- Email: [email protected]
Laryngeal mask airway versus endotracheal tube for outpatient surgery: analysis of anesthesia-controlled time
Original Contributions Laryngeal Mask Airway versus Endotracheal Tube for Outpatient Surgery: Analysis of Anesthesia-Controlled Time Bernd Hartmann, MD,* Anne Banzhaf, MD,* Axel Junger, MD,† Rainer Ro¨hrig, MD,* Matthias Benson, MD,‡ Rainer Schu¨rg, MD,* Gunter Hempelmann, MD§ Department of Anesthesiology, Intensive Care, and Pain Therapy, University Hospital, Giessen, Germany *Dr. med. †PD Dr. med., MBA(FIT) ‡PD Dr. med. §Prof. Dr. med. Dr. h.c. Address correspondence to PD Dr. med. A. Junger at the Abteilung Anaesthesiologie, Intensivmedizin, Schmerztherapie, Universita¨tsklinikum Giessen, Rudolf-Buchheim-Str. 7, 35392 Giessen, Germany. E-mail: [email protected] Supported in part by a grant from IMESO GmbH, Hu¨ttenberg, Germany. The founding agreement ensured the authors’ independence in designing the study, interpreting the data, writing and publishing the report. PD Dr. med. M. Benson is a partner of the IMESO GmbH (Hu¨ttenberg, Germany) and is a faculty member of the University Hospital Giessen. None of the other authors or participants has any financial interest in the subject matter, materials, or equipment discussed or in competing materials. MoReData GmbH, Giessen, Germany, helped with the data management and statistical evaluation. Presented in part at the 50th German Congress of Anesthesiology, Munich, April 9th to 12th, 2003. Received for publication April 13, 2003; revised manuscript accepted for publication July 17, 2003. Journal of Clinical Anesthesia 16:195–199, 2004 © 2004 Elsevier Inc. All rights reserved. 360 Park Avenue, South, New York, NY 10010
Study Objective: To show that efficiency of operating room times can be improved significantly using rapid changes between operative procedures. Design: Randomized, prospective clinical study. Setting: Tertiary care university hospital, elective peripheral trauma-related orthopedic surgery. Patients: 72 adult, ASA physical status I, II, and III patients scheduled for elective peripheral trauma-related orthopedic surgery requiring general anesthesia. Interventions: Patient airways were managed using either a Laryngeal Mask Airway (LMA) or an endotracheal tube (ETT) in the hands of anesthesiologists experienced in both. They were not informed as to the primary intention of the study. All perioperative data, including the preoperative and postoperative outpatient stay at the outpatient surgical ward, were recorded with an anesthesia information management system. Measurements: The primary outcome measures were: time needed for anesthesia induction and emergence from anesthesia. All manual recording into the anesthesia information management system during anesthesia was accomplished by nurses who were uninformed as to the aim of the study. Main Results: Anesthesia induction was significantly (p ⬍ 0.01) shorter using LMAs (means ⫾ SD, medians, [interquartile ranges]) (LMA: 5.8 ⫾ 1.5, 5, [5;7] vs. ETT: 7.4 ⫾ 1.8, 7, [7;8] min), whereas emergence from anesthesia was not different (LMA: 11.8 ⫾ 3.3, 11, [9;14] vs. ETT: 13.2 ⫾ 4.8; 12, [10;16] min). Conclusion: The clinical relevance of reduced anesthesia induction time using LMA is questionable. The lack of difference in emergence time could be a result of the use of total intravenous anesthesia. © 2004 by Elsevier Inc. Keywords: Computerized patient record; data management system; health economics; Laryngeal Mask Airway; operating room information systems.
Introduction Process analysis in surgical and anesthesia procedures are becoming more and more important because it may reveal opportunities for optimizing resource 0952-8180/04/$–see front matter doi:10.1016/j.jclinane.2003.07.008
Original Contributions
utilization. This is especially important in the operating room (OR), because 33% of total hospital costs for surgical care are spent on OR costs.1 Anesthesia care could contribute to reducing OR costs by decreasing anesthesia-controlled time (ACT). Therefore, anesthesia induction and emergence from anesthesia have to be time-optimized without neglecting patient care, especially when not performed in separate rooms, thereby interrupting the surgical workflow in the OR. For several years, new anesthetic drugs have been available for clinical use, allowing so-called ‘fast track’ anesthesia to be performed.2,3 Using the Laryngeal Mask Airway (LMA), muscle relaxation can be avoided. Since its introduction into routine clinical practice by Brain in 1983,4 the LMA has been broadly accepted as simplifying airway management for ambulatory surgery.5,6 All these factors together might lead to an acceleration of clinical routine in the OR. There are several studies6 – 8 comparing the use of the LMA with endotracheal tube (ETT). Todd7 found a significantly shorter recovery time for the LMA for ambulatory oral surgery, whereas Joshi et al.6 found no significant differences between times needed for placement or for removal of the two airway devices during ambulatory surgery. Azma et al.8 found both a significantly shorter time for airway placement and removal of the LMA compared with ETT in patients undergoing colorectal surgery with combined general and regional anesthesia. To evaluate potential resources for reducing time in the OR, we performed a prospective, randomized study to compare anesthesia-related times between total intravenous anesthesia (TIVA) with LMA or ETT facilitated by nondepolarizing muscle relaxation with mivacurium in ambulatory trauma-related peripheral orthopedic surgery.
Materials and Methods Patients The study was approved by the ethics committee of the medical faculty of the Justus-Liebig-University Giessen, Germany. We studied surgical outpatients aged 18 to 65 years with an ASA physical status I, II, or III9 undergoing elective trauma-related orthopedic surgery under general anesthesia and an expected duration of surgery less than 1 hour. Patients with anticipated airway difficulties (e.g., Mallampati classification ⬎II10), those at increased risk of regurgitation (e.g., body mass index ⬎30 kg m⫺2, history of gastroesophageal reflux, hiatal hernia, history of gastric surgery), those with local pharyngeal or laryngeal pathology (e.g., tumor, abscess), known allergies against study medications and other contraindications for the use of the LMA were excluded. Patients were informed about the available types of airway management, LMA or ETT, for general anesthesia and were randomized to one of the two types. Patients were enrolled in the study after giving their written informed consent.
