Chapter 24. The Impact Of Intraoperative Monitoring On Patient Safety
Salim D. Islam, M.D.
Andrew D. Auerbach, M.D., M.P.H.
University of California, San Francisco School of Medicine
Until the 1960s, intraoperative monitoring consisted of a blood pressure cuff, electrocardiogram (ECG), stethoscope, and the vigilance of the anesthesiologist. Over the next 2 decades, the array of available monitors burgeoned, and clinical practice varied widely. In 1986, in an effort to improve patient safety, standards for intraoperative monitoring were developed and implemented by the American Society of Anesthesiologists (ASA).1 They have been almost universally adopted by anesthesia providers in the United States and now form the standard of care in this country. The ASA standards are summarized in Figure 24.1.
Concurrently with the implementation of better monitoring, anesthesia-related mortality has fallen sharply. Proponents of monitoring claim that better monitoring is the reason for improvement in patient safety.2-4 Others have claimed that advances in knowledge and training combined with the development of safer medications have had as much impact on patient safety as the adoption of monitoring standards.5,6 In this chapter, we evaluate the evidence linking intraoperative monitoring to patient safety.
Intraoperative monitoring involves the use of mechanical devices to record and display physiologic parameters such as heart rate, blood pressure, oxygen saturation, and temperature. Standard routine monitoring is noninvasive, employing blood pressure cuff, ECG, and pulse oximetry.
Invasive monitors such as arterial and central venous catheters and transesophageal echocardiography may provide more detailed and timely physiologic information, but also pose an increased risk for iatrogenic complications. In practice these monitors are used selectively, and are not reviewed here.
Prevalence and Severity of the Target Safety Problem
Death due to anesthesia has become rare. In one large Canadian study involving 27,184 inpatients who underwent anesthesia, physician review of 115 randomly selected "major events" classified less than 20% as having any anesthetic involvement, with no deaths even partially attributed to anesthesia.7 In the United States, the mortality due to general anesthesia has been estimated at approximately 5000 deaths per year (in the 1970s),8 with approximately half that number estimated in the 1980s.9 Thus, general anesthesia represents the one aspect of healthcare where the risk of death is low enough to rival the safety record achieved in other high-risk industries such as aviation.10
By contrast, morbid events (complications) related to anesthetic care are likely more prevalent and difficult to classify as preventable or unavoidable. Because certain aspects of monitoring may reduce the incidence of morbid events unrelated to anesthesia, assessing the impact of monitoring practices solely on anesthetic outcomes may be inappropriate. For example, detection of a consistent decrease in intraoperative blood pressure may signal unrecognized bleeding, allowing the anesthesiologist to alert the surgeon to this possibility and prompting appropriate management. While intraoperative hemorrhage does not represent an anesthetic complