Rapacuronium, with a rapid onset of peak effect and a short recovery time, was introduced into clinical practice as a potential alternative to succinylcholine. It provides comparable intubating conditions without the associated side effects of hyperkalemia, myalgia, and the risk of triggering malignant hyperthermia (1–4).
Rapacuronium became available for clinical use at The Children’s Hospital of Philadelphia on March 1, 2000. Several severe cases of bronchospasm presented within the first few months of clinical use that appeared to be temporally related to the administration of rapacuronium prompted two related studies. All of these cases were reported to the Food and Drug Administration (FDA), and rapacuronium was withdrawn from the hospital’s formulary on July 31, 2000 pending results of the studies. First, a retrospective cohort study was performed to determine which risk factors, if any, existed for the development of bronchospasm after the administration of rapacuronium. Second, a case-control study was designed to answer the question, “Are children with bronchospasm during induction of anesthesia more likely to have received rapacuronium compared with other muscle relaxants?”
The commercially available computerized anesthesia record system (CompuRecord®; Philips Medical Systems, Andover, MA) at our institution is an integrated information system that consists of three linked databases—a preanesthetic evaluation, an intraoperative anesthesia record, and a continuous quality improvement (CQI) section. Information is manually entered into the preanesthetic evaluation database during the preoperative visit for elective surgical procedures. The system is configured to permit entries as simple “yes” or “no” checkbox options and space is available for additional comments. No default values are available, and a blank space will appear if data are not entered in the checkbox. Selected information from the preanesthetic evaluation is transferred into the intraoperative anesthesia record. The CompuRecord® captures and stores data from the various monitoring devices and permits manual entry of additional information in user-selected checkboxes along with standard or narrative comments. Preselected configurations of these data can be printed as a hard-copy permanent medical record of intraoperative events. The CQI section consists of checkboxes for voluntarily reported categories of perioperative occurrences with space for an accompanying narrative description of the event.
The timeframe for both the retrospective cohort and the case-control studies was limited to the period during which rapacuronium was available for clinical use (March 1, 2000 to July 31, 2000). For the retrospective cohort study, our intraoperative anesthesia record database was queried using the research module of the CompuRecord® system to identify all cases in which rapacuronium was administered to patients 0–18 yr old. Two independent reviewers inspected these records to determine which patients met the criteria for a diagnosis of bronchospasm during induction of anesthesia. Criteria for the diagnosis of bronchospasm during induction included either a statement in the narrative of the anesthesia record that the patient had developed bronchospasm or documentation of the use of a bronchodilator within 15 min after induction of anesthesia. The following data were abstracted from each anesthesia record: age, weight, sex, ASA physical status, history of reactive airways disease (RAD), history of drug allergy, rapacuronium dose (mg/kg), mode of induction of anesthesia (rapid or modified rapid sequence induction [RSI] versus routine inhaled or IV induction), IV induction drug (propofol versus thiopental), and use of morphine. Patients were considered to have prior RAD only if their ASA physical status was II or more and if at least one of the following items were documented in the preanesthetic evaluation or intraoperative anesthesia record: a) RAD was listed as a concomitant diagnosis or b) the patient was receiving at least one medication used for the treatment of RAD (e.g., inhaled or oral bronchodilators, inhaled steroids or cromolyn sodium).
In addition to the cases of bronchospasm during induction of anesthesia with rapacuronium identified in the retrospective cohort study, cases of bronchospasm during induction of anesthesia with other muscle relaxants were identified for the case-control study. First our intraoperative anesthesia record was queried using the research module for the standard comment “albuterol aerosol treatment.” The CQI database was queried to find all cases of voluntarily reported bronchospasm during the study time frame. Two independent reviewers applied the same criteria as in the cohort study for identification of bronchospasm during induction of anesthesia. Four controls were identified for each case of bronchospasm. The controls were randomly selected from among the patients undergoing procedures on the same day as each case by matching the last digit of medical record number with a table of random digits. Patients were excluded from the Control group if an oral or nasal tracheal tube or tracheostomy tube was present before induction of anesthesia, if a muscle relaxant was not administered, or if the patient was >18 yr of age. Data abstracted from each anesthesia record were the same as those for the retrospective cohort study with the addition of the specific muscle relaxant administered. The criteria for determining a history of RAD were the same as that used for the retrospective cohort study.