Anesthesia Techniques Patients were generally premedicated with midazolam 3.75 to 7.5 mg orally 45 minutes prior to induction at the 196
J. Clin. Anesth., vol. 16, May 2004
outpatient surgery ward. Monitoring consisted of noninvasive blood pressure measurement, heart rate via electrocardiogram, hemoglobin oxygen saturation, end-tidal CO2 values, and inspiratory and expiratory oxygen concentrations. Following 3 minutes of preoxygenation, general anesthesia was induced with remifentanil (0.1 to 0.5 g kg⫺1min⫺1) and propofol (2 to 3 mg kg⫺1). LMA (LMAClassic™, Laryngeal Mask Company Limited, Oxon, UK; female: size 3 or 4; male: size 4 or 5, according to the patient’s weight) was inserted after the loss of the eyelash reflex by an experienced anesthesiologist. In the ETT group, mivacurium (0.2 mg kg⫺1; without repetitive doses) was used to facilitate the placement of a Magill tube (Mallinckrodt™, St. Louis, MO; internal diameter - female: 7.5 or 8.0 mm, male: 8.0 or 8.5 mm), which also was accomplished by an experienced anesthesiologist. Total intravenous anesthesia was maintained continuously with remifentanil (0.05 to 0.25 g kg⫺1min⫺1) and propofol (6 to 12 mg kg⫺1h⫺1) via motor-driven infusion pump. The anesthetic dose was adjusted to provide adequate anesthesia as indicated by the absence of intentional movement and hemodynamic stability. Pressure-controlled ventilation (peak pressure ⱕ20 cmH2O) with a positive endexpiratory pressure at approximately 5 cm H2O was maintained with an anesthesia ventilator (Julian, Dra¨ ger Medical AG & Co. KGaA, Germany) with an FiO2 ⫽ 0.35 and a respiratory rate of 10 to 12 min⫺1. It was altered as necessary to maintain an end-tidal CO2 concentration at 33 to 44 mmHg. At the end of surgery, all anesthetic infusions were stopped and the airway device was removed when patients were breathing sufficiently spontaneously and were able to follow simple commands. Neuromuscular blockade was only reversed when clinically necessary. For postoperative analgesia, all patients received a 50to 100-mg diclofenac suppository after anesthesia induction, which could be repeated at the outpatient surgery ward before discharge. At the postanesthesia care unit (PACU), intravenous (IV) piritramide (2.5 to 3 mg every 10 min if required) was used as further analgesic. Droperidol IV (0.625 mg) was used as an antiemetic and clonidine IV (75 g) to prevent postoperative shivering.
Computerized Anesthesia Recordkeeping All anesthetic procedures were recorded with the anesthesia record keeping system NarkoData Version 4 (Imeso GmbH, Hu¨ttenberg, Germany).11 This system collects all data from the preoperative anesthetic rounds to all routinely registered perioperative clinical anesthetic and surgical data, including the preoperative and postoperative stay at the outpatient surgery ward and the administered drugs. This system records relevant perioperative events and presents them on a graphical anesthetic chart. Beginning and ending of anesthesia/surgical procedure and the start of surgical preparation and positioning can only be recorded once per anesthetic procedure. All other times can be recorded repeatedly. Relevant data (start and end of anesthesia and surgery) must be complete and no missing values are tolerated. The logical order of the recorded times is checked using special algorithms. Thus,
0.47 0.60 0.50 0.86
Data are presented as numbers (n) and percentages (%) or mean value (MV), standard deviation (SD), median (X0.5), interquartile ranges (25th and 75th percentiles of the distribution; IQA), and 95% confidence interval (CI).
n CI IQA % n Table 1. Biometric Data
Sex (male/female) 24/12 66.7/33.3 20/16 55.6/44.4 ASA (I/II/III) 23/12/1 63.9/33.3/2.8 24/12/0 66.7/33.3/0.0 Age (years) 35.8 ⫾ 11.8 33 [25.8 ; 42.0] [31.8 ; 39.7] 37.6 ⫾ 11.0 33 [30.0 ; 48.5] [33.9 ; 41.4] 24.6 ⫾ 2.9 24.8 [22.5 ; 27.2] [23.6 ; 25.6] 24.5 ⫾ 2.7 24.6 [22.9 ; 26.2] [23.6 ; 25.4] Body mass index [kg/m2]
MV ⴞ SD X0.5 MV ⴞ SD X0.5
%
Tracheal Intubation (n ⴝ 36) Laryngeal Mask (n ⴝ 36)
IQA
CI
p-Value
Computerized anesthesia records for time analysis: Hartmann et al. AQ:1
an exact definition of the anesthesia record’s time frame is achieved and a complete data set at the end of anesthesia care can be obtained. Different time periods were calculated from these data. At the end of the outpatient stay before discharge, the record is imported into a relational database (Oracle™ Corporation, 500 Oracle Parkway, Redwood Shores, CA). For this prospective study, a standardized electronic anesthesia record was configured according to the study protocol, thus assuring a highly uniform quality of data for all required perioperative variables. To precisely document the ACT, a study nurse who was not blinded to airway technique entered all manually recorded data into the electronic anesthesia record in real time, including the anesthesia and surgery times and the beginning and end of the administration of anesthetic drugs.
Objective The primary outcome measures were the times needed for anesthesia induction and the emergence from anesthesia after the end of surgery, calculated in minutes. Anesthesia induction was defined as the time from the beginning of preoxygenation (appropriate to the start of the remifentanil administration) to the correct placement of the airway device (synonymous with the start of surgical preparation and positioning). Emergence from anesthesia was defined as the time from the end of surgical procedure (appropriate to the end of remifentanil and propofol administration) to extubation after the patient regained consciousness and was breathing spontaneously. Discharge from the OR to recovery occurred after extubation. The calculated sample size was oriented towards an assumed minimal clinically important difference of 2 to 3 minutes for the induction time between LMA and ETT. In their work, Joshi et al.6 calculated a standard deviation () of 3 minutes. Our hypothesis, ‘there is no difference between LMA and ETT in the induction time’, led to a minimum sample size per group of n ⫽ 2912,13 with a statistical significance of ␣ ⫽ 0.05 for “induction time” and a power of 1 ⫺  ⫽ 0.80 ( ⫽ 0.20). Because the sample size was calculated assuming a normal distribution pattern and homogeneity of variance of the main outcome measures, we increased the sample collective by 16%14 to account for a possible loss of power. Thus, a total sample size of n ⫽ 68 patients was attained.
