard; Boston, MA), ECG, and noninvasive BP monitoring (model M1008B; Hewlett- Packard). Patients, bronchoscopists, and anesthetists were unaware of the antisialagogue given prior to the procedure, and each party was asked to carefully observe the degree of respiratory tract secretions, the amount of cough experienced during the procedure, and the overall degree of patient comfort. Observations of the antisialagogic effect of the medication and controls were recorded using a modified visual analog scale described by Hatton et al. 19 Visual analog scales were 100 mm in length; respondents placed a vertical line at the point on the scale that best described their observation or experience. More favorable qualities (eg, dry, excellent visualization of the airways) were placed toward the right side of the scale and vice versa. A categorical questionnaire adapted from Rees et al 7 was completed by the bronchoscopists only. Patient questionnaires were completed before being released from the recovery area (approximately 90 min after the procedure), or on the following day. Six different bronchoscopists and five different anesthetists participated in the study after the correct method for completing the questionnaire was thoroughly explained to each. Adverse events noted during and after the procedure were recorded. These included the following: tachycardia or bradycardia (a change of 30 beats/min from baseline or 55 beats/min for bradycardia), hypertension (a rise in systolic or diastolic BP 25 mm Hg), other cardiac arrhythmias, desaturation (arterial oxygen saturation 88% during procedure), wheezing, or coughing that interrupted the procedure. Statistical Analysis Binomial outcomes were compared using the 2 test. Comparisons of ordinal data among the three groups were made using Kruskal-Wallis one-way analysis of variance, and Wilcoxon s signed-rank test was applied to analyze changes within the groups. This study was designed to have 80% statistical power ( ) of detecting a 12% difference between the mean values recorded on the visual analog scale for the primary end point of control of respiratory tract secretions. This difference was estimated to be the minimum required to make the use of these medications clinically generalizable. To achieve this power, approximately 210 patients would have been required for the study, 70 in each group. Secondary end points included cough suppression, complications, and overall patient comfort. Results The characteristics of each patient group were similar in terms of age and gender (Table 1). Equivalent doses of medications were utilized, and there was no appreciable difference in the time required to complete the bronchoscopy for each subgroup of patients. Indications for bronchoscopy and procedures performed during bronchoscopy are given in Table 2. Five bronchoscopies (2.3% of the total) required early termination of the procedure: atropine (n 2), glycopyrrolate (n 1), and placebo (n 1). All of these procedures were associated with severe cough, preventing safe collection of transbronchial biopsy samples. None of the complications required hospital admission (Table 3). No patient developed bradycardia (pulse 55 beats/min). Three patients who received glycopyrrolate developed sinus tachycardia (one patient experienced frequent premature ventricular contractions), as did two patients who re- CHEST / 118 / 1/ JULY,

3 Table 1 Patient Characteristics* Variables n 75 p Value Gender, No. Male Female Age, yr Duration of procedure, min Medications utilized, mean dose Lidocaine, ml Midazolam, mg *Data are presented as mean SD unless otherwise indicated. ceived atropine. No tachycardias were recorded for patients who received placebo treatment. There was no significant difference in antisialagogic effect for either atropine or glycopyrrolate (as compared to placebo) based on the visual analog scores from the bronchoscopists and anesthetists (Table 4). Similarly, the bronchoscopists and anesthetists noted no significant difference in cough suppression and overall patient comfort for either medication as compared to placebo. Based on the Table 2 Summary of Indications for Bronchoscopy and Procedures Performed* Variables n 75 % Indications, No. Nodule or mass Cough Infiltrate Hemoptysis Positive sputum cytology Photodynamic therapy Adenopathy Airway stenosis Bronchopleural fistula Foreign body Volume loss Broncholithiasis Pleural effusion Previous carcinoma Other Procedures performed, No. Biopsy Inspection only Biopsy/brush Brushing TTNA Biopsy/TTNA Biopsy/brush/TTNA Brush/TTNA Biopsy/BAL BAL Inspection prior to stent placement Biopsy/brush/BAL *TTNA transtracheal needle aspiration. patient recollection, the visual analog score was slightly greater for glycopyrrolate (as compared to placebo) for overall patient comfort, but was similar for all other variables. Discussion The use of anticholinergic agents to prevent reflex bronchoconstriction and bradycardia, reduce secretions, and minimize cough has been commonly employed by bronchoscopists throughout North America. 