The statistical approach was similar for both the retrospective cohort and the case-control studies. Univariate analyses were conducted using an unpaired Student’s t-test for parametric variables. Fisher’s exact test and relative risk (RR) (cohort study) or odds ratio (OR) (case-control study) calculated for categorical variables. The 95% CIs were calculated for both the RR and OR. A P value < 0.05 was considered significant as were RR or OR whose 95% CI excluded unity. The Kappa statistic was calculated to assess degree of agreement between the two reviewers.
Retrospective Cohort Study
Two-hundred-eighty-seven patients who received rapacuronium were identified, of whom 12 (4.2%) developed bronchospasm during induction of anesthesia. Agreement on the identification of bronchospasm during induction of anesthesia was 100% between the two reviewers (DMR and MSS, kappa = 1.0). Comments regarding the severity of the bronchospasm were notable in many of the 12 operative records (Table 1). In five episodes of bronchospasm there were comments describing an inability to move the chest after intubation and absent end-tidal carbon dioxide (Petco2) on the capnogram even after visual confirmation of correct positioning of the tracheal tube (Table 1). Despite the severity of the episode of bronchospasm, all 12 patients responded to therapy and subsequently underwent the scheduled surgical procedures.
There were no significant differences between patients with bronchospasm and patients without bronchospasm with respect to age, sex, ASA physical status, history of drug allergy, rapacuronium dose, use of morphine, and choice of IV induction drug (Table 2). Eleven of the 12 patients who developed bronchospasm with rapacuronium had undergone RSI (RR, 17.5; 95% CI, 2.9–∞ for RSI). A definite history of the presence or absence of RAD was noted in the preoperative evaluation or intraoperative anesthesia record of 186 of the 287 patients who received rapacuronium. The remaining 101 patients were classified as not having RAD, giving an overall incidence of RAD of 23%. The RR of developing bronchospasm was significantly increased (RR, 4.6) for patients with a history of prior RAD (Table 2).
During the 5-mo time period, 7558 anesthetic procedures were performed at The Children’s Hospital of Philadelphia, of which 6895 were recorded with a computerized anesthesia record system. The other 663 cases were performed outside the operating room suites where the CompuRecord® system was not available. We identified the use of vecuronium in 61.1% of the 4354 patients who received muscle relaxants before intubation. Pancuronium was used in 18.5%, succinylcholine in 8.2%, mivacurium in 4.8%, rapacuronium in 6.1%, and cisatracurium in 1.3%. Seventy-nine patients receiving general anesthesia were identified to have possibly experienced intraoperative bronchospasm. Fifty-six cases were excluded because the episode in question developed more than 15 min after the induction of anesthesia, the patient did not receive a muscle relaxant, or the anesthetic record contained insufficient data to support the diagnosis of bronchospasm. The 23 remaining cases and 92 controls constitute the 115 subjects for the case-control study. Twelve of the 23 patients with postinduction bronchospasm received rapacuronium and 11 received other muscle relaxants (Table 3). Agreement between the two independent reviewers (DMR and MSS) was high on the initial evaluation (kappa 0.87). During a second review of the four discordant cases a consensus was reached between the reviewers, with three patients classified as having developed bronchospasm and one patient as not developing bronchospasm.
There were no significant differences between patients with bronchospasm and patients without bronchospasm with respect to age, sex, ASA physical status, history of drug allergy, rapacuronium dose, use of morphine, or choice of IV induction drug (Table 3). A definite history of the presence or absence of RAD was noted in the preoperative evaluation or intraoperative anesthesia record of 82 of the 115 cases and controls. The remaining 33 patients were classified as not having RAD. Cases were more likely than controls to have a history of RAD (OR, 3.3) (Table 3). Children with bronchospasm during induction of anesthesia were several times more likely to have received rapacuronium than other muscle relaxants compared with controls (OR, 10.1) (Table 3). The narrative comments in the anesthesia records for patients with bronchospasm who received muscle relaxants other than rapacuronium did not reflect the same severity as for the patients who received rapacuronium (Table 1).
The evidence from both the retrospective cohort study and the case-control study suggests that there was a substantial increase in risk for developing broncho-spasm in pediatric patients who received rapacuronium as a muscle relaxant. The written comments in the anesthesia records suggest that the severity of the episodes of bronchospasm may have been more intense than those associated with the use of other muscle relaxants. Case reports from other institutions have also suggested safety issues regarding rapacuronium administration that had not been detected in the prerelease testing of this drug, and this prompted voluntary withdrawal of the product by the manufacturer on March 27, 2001 (5). Our data go beyond the anecdotal information provided by previous case reports by estimating the incidence of the problem and by identifying possible associated risk factors including a history of RAD and RSI.