Statistics For statistical evaluation, data were exported from the database into the SPSS statistics program (SPSS威 GmbH Software, Munich, Germany) using the Voyant™ software (Brossco Systems, Espoo, Finland), a standard structured query language (SQL) tool). Variables were tested for normal distribution using the Shapiro-Wilk test. The Mann-Whitney U-test was used for nonparametric variables for further analysis. Categorical variables were tested using either Chi-square statistics or the exact Fisher test. A p-value of less than 0.05 was considered statistically significant after correction with Bonferroni (two target J. Clin. Anesth., vol. 16, May 2004
197
Original Contributions
Table 2. Anesthesia-Related Times in Minutes Laryngeal Mask (n ⴝ 36) MV ⴞ SD X0.5
Anesthesia-Related Times Induction of anesthesia (min) Emergence from anesthesia (min) Incision-suture time (min) Time to awake (min)
5.8 ⫾ 1.5 11.8 ⫾ 3.3 20.4 ⫾ 15.4 4.5 ⫾ 3.0
5 11 16 4
IQA [5 [9 [9 [2
; ; ; ;
7] 14] 27] 5]
Tracheal Intubation (n ⴝ 36) MV ⴞ SD X0.5
CI
[5.2 ; 6.3] 7.4 ⫾ 1.8 [10.7 ; 12.9] 13.2 ⫾ 4.8 [15.2 ; 25.6] 21.7 ⫾ 11.2 [3.5 ; 5.5] 7.1 ⫾ 4.1
7 12 20 6.5
IQA
CI
p-Value
[7 ; 8] [10 ; 16] [12 ; 27] [5 ; 9]
[6.8 ; 8.0] [11.5 ; 14.8] [17.9 ; 25.5] [5.8 ; 8.5]
⬍0.01 0.28 0.32 ⬍0.01
Data are presented as mean value (MV), standard deviation (SD), median (X0.5), interquartile ranges (25th and 75th percentiles of the distribution; IQA), and 95% confidence interval (CI).
anesthesia time parameters ‘induction of anesthesia’ and ‘emergence from anesthesia’). Data are expressed as mean values (MV), standard deviation (SD), median, 25th and 75th percentiles of the distribution (interquartile range ⫽ IQA), and 95% confidence interval (CI), or as numbers and percentages.
in both groups. Clinically, there was no indication of regurgitation or aspiration. Additionally, postoperative sore throat was neither reported in the evening nor the day after the intervention, as checked by a routinely used standardized telephone survey by the responsible anesthesiologist of the outpatient surgical ward.
Results
Discussion
The electronic charts contained complete data sets of relevant perioperative time points, biometric data, and administered drugs for 72 studied patients. After randomization, 36 patients were studied in each group between February 8, 2001 and March 20, 2002. Table 1 shows the biometric data of the two groups without any significant differences in sex, ASA physical status, age, or body mass index. Anesthesia induction was significantly (p ⬍ 0.01) 1.5 minutes shorter using LMA, whereas emergence from anesthesia was not different (p ⫽ 0.28; power ⫽ 0.40) (Table 2). The analysis of all administered anesthetic drugs is presented in Table 3. Only the additional use of mivacurium for muscle relaxation facilitating endotracheal intubation varied according to the study protocol, whereas no differences in the need for perioperative anesthetics were noticed. The dose of midazolam for oral premedication and the requirements of postoperative analgesics (piritramide and diclofenac), antiemetics (droperidol: one patient in each group) and clonidine against postoperative shivering (LMA: 6 patients; ETT: 2 patients) were similar
Considering the increasing cost pressure in today’s health system, it is essential that the field of anesthesia analyze the cost-effectiveness in clinical routine. Process analysis allows the foundation of potential cost reduction, which in turn allows the optimization of ever decreasing resources. In Germany, as of 2003, a system of Diagnosis Related Groups (DRG) will take its part in hospital billing, following the Australian DRG model. This system will become mandatory for all insurance companies in hospital patient cases as of 2005. This means that the current system of billing anesthesia minutes internally will not be valid for anesthesia in the OR any longer. Rather, one sum will be paid per case, which is then divided up internally according to different measures. To remain competitive in the operating field, the selection of anesthesia technique has to be determined: cheaper anesthetic agents versus least possible ACT. A recently published retrospective data analysis15 demonstrated that in outpatient procedures, the use of a laryngeal mask led to a faster recovery and quicker OR release for the next procedure when compared with endotracheal intubation. In our prospective study, we found a slightly
Table 3. Total Dose of Administrated Drugs During the Perioperative Period Laryngeal Mask (n ⴝ 36) MV ⴞ SD Midazolam p.o. (mg) Propofol i.v. (mg) Remifentanil i.v. (g) Piritramide i.v. (mg) Diclofenac supp. (mg) Mivacurium i.v. (mg)
X0.5
7.0 ⫾ 2.2 7.5 596 ⫾ 143 573 511 ⫾ 143 486 9.6 ⫾ 6.1 7.5 104 ⫾ 30 100
Tracheal Intubation (n ⴝ 36)
IQA
CI
[7.5 ; 7.5] [492 ; 708] [408 ; 620] [6.3 ; 10.0] [100 ; 100]
[6.2 ; 7.7] [534 ; 658] [449 ; 573] [7.5 ; 11.6] [94 ; 114]
MV ⴞ SD
X0.5
7.0 ⫾ 1.8 7.5 640 ⫾ 161 610 611 ⫾ 239 550 8.8 ⫾ 5.1 7.5 109 ⫾ 43 100 15.0 ⫾ 5.4 15.3
IQA
CI
[7.5 ; 7.5] [549 ; 707] [388 ; 822] [5.0 ; 10.8] [100 ; 100] [13.0 ; 16.2]
[6.4 ; 7.6] [569 ; 712] [505 ; 717] [7.1 ; 10.5] [94 ; 123] [13.2 ; 16.8]
p-Value p p p p p
⫽ ⫽ ⫽ ⫽ ⫽
0.74 0.36 0.32 0.77 0.53
Premedication per oral (p.o.), intravenous anesthetics (i.v.), postoperative analgesics; suppository (supp.) presented as mean value (MV), standard deviation (SD), median (X0.5), interquartile ranges (25th and 75th percentiles of the distribution; IQA), and 95% confidence interval (CI). 198
J. Clin. Anesth., vol. 16, May 2004
Computerized anesthesia records for time analysis: Hartmann et al. AQ:1