1 As with all IM medication delivery, the use of these agents adds a certain amount of patient discomfort, a small but present overall risk, and, when summed over many patients, a significant increase in cost. Unfortunately, studies designed to examine the safety and efficacy of these drugs are limited by the subjectivity of determining the antisialagogic efficacy of each medication. We used a modified visual analog scale completed by the bronchoscopist, anesthetist, and the patient in an effort to quantitate the inherently subjective task of measuring airway secretions. These scoring systems have been validated to measure qualities such as comfort, intensity of cough, and severity of pain, but not to measure antisialagogic efficacy specifically. Therefore, the bronchoscopists completed an ordinal scale of antisialagogic effectiveness as well, producing similar results (Table 4). While actual sampling of secretions might be possible to develop a more objective measure of efficacy, doing so would be limited in clinical generalizability and undoubtedly suffer from precision errors. In this sample of relatively healthy outpatients, bronchoscopists noted no significant difference among or between atropine, glycopyrrolate, and placebo for secretion control, cough suppression, and overall patient comfort. This supports findings previously reported by several other authors. 5,7,17 Although this study lacked the statistical power to detect small differences ( 12% in visual analog score) between the two medications, that difference would not likely be appreciable in a clinical context. 190 Bronchoscopy

4 Table 3 Complications Recorded During Bronchoscopy* Complications No. No. No. % Tachycardia Hypertension Arrythmia Desaturation Wheezing Severe cough Bradycardia *Total complication rate, 8.0%; rate of complications requiring termination of procedure, 2.3%. The amount of medication required to suppress cough was similar for each anticholinergic agent and not statistically different from placebo (Table 1). This occurred despite the fact that more than four times as many patients were randomized to placebo who were undergoing bronchoscopy to assess anatomic causes of chronic cough (Table 2). Patients recorded higher visual analog scores for glycopyrrolate for secretion control. However, this variable was barely statistically significant (p 0.04) and may have been influenced by the use of IV sedation during the procedure. Interestingly, there was no significant difference in the number of complications experienced by patients receiving either medication when compared to placebo. We recorded no bradycardic episodes in the entire study. Vasovagal events have been reported to occur uncommonly during bronchoscopy, although atropine has been reported to correct it and possibly prevent its occurrence. 22 The number of untoward events were few and likely reflected the generally healthy study population and the use of IV sedation. The use of IV sedation in this study reflects the general clinical practice trend of North American bronchoscopists 1 and may be responsible for improved patient comfort, with less complications associated with bradyarrhythmias and severe cough. The study design may be limited in several respects. Although the visual analog scoring process was explained carefully to each individual participating in the study, having multiple bronchoscopists and anesthetists may have introduced bias in terms of interrater reliability. However, when stratified by individual bronchoscopist, the same trends in visual analog scores emerged. All patients in this study underwent oropharyngeal intubation with an uncuffed wire spiral endotracheal tube, a practice standard at our institution (to assist with rapidity of instrument removal and reinsertion and to better control the airway), but not necessarily in the general bronchoscopy community. Results may have differed using a nasopharyngeal approach, but may have modified the results in favor of anticholinergic medications due to increased vagal response with resulting bradycardia during introduction of the bronchoscope