We identified rapacuronium as a possible cause of bronchospasm in part because of the severity of several of the episodes and withdrew rapacuronium from clinical use at the end of July 2000. There have been case reports of severe bronchospasm temporally related to the administration of rapacuronium during the induction of anesthesia including reports of five deaths (6). In recent published reports, 14 of 21 patients had very severe bronchospasm after rapacuronium use (7–9). Descriptions of the events were laden with catastrophic phrases such as inability to ventilate, no chest wall movement, lack of detection of Petco2 on capnography despite visual confirmation of the correct placement of the tracheal tube, and oxygen desaturation. Narrative comments in our anesthetic records were similar to those previously reported. (Table 1).
Bronchospasm has been reported with the use of many neuromuscular blocking drugs including succinylcholine, d-tubocurarine, atracurium, rocuronium, mivacurium, and rapacuronium (7–19). A 10.7% incidence of pulmonary side effects (bronchospasm and increased airway pressure) in patients receiving rapacuronium has been reported with only one case of bronchospasm described as a serious adverse experience (20). Another study reported a 14.7% incidence of bronchospasm with rapacuronium administration without any increases in plasma histamine levels (21). None of these patients was difficult to ventilate. We found that pediatric patients with bronchospasm during induction of anesthesia were 10 times more likely to have received rapacuronium than another neuromuscular blocking drug. We estimated the incidence of bronchospasm for patients receiving rapacuronium at 4.2%. The manufacturer’s estimated incidence in the package insert is 3.2%, which lies within the 95% CI of our estimate.
Many nondepolarizing drugs have been administered in larger than usual doses to speed onset of clinical effect (22). Because RSI was a risk factor for bronchospasm, it might have been expected that patients with bronchospasm would have received larger doses of rapacuronium. However, we found no difference in the dose of rapacuronium between patients who developed bronchospasm and those who did not. This suggests the possibility that the rapidity of drug administration during RSI rather than the magnitude of the dose may play a role in the development of bronchospasm.
Historically, the anesthesia record has been seen as a means of documenting the patient’s condition. Because the advent of the computerized anesthesia record has increased the ease of extracting and interpreting data, the role of the anesthesia record has expanded; it is now valuable as a research tool. However, research using the automated anesthesia recording system has not yet been systematically conducted in pediatrics, and we are aware of only two publications that have used these records to determine outcomes in anesthetized adults (23,24). This study shows the importance of a computerized anesthesia record for research, especially in investigating adverse events. It is extremely useful in conducting retrospective cohort and case-control studies. Although these observational studies cannot prove cause and effect, they can offer valuable information that cannot be obtained in a randomized clinical trial because of ethical or practical considerations. This may be particularly useful for identification of clinical problems that occur infrequently.
One of the major limitations of our study was the possibility that we did not capture all the reports of postinduction bronchospasm for patients who did not receive rapacuronium. In our computerized anesthesia record system there is a check box for a standard comment of albuterol aerosol treatment. Some of the anesthesiologists may have entered a narrative comment of the use of albuterol rather than checking the standard comment. Although our computerized anesthesia recording system allows for easy querying of common drugs and standardized comments, it does not allow us to query narrative comments in the body of the anesthesia record. We can be confident that our study captured all the patients who received rapacuronium because we examined the record of every patient who received this drug. However, we may not have captured all cases where bronchospasm occurred if another muscle relaxant was used. Unless the anesthesiologist voluntarily reported this complication to our CQI database, it would be irretrievable. Although voluntary reporting in CQI is not reliable for minor events that are not life-threatening, it is more likely that all severe cases will be reported (25). Although only 6 of the 12 cases of bronchospasm after the administration of rapacuronium were reported to the CQI database, these were the severe cases. A better design of the record and increased education of the users may resolve this problem.
The presence or absence of RAD could not be definitely determined for 35.2% of the retrospective cohort and 28.7% in the case-control subjects. However, we took the conservative approach by classifying all of these patients as not having a history of RAD. Despite this approach, the RR for a history of RAD was 4.6 (95% CI, 1.5–14.3) in the Bronchospasm group. If we assumed that the total incidence of RAD in the entire cohort was the same as the incidence in patients in whom a definite presence or absence of RAD history was established, the RR for RAD would be even larger at 5.6 (95% CI, 1.5–20.3).
In conclusion, this retrospective cohort study has determined that a history of RAD and the RSI were risk factors for the development of bronchospasm after induction of anesthesia and rapacuronium administration. In a case-control study we have determined that children with bronchospasm during induction of anesthesia were several times more likely to have received rapacuronium than other muscle relaxants.
1. Fisher DM. Neuromuscular blocking agents in paediatric anaesthesia. BJA 1999; 83: 58–64.
2. Kadar AG, Kramer BA, Barth MC, White PF. Rapacuronium: an alternative to succinylcholine for electroconvulsive therapy. Anesth Analg 2001; 92: 1171–2.
3. Meakin GH, Meretoja OA, Motsch J, et al. A dose-ranging study of rapacuronium in pediatric patients. Anesthesiology 2000; 92: 1002–9.
4. Onrust SV, Foster RH. Rapacuronium bromide: a review of its use in anaesthetic practice. Drugs 1999; 58: 887–918.
7. Kron SS. Severe bronchospasm and desaturation in a child associated with rapacuronium. Anesthesiology 2001; 94: 923–4.
8. Meakin GH, Pronske EH, Lerman J, et al. Bronchospasm after rapacuronium in infants and children. Anesthesiology 2001; 94: 926–7.
9. Naguib M. How serious is the bronchospasm induced by rapacuronium? Anesthesiology 2001; 94: 924–5.
10. Brandom BW. Neuromuscular blocking drugs in pediatric patients. Anesth Analg 2000; 90 Suppl: 1–8.
11. Fisher DM. Neuromuscular blocking agents in paediatric anaesthesia. BJA 1999; 83: 58–64.
12. Goudsouzian NG. Rapacuronium and bronchospasm. Anesthesiology 2001; 94: 727–8.
13. Moss J. Muscle relaxants and histamine release. Acta Anaesthesiol Scand Suppl 1995; 106: 7–12.
14. Naguib M, Magboul MM. Adverse effects of neuromuscular blockers and their antagonists. Mid East J Anesthesiol 1998; 4: 341–73.
15. O’Callaghan AC, Scadding G, Watkins J. Bronchospasm following the use of vecuronium. Anaesthesia 1986; 41: 940–2.
16. Sale JP. Bronchospasm following the use of atracurium. Anaesthesia 1983; 38: 511–2.
17. Sanchez-Guerrero IM, Tortosa JA, Hernandez-Palazon J, Escudero AI. Anaphylactoid reaction induced by pancuronium during general anaesthesia. Eur J Anaesthesiol 1998; 15: 613–4.
18. Siler JN, Mager JG Jr., Wyche MQ Jr. Atracurium: hypotension, tachycardia and bronchospasm. Anesthesiology 1985; 62: 645–6.
19. Watkins J. Adverse reaction to neuromuscular blockers: frequency, investigation, and epidemiology. Acta Anaesthesiol Scand Suppl 1994; 102: 6–10.
20. Sparr HJ, Mellinghoff H, Blobner M, Noldge-Schomburg G. Comparison of intubating conditions after rapacuronium (Org 9487) and succinylcholine following rapid sequence induction in adult patients. BJA 1999; 82: 537–41.
21. Levy JH, Pitts M, Thanopoulous A, et al. The effects of rapacuronium on histamine release and hemodynamics in adult patients undergoing general anesthesia. Anesth Analg 1999; 89: 290–5.
22. Stoelting RK, Miller RD. Basics of Anesthesia, 3rd ed. New York: Churchill Livingstone Inc., 1994: 94–5.
23. Sanborn KV, Castro J, Kuroda M, Thys DM. Detection of intraoperative incidents by electronic scanning of computerized anesthesia records: comparison with voluntary reporting. Anesthesiology 1996; 85: 977–87.
24. Reich DL, Bodian CA, Krol M, et al. Intraoperative hemodynamic predictors of mortality, stroke, and myocardial infarction after coronary artery bypass surgery. Anesth Analg 1999; 89: 814–22.
25. Cooper JB. Is voluntary reporting of critical events effective for quality assurance? Anesthesiology 1996; 85: 961–4.