shorter induction time and identical recovery times when using the laryngeal mask. However, the clinical relevance of an average 90 seconds faster induction is questionable, especially because the case number calculation was calculated with a clinically relevant difference of between 2 and 3 minutes, based on a standard deviation of 3 minutes as with Joshi et al.6 This group found comparative induction and extubation times of approximately 5 minutes, whereas the recovery time was shorter. In clinical practice, the patient may be transferred to the postoperative care unit with the LMA in situ, but not when still intubated. This would be within the scope of any comparative study. In addition, patients having ultra-short operations in the ETT group will presumably have a long emergence time, not as a result of the airway device, but to the offset of relaxation. Using mivacurium in our study, we tried to eliminate this possible effect. In his study, Todd7 found a significantly shorter recovery time for ambulatory oral surgery patients for the LMA group, leading to lower costs per case. In a small patient sample, the use of LMA (n ⫽ 10) in combination with central neuraxis anesthesia facilitates anesthesia induction an emergence from anesthesia in patients undergoing colorectal surgery as compared with ETT (n ⫽ 11).8 In clinical practice, these differences may seem marginal at first. Dexter et al.16 could show that the reduction of ACT saved no time for an additional procedure, but could help to reduce expensive overtime during on-call shifts. In comparison, in outpatient ORs with many shorter procedures, small amounts of time saved can lead to a relevant decrease in total OR time. Dexter and Macario17 could show using a computer simulation that a decrease in total OR time per case saves work time and allows additional procedures to be performed within the normal working hours. We calculated a decreased use of anesthetic agents in the LMA group, which was however statistically insignificant (excluding muscle relaxation). Joshi et al.6 came to similar conclusions. Muscle relaxation can be avoided altogether using an LMA. Avoiding painful laryngoscopies and intubation allows less use of anesthetic agents.18 However, this has no influence on recovery behavior. Cork et al.5 reported an increased use of anesthetics for ETT versus placing an LMA. They also noted a decreased incidence of adverse effects (nausea, vomiting, sore throat) using an LMA instead of an ETT.5 Our study showed no relevant postoperative complaints during the postoperative examination before releasing the patient. In analyzing the drugs given postoperatively, we could find no differences in treating nausea, vomiting, or shivering. In viewing the individual aspects of LMA versus ETT, decreased induction times, reduced anesthetic use, the lack of muscle relaxants, and decreased postoperative complications clearly mark the advantages of LMA over ETT. Additionally, Macario et al.19 could show with their cost analysis that the use of an LMA costs less than ETT for outpatient procedures with an anesthesia duration of more than 40 minutes and the prerequisite that a laryngeal mask can be used more than 40 times. From this point of view, use of a LMA has marked advantages over ETT.
Acknowledgments We would like to thank MoReData GmbH, Giessen, Germany, for their help in data management and statistical evaluation.
References 1. Macario A, Vitez TS, Dunn B, McDonald T: Where are the costs in perioperative care? Analysis of hospital costs and charges for inpatient surgical care. Anesthesiology 1995;83:1138 –44. 2. Joshi GP, Twersky RS: Fast tracking in ambulatory surgery. Ambul Surg 2000;8:185–90. 3. Scholz J, Steinfath M: Is remifentanil an ideal opioid for anesthesiologic management in the 21st century? [First Remifentanil ein ideales Opioid fu¨ r das ana¨ sthesiologische Management im 21. Jahrhundert?] Ana¨sthesiol Intensivmed Notfallmed Schmerzther 1996; 31:592–607. 4. Brain AI: The laryngeal mask—a new concept in airway management. Br J Anaesth 1983;55:801–5. 5. Cork RC, Depa RM, Standen JR: Prospective comparison of use of the laryngeal mask and endotracheal tube for ambulatory surgery. Anesth Analg 1994;79:719 –27. 6. Joshi GP, Inagaki Y, White PF, et al: Use of the laryngeal mask airway as an alternative to the tracheal tube during ambulatory anesthesia. Anesth Analg 1997;85:573–7. 7. Todd DW: A comparison of endotracheal intubation and use of the laryngeal mask airway for ambulatory oral surgery patients. J Oral Maxillofac Surg 2002;60:2–4. 8. Azma T, Kawai K, Okida M, Okada K, Tamura H: Use of the laryngeal mask airway in combination with regional anesthesia facilitates induction and emergence from general anesthesia in patients undergoing colorectal surgery. Hiroshima J Med Sci 2002;51:89 –92. 9. New classification of physical status. House of Delegates, American Society of Anesthesiologists. Anesthesiology 1963;24:111. 10. Mallampati SR: Clinical signs to predict difficult tracheal intubation (hypothesis). Can Anaesth Soc J 1983;30(3 Pt 1):316 –7. 11. Benson M, Junger A, Quinzio L, et al: Clinical and practical requirements of online software for anesthesia documentation an experience report. Int J Med Inf 2000;57:155–64. 12. Campbell MJ, Julious SA, Altman DG: Estimating sample sizes for binary, ordered categorical, and continuous outcomes in two group comparisons. BMJ 1995;311:1145–8. 13. Machin D, Campbell MJ, Fayers P, Pinol A: Sample Size Tables for Clinical Studies. Oxford: Blackwell Science, 1997. 14. Bu¨ ning H, Trenkler A: Nichtparametrische statistische Methoden. Berlin: Walter de Gruyter, 1994. 15. Junger A, Klasen J, Hartmann B, et al: Shorter discharge time after regional or intravenous anaesthesia in combination with laryngeal mask airway compared with balanced anaesthesia with endotracheal intubation. Eur J Anaesthesiol 2002;19:119 –24. 16. Dexter F, Coffin S, Tinker JH: Decreases in anesthesia-controlled time cannot permit one additional surgical operation to be reliably scheduled during the workday. Anesth Analg 1995;81:1263–8. 17. Dexter F, Macario A: Decrease in case duration required to complete an additional case during regularly scheduled hours in an operating room suite: a computer simulation study. Anesth Analg 1999;88:72–6. 18. Wilkins CJ, Cramp PG, Staples J, Stevens WC: Comparison of the anesthetic requirement for tolerance of laryngeal mask airway and endotracheal tube. Anesth Analg 1992;75:794 –7. 19. Macario A, Chang PC, Stempel DB, Brock-Utne JG: A cost analysis of the laryngeal mask airway for elective surgery in adult outpatients. Anesthesiology 1995;83:250 –7.