into the nasopharynx. Although medications were to be administered 15 to 45 min prior to the procedure, we did not record the exact amount of time that elapsed between the time of administration to the beginning of the procedure for each patient. Therefore, we could not determine if one subgroup of patients may have received medication too close to the time of starting the bronchoscopy. In summary, we found no significant difference in secretion control between atropine and glycopyrrolate when compared to placebo for bronchoscopy outpatients who received IV sedation. Complications Table 4 Summary of Visual Analogue Scales (0 to 100 mm) Variables n 75 p Value Secretion control (median), mm Bronchoscopist Patient * Effectiveness of cough suppression Bronchoscopist Anesthetist Patient Overall patient comfort Bronchoscopist Anesthetist Patient Bronchoscopist rating of secretion control (ordinal scale), No. Dry (perfect conditions) Moist (acceptable) Wet (unacceptable) *A higher score indicates better secretion control and patient comfort. CHEST / 118 / 1/ JULY,

5 were minimal and similar for all groups. With the largest reported experience examining the use of anticholinergic agents prior to bronchoscopy, we suggest that routine use of these medications be discontinued. ACKNOWLEDGMENT: We wish to thank the nurses of the Mayo Medical Center, Department of Surgical Services, for their assistance in administering the patient visual analog surveys. We also wish to acknowledge Darrell R. Schroeder, Mayo Medical Center, Department of Health Sciences Research, for assistance with statistical analysis. References 1 Prakash UB, Offord KP, Stubbs SE. Bronchoscopy in North America: the ACCP Survey. Chest 1991; 100: Reed AP. Preparation of the patient for awake flexible fiberoptic bronchoscopy. Chest 1992; 101: Simpson FG, Arnold AG, Purvis A, et. al. Postal survey of bronchoscopic practice by physicians in the United Kingdom. Thorax 1986; 41: Zavala DC. Complications following fiberoptic bronchoscopy: the good news and the bad news [editorial]. Chest 1978; 73: Goroszeniuk T, Nicholas IH, Marchant P, et al. Premedication for fiberoptic bronchoscopy: fentanyl, diazepam, and atropine compared with papaveretum and lidocaine. BMJ 1980; 281: Prakash UBS, Stubbs SE. Bronchoscopy: indications and technique. Semin Respir Med 1981; 3: Rees PJ, Hay JG, Webb JR. Premedication for fiberoptic bronchoscopy. Thorax 1983; 38: Mirakhur RK, Jones CJ, Dundee JW. Effects of intravenous administration of glycopyrrolate and atropine in anesthetized patients. Anesthesia 1980; 35: Mcozanitis DA, Dundee JW, Merrett JD, et al. Evaluation of glycopyrrolate and atropine as adjuncts to reversal of nondepolarizing agents in a true-to-life situation. Br J Anaesth 1980; 52: Mmirakhur RK, Clarke RSJ, Elliott J, et al. Atropine and glycopyrronium premedication: a comparison of the effects on cardiac rate and rhythm during induction of anesthesia. Anesthesia 1978; 33: Orko R, Rosenberg PH. Comparison of some postanaesthetic effects of atropine and glycopyrrolate with particular emphasis on urinary problems. Acta Anaesthesiol Scand 1984; 28: Kongsrud F, Sponheim S. A comparison of atropine and glycopyrrolate in anesthetic practice. Acta Anaesthesiol Scand 1982; 26: Mogensen F, Muller D, Valentin N. Glycopyrrolate during ketamine/diazepam anesthesia: a double-blind comparison with atropine. Acta Anaesthesiol Scand 1986; 30: Greeman J, Prasad J. Comparison of the ocular effect of atropine and glycopyrrolate with two intravenous induction agents. Br J Anaesth 1985; 57: Roffe C, Smith MJ, Basran GS. Anticholinergic premedication for fiberoptic bronchoscopy. Monaldi Arch Chest Dis 1994; 49: Grønnebech H, Johansson G, Smedebøl M, et al. Glycopyrrolate vs atropine during anesthesia for laryngoscopy and bronchoscopy. Acta Anaesthesiol Scand 1993; 37: Williams T, Brooks T, Ward C. The role of atropine premedication in fiberoptic bronchoscopy using intravenous midazolam sedation. Chest 1998; 113: Baer GA, Annila PA. Atropine as premedication for anesthesia and bronchoscopy [letter]. Lancet 1995; 345: Hatton MQF, Allen MB, Vathenen AS, et al. Does sedation help in fiberoptic bronchoscopy? BMJ 1994; 309: SAS/STAT, Version 6, 4th ed. Vol 1& 2. Cary, NC: SAS Institute, Fishman GS, Moore LR. A statistical evaluation of multiplicative congruential generators with modulus (2 31 1). J Am Stat Assoc 1982; 77: Zavala DC, Godsey K, Bedell GN. The response to atropine sulfate given by aerosol and intramuscular routes to patients undergoing fiberoptic bronchoscopy. Chest 1981; 79: Bronchoscopy