J. Clin. Anesth., vol. 16, May 2004
199
Study Objective: To show that efficiency of operating room times can be improved significantly using rapid changes between operative procedures. Design: Randomized, prospective clinical study. Setting: Tertiary care university hospital, elective peripheral trauma-related orthopedic surgery. Patients: 72 adult, ASA physical status I, II, and III patients scheduled for elective peripheral trauma-related orthopedic surgery requiring general anesthesia. Interventions: Patient airways were managed using either a Laryngeal Mask Airway (LMA) or an endotracheal tube (ETT) in the hands of anesthesiologists experienced in both. They were not informed as to the primary intention of the study. All perioperative data, including the preoperative and postoperative outpatient stay at the outpatient surgical ward, were recorded with an anesthesia information management system. Measurements: The primary outcome measures were: time needed for anesthesia induction and emergence from anesthesia. All manual recording into the anesthesia information management system during anesthesia was accomplished by nurses who were uninformed as to the aim of the study. Main Results: Anesthesia induction was significantly (p ⬍ 0.01) shorter using LMAs (means ⫾ SD, medians, [interquartile ranges]) (LMA: 5.8 ⫾ 1.5, 5, [5;7] vs. ETT: 7.4 ⫾ 1.8, 7, [7;8] min), whereas emergence from anesthesia was not different (LMA: 11.8 ⫾ 3.3, 11, [9;14] vs. ETT: 13.2 ⫾ 4.8; 12, [10;16] min). Conclusion: The clinical relevance of reduced anesthesia induction time using LMA is questionable. The lack of difference in emergence time could be a result of the use of total intravenous anesthesia. © 2004 by Elsevier Inc. Keywords: Computerized patient record; data management system; health economics; Laryngeal Mask Airway; operating room information systems.
Introduction Process analysis in surgical and anesthesia procedures are becoming more and more important because it may reveal opportunities for optimizing resource 0952-8180/04/$–see front matter doi:10.1016/j.jclinane.2003.07.008
Original Contributions
utilization. This is especially important in the operating room (OR), because 33% of total hospital costs for surgical care are spent on OR costs.1 Anesthesia care could contribute to reducing OR costs by decreasing anesthesia-controlled time (ACT). Therefore, anesthesia induction and emergence from anesthesia have to be time-optimized without neglecting patient care, especially when not performed in separate rooms, thereby interrupting the surgical workflow in the OR. For several years, new anesthetic drugs have been available for clinical use, allowing so-called ‘fast track’ anesthesia to be performed.2,3 Using the Laryngeal Mask Airway (LMA), muscle relaxation can be avoided. Since its introduction into routine clinical practice by Brain in 1983,4 the LMA has been broadly accepted as simplifying airway management for ambulatory surgery.5,6 All these factors together might lead to an acceleration of clinical routine in the OR. There are several studies6 – 8 comparing the use of the LMA with endotracheal tube (ETT). Todd7 found a significantly shorter recovery time for the LMA for ambulatory oral surgery, whereas Joshi et al.6 found no significant differences between times needed for placement or for removal of the two airway devices during ambulatory surgery. Azma et al.8 found both a significantly shorter time for airway placement and removal of the LMA compared with ETT in patients undergoing colorectal surgery with combined general and regional anesthesia. To evaluate potential resources for reducing time in the OR, we performed a prospective, randomized study to compare anesthesia-related times between total intravenous anesthesia (TIVA) with LMA or ETT facilitated by nondepolarizing muscle relaxation with mivacurium in ambulatory trauma-related peripheral orthopedic surgery.
Materials and Methods Patients The study was approved by the ethics committee of the medical faculty of the Justus-Liebig-University Giessen, Germany. We studied surgical outpatients aged 18 to 65 years with an ASA physical status I, II, or III9 undergoing elective trauma-related orthopedic surgery under general anesthesia and an expected duration of surgery less than 1 hour. Patients with anticipated airway difficulties (e.g., Mallampati classification ⬎II10), those at increased risk of regurgitation (e.g., body mass index ⬎30 kg m⫺2, history of gastroesophageal reflux, hiatal hernia, history of gastric surgery), those with local pharyngeal or laryngeal pathology (e.g., tumor, abscess), known allergies against study medications and other contraindications for the use of the LMA were excluded. Patients were informed about the available types of airway management, LMA or ETT, for general anesthesia and were randomized to one of the two types. Patients were enrolled in the study after giving their written informed consent.
Anesthesia Techniques Patients were generally premedicated with midazolam 3.75 to 7.5 mg orally 45 minutes prior to induction at the 196
J. Clin. Anesth., vol. 16, May 2004
outpatient surgery ward. Monitoring consisted of noninvasive blood pressure measurement, heart rate via electrocardiogram, hemoglobin oxygen saturation, end-tidal CO2 values, and inspiratory and expiratory oxygen concentrations. Following 3 minutes of preoxygenation, general anesthesia was induced with remifentanil (0.1 to 0.5 g kg⫺1min⫺1) and propofol (2 to 3 mg kg⫺1). LMA (LMAClassic™, Laryngeal Mask Company Limited, Oxon, UK; female: size 3 or 4; male: size 4 or 5, according to the patient’s weight) was inserted after the loss of the eyelash reflex by an experienced anesthesiologist. In the ETT group, mivacurium (0.2 mg kg⫺1; without repetitive doses) was used to facilitate the placement of a Magill tube (Mallinckrodt™, St. Louis, MO; internal diameter - female: 7.5 or 8.0 mm, male: 8.0 or 8.5 mm), which also was accomplished by an experienced anesthesiologist. Total intravenous anesthesia was maintained continuously with remifentanil (0.05 to 0.25 g kg⫺1min⫺1) and propofol (6 to 12 mg kg⫺1h⫺1) via motor-driven infusion pump. The anesthetic dose was adjusted to provide adequate anesthesia as indicated by the absence of intentional movement and hemodynamic stability. Pressure-controlled ventilation (peak pressure ⱕ20 cmH2O) with a positive endexpiratory pressure at approximately 5 cm H2O was maintained with an anesthesia ventilator (Julian, Dra¨ ger Medical AG & Co. KGaA, Germany) with an FiO2 ⫽ 0.35 and a respiratory rate of 10 to 12 min⫺1. It was altered as necessary to maintain an end-tidal CO2 concentration at 33 to 44 mmHg. At the end of surgery, all anesthetic infusions were stopped and the airway device was removed when patients were breathing sufficiently spontaneously and were able to follow simple commands. Neuromuscular blockade was only reversed when clinically necessary. For postoperative analgesia, all patients received a 50to 100-mg diclofenac suppository after anesthesia induction, which could be repeated at the outpatient surgery ward before discharge. At the postanesthesia care unit (PACU), intravenous (IV) piritramide (2.5 to 3 mg every 10 min if required) was used as further analgesic. Droperidol IV (0.625 mg) was used as an antiemetic and clonidine IV (75 g) to prevent postoperative shivering.
Computerized Anesthesia Recordkeeping All anesthetic procedures were recorded with the anesthesia record keeping system NarkoData Version 4 (Imeso GmbH, Hu¨ttenberg, Germany).11 This system collects all data from the preoperative anesthetic rounds to all routinely registered perioperative clinical anesthetic and surgical data, including the preoperative and postoperative stay at the outpatient surgery ward and the administered drugs. This system records relevant perioperative events and presents them on a graphical anesthetic chart. Beginning and ending of anesthesia/surgical procedure and the start of surgical preparation and positioning can only be recorded once per anesthetic procedure. All other times can be recorded repeatedly. Relevant data (start and end of anesthesia and surgery) must be complete and no missing values are tolerated. The logical order of the recorded times is checked using special algorithms. Thus,
0.47 0.60 0.50 0.86
Data are presented as numbers (n) and percentages (%) or mean value (MV), standard deviation (SD), median (X0.5), interquartile ranges (25th and 75th percentiles of the distribution; IQA), and 95% confidence interval (CI).
n CI IQA % n Table 1. Biometric Data
Sex (male/female) 24/12 66.7/33.3 20/16 55.6/44.4 ASA (I/II/III) 23/12/1 63.9/33.3/2.8 24/12/0 66.7/33.3/0.0 Age (years) 35.8 ⫾ 11.8 33 [25.8 ; 42.0] [31.8 ; 39.7] 37.6 ⫾ 11.0 33 [30.0 ; 48.5] [33.9 ; 41.4] 24.6 ⫾ 2.9 24.8 [22.5 ; 27.2] [23.6 ; 25.6] 24.5 ⫾ 2.7 24.6 [22.9 ; 26.2] [23.6 ; 25.4] Body mass index [kg/m2]
MV ⴞ SD X0.5 MV ⴞ SD X0.5
%
Tracheal Intubation (n ⴝ 36) Laryngeal Mask (n ⴝ 36)
IQA
CI
p-Value
Computerized anesthesia records for time analysis: Hartmann et al. AQ:1
an exact definition of the anesthesia record’s time frame is achieved and a complete data set at the end of anesthesia care can be obtained. Different time periods were calculated from these data. At the end of the outpatient stay before discharge, the record is imported into a relational database (Oracle™ Corporation, 500 Oracle Parkway, Redwood Shores, CA). For this prospective study, a standardized electronic anesthesia record was configured according to the study protocol, thus assuring a highly uniform quality of data for all required perioperative variables. To precisely document the ACT, a study nurse who was not blinded to airway technique entered all manually recorded data into the electronic anesthesia record in real time, including the anesthesia and surgery times and the beginning and end of the administration of anesthetic drugs.
Objective The primary outcome measures were the times needed for anesthesia induction and the emergence from anesthesia after the end of surgery, calculated in minutes. Anesthesia induction was defined as the time from the beginning of preoxygenation (appropriate to the start of the remifentanil administration) to the correct placement of the airway device (synonymous with the start of surgical preparation and positioning). Emergence from anesthesia was defined as the time from the end of surgical procedure (appropriate to the end of remifentanil and propofol administration) to extubation after the patient regained consciousness and was breathing spontaneously. Discharge from the OR to recovery occurred after extubation. The calculated sample size was oriented towards an assumed minimal clinically important difference of 2 to 3 minutes for the induction time between LMA and ETT. In their work, Joshi et al.6 calculated a standard deviation () of 3 minutes. Our hypothesis, ‘there is no difference between LMA and ETT in the induction time’, led to a minimum sample size per group of n ⫽ 2912,13 with a statistical significance of ␣ ⫽ 0.05 for “induction time” and a power of 1 ⫺  ⫽ 0.80 ( ⫽ 0.20). Because the sample size was calculated assuming a normal distribution pattern and homogeneity of variance of the main outcome measures, we increased the sample collective by 16%14 to account for a possible loss of power. Thus, a total sample size of n ⫽ 68 patients was attained.
Statistics For statistical evaluation, data were exported from the database into the SPSS statistics program (SPSS威 GmbH Software, Munich, Germany) using the Voyant™ software (Brossco Systems, Espoo, Finland), a standard structured query language (SQL) tool). Variables were tested for normal distribution using the Shapiro-Wilk test. The Mann-Whitney U-test was used for nonparametric variables for further analysis. Categorical variables were tested using either Chi-square statistics or the exact Fisher test. A p-value of less than 0.05 was considered statistically significant after correction with Bonferroni (two target J. Clin. Anesth., vol. 16, May 2004
197
Original Contributions
Table 2. Anesthesia-Related Times in Minutes Laryngeal Mask (n ⴝ 36) MV ⴞ SD X0.5
Anesthesia-Related Times Induction of anesthesia (min) Emergence from anesthesia (min) Incision-suture time (min) Time to awake (min)
5.8 ⫾ 1.5 11.8 ⫾ 3.3 20.4 ⫾ 15.4 4.5 ⫾ 3.0
5 11 16 4
IQA [5 [9 [9 [2
; ; ; ;
7] 14] 27] 5]
Tracheal Intubation (n ⴝ 36) MV ⴞ SD X0.5
CI
[5.2 ; 6.3] 7.4 ⫾ 1.8 [10.7 ; 12.9] 13.2 ⫾ 4.8 [15.2 ; 25.6] 21.7 ⫾ 11.2 [3.5 ; 5.5] 7.1 ⫾ 4.1
7 12 20 6.5
IQA
CI
p-Value
[7 ; 8] [10 ; 16] [12 ; 27] [5 ; 9]
[6.8 ; 8.0] [11.5 ; 14.8] [17.9 ; 25.5] [5.8 ; 8.5]
⬍0.01 0.28 0.32 ⬍0.01
Data are presented as mean value (MV), standard deviation (SD), median (X0.5), interquartile ranges (25th and 75th percentiles of the distribution; IQA), and 95% confidence interval (CI).
anesthesia time parameters ‘induction of anesthesia’ and ‘emergence from anesthesia’). Data are expressed as mean values (MV), standard deviation (SD), median, 25th and 75th percentiles of the distribution (interquartile range ⫽ IQA), and 95% confidence interval (CI), or as numbers and percentages.
in both groups. Clinically, there was no indication of regurgitation or aspiration. Additionally, postoperative sore throat was neither reported in the evening nor the day after the intervention, as checked by a routinely used standardized telephone survey by the responsible anesthesiologist of the outpatient surgical ward.
Results
Discussion
The electronic charts contained complete data sets of relevant perioperative time points, biometric data, and administered drugs for 72 studied patients. After randomization, 36 patients were studied in each group between February 8, 2001 and March 20, 2002. Table 1 shows the biometric data of the two groups without any significant differences in sex, ASA physical status, age, or body mass index. Anesthesia induction was significantly (p ⬍ 0.01) 1.5 minutes shorter using LMA, whereas emergence from anesthesia was not different (p ⫽ 0.28; power ⫽ 0.40) (Table 2). The analysis of all administered anesthetic drugs is presented in Table 3. Only the additional use of mivacurium for muscle relaxation facilitating endotracheal intubation varied according to the study protocol, whereas no differences in the need for perioperative anesthetics were noticed. The dose of midazolam for oral premedication and the requirements of postoperative analgesics (piritramide and diclofenac), antiemetics (droperidol: one patient in each group) and clonidine against postoperative shivering (LMA: 6 patients; ETT: 2 patients) were similar
Considering the increasing cost pressure in today’s health system, it is essential that the field of anesthesia analyze the cost-effectiveness in clinical routine. Process analysis allows the foundation of potential cost reduction, which in turn allows the optimization of ever decreasing resources. In Germany, as of 2003, a system of Diagnosis Related Groups (DRG) will take its part in hospital billing, following the Australian DRG model. This system will become mandatory for all insurance companies in hospital patient cases as of 2005. This means that the current system of billing anesthesia minutes internally will not be valid for anesthesia in the OR any longer. Rather, one sum will be paid per case, which is then divided up internally according to different measures. To remain competitive in the operating field, the selection of anesthesia technique has to be determined: cheaper anesthetic agents versus least possible ACT. A recently published retrospective data analysis15 demonstrated that in outpatient procedures, the use of a laryngeal mask led to a faster recovery and quicker OR release for the next procedure when compared with endotracheal intubation. In our prospective study, we found a slightly
Table 3. Total Dose of Administrated Drugs During the Perioperative Period Laryngeal Mask (n ⴝ 36) MV ⴞ SD Midazolam p.o. (mg) Propofol i.v. (mg) Remifentanil i.v. (g) Piritramide i.v. (mg) Diclofenac supp. (mg) Mivacurium i.v. (mg)
X0.5
7.0 ⫾ 2.2 7.5 596 ⫾ 143 573 511 ⫾ 143 486 9.6 ⫾ 6.1 7.5 104 ⫾ 30 100
Tracheal Intubation (n ⴝ 36)
IQA
CI
[7.5 ; 7.5] [492 ; 708] [408 ; 620] [6.3 ; 10.0] [100 ; 100]
[6.2 ; 7.7] [534 ; 658] [449 ; 573] [7.5 ; 11.6] [94 ; 114]
MV ⴞ SD
X0.5
7.0 ⫾ 1.8 7.5 640 ⫾ 161 610 611 ⫾ 239 550 8.8 ⫾ 5.1 7.5 109 ⫾ 43 100 15.0 ⫾ 5.4 15.3
IQA
CI
[7.5 ; 7.5] [549 ; 707] [388 ; 822] [5.0 ; 10.8] [100 ; 100] [13.0 ; 16.2]
[6.4 ; 7.6] [569 ; 712] [505 ; 717] [7.1 ; 10.5] [94 ; 123] [13.2 ; 16.8]
p-Value p p p p p
⫽ ⫽ ⫽ ⫽ ⫽
0.74 0.36 0.32 0.77 0.53
Premedication per oral (p.o.), intravenous anesthetics (i.v.), postoperative analgesics; suppository (supp.) presented as mean value (MV), standard deviation (SD), median (X0.5), interquartile ranges (25th and 75th percentiles of the distribution; IQA), and 95% confidence interval (CI). 198
J. Clin. Anesth., vol. 16, May 2004
Computerized anesthesia records for time analysis: Hartmann et al. AQ:1
shorter induction time and identical recovery times when using the laryngeal mask. However, the clinical relevance of an average 90 seconds faster induction is questionable, especially because the case number calculation was calculated with a clinically relevant difference of between 2 and 3 minutes, based on a standard deviation of 3 minutes as with Joshi et al.6 This group found comparative induction and extubation times of approximately 5 minutes, whereas the recovery time was shorter. In clinical practice, the patient may be transferred to the postoperative care unit with the LMA in situ, but not when still intubated. This would be within the scope of any comparative study. In addition, patients having ultra-short operations in the ETT group will presumably have a long emergence time, not as a result of the airway device, but to the offset of relaxation. Using mivacurium in our study, we tried to eliminate this possible effect. In his study, Todd7 found a significantly shorter recovery time for ambulatory oral surgery patients for the LMA group, leading to lower costs per case. In a small patient sample, the use of LMA (n ⫽ 10) in combination with central neuraxis anesthesia facilitates anesthesia induction an emergence from anesthesia in patients undergoing colorectal surgery as compared with ETT (n ⫽ 11).8 In clinical practice, these differences may seem marginal at first. Dexter et al.16 could show that the reduction of ACT saved no time for an additional procedure, but could help to reduce expensive overtime during on-call shifts. In comparison, in outpatient ORs with many shorter procedures, small amounts of time saved can lead to a relevant decrease in total OR time. Dexter and Macario17 could show using a computer simulation that a decrease in total OR time per case saves work time and allows additional procedures to be performed within the normal working hours. We calculated a decreased use of anesthetic agents in the LMA group, which was however statistically insignificant (excluding muscle relaxation). Joshi et al.6 came to similar conclusions. Muscle relaxation can be avoided altogether using an LMA. Avoiding painful laryngoscopies and intubation allows less use of anesthetic agents.18 However, this has no influence on recovery behavior. Cork et al.5 reported an increased use of anesthetics for ETT versus placing an LMA. They also noted a decreased incidence of adverse effects (nausea, vomiting, sore throat) using an LMA instead of an ETT.5 Our study showed no relevant postoperative complaints during the postoperative examination before releasing the patient. In analyzing the drugs given postoperatively, we could find no differences in treating nausea, vomiting, or shivering. In viewing the individual aspects of LMA versus ETT, decreased induction times, reduced anesthetic use, the lack of muscle relaxants, and decreased postoperative complications clearly mark the advantages of LMA over ETT. Additionally, Macario et al.19 could show with their cost analysis that the use of an LMA costs less than ETT for outpatient procedures with an anesthesia duration of more than 40 minutes and the prerequisite that a laryngeal mask can be used more than 40 times. From this point of view, use of a LMA has marked advantages over ETT.
Acknowledgments We would like to thank MoReData GmbH, Giessen, Germany, for their help in data management and statistical evaluation.
References 1. Macario A, Vitez TS, Dunn B, McDonald T: Where are the costs in perioperative care? Analysis of hospital costs and charges for inpatient surgical care. Anesthesiology 1995;83:1138 –44. 2. Joshi GP, Twersky RS: Fast tracking in ambulatory surgery. Ambul Surg 2000;8:185–90. 3. Scholz J, Steinfath M: Is remifentanil an ideal opioid for anesthesiologic management in the 21st century? [First Remifentanil ein ideales Opioid fu¨ r das ana¨ sthesiologische Management im 21. Jahrhundert?] Ana¨sthesiol Intensivmed Notfallmed Schmerzther 1996; 31:592–607. 4. Brain AI: The laryngeal mask—a new concept in airway management. Br J Anaesth 1983;55:801–5. 5. Cork RC, Depa RM, Standen JR: Prospective comparison of use of the laryngeal mask and endotracheal tube for ambulatory surgery. Anesth Analg 1994;79:719 –27. 6. Joshi GP, Inagaki Y, White PF, et al: Use of the laryngeal mask airway as an alternative to the tracheal tube during ambulatory anesthesia. Anesth Analg 1997;85:573–7. 7. Todd DW: A comparison of endotracheal intubation and use of the laryngeal mask airway for ambulatory oral surgery patients. J Oral Maxillofac Surg 2002;60:2–4. 8. Azma T, Kawai K, Okida M, Okada K, Tamura H: Use of the laryngeal mask airway in combination with regional anesthesia facilitates induction and emergence from general anesthesia in patients undergoing colorectal surgery. Hiroshima J Med Sci 2002;51:89 –92. 9. New classification of physical status. House of Delegates, American Society of Anesthesiologists. Anesthesiology 1963;24:111. 10. Mallampati SR: Clinical signs to predict difficult tracheal intubation (hypothesis). Can Anaesth Soc J 1983;30(3 Pt 1):316 –7. 11. Benson M, Junger A, Quinzio L, et al: Clinical and practical requirements of online software for anesthesia documentation an experience report. Int J Med Inf 2000;57:155–64. 12. Campbell MJ, Julious SA, Altman DG: Estimating sample sizes for binary, ordered categorical, and continuous outcomes in two group comparisons. BMJ 1995;311:1145–8. 13. Machin D, Campbell MJ, Fayers P, Pinol A: Sample Size Tables for Clinical Studies. Oxford: Blackwell Science, 1997. 14. Bu¨ ning H, Trenkler A: Nichtparametrische statistische Methoden. Berlin: Walter de Gruyter, 1994. 15. Junger A, Klasen J, Hartmann B, et al: Shorter discharge time after regional or intravenous anaesthesia in combination with laryngeal mask airway compared with balanced anaesthesia with endotracheal intubation. Eur J Anaesthesiol 2002;19:119 –24. 16. Dexter F, Coffin S, Tinker JH: Decreases in anesthesia-controlled time cannot permit one additional surgical operation to be reliably scheduled during the workday. Anesth Analg 1995;81:1263–8. 17. Dexter F, Macario A: Decrease in case duration required to complete an additional case during regularly scheduled hours in an operating room suite: a computer simulation study. Anesth Analg 1999;88:72–6. 18. Wilkins CJ, Cramp PG, Staples J, Stevens WC: Comparison of the anesthetic requirement for tolerance of laryngeal mask airway and endotracheal tube. Anesth Analg 1992;75:794 –7. 19. Macario A, Chang PC, Stempel DB, Brock-Utne JG: A cost analysis of the laryngeal mask airway for elective surgery in adult outpatients. Anesthesiology 1995;83:250 –7.
J. Clin. Anesth., vol. 16, May 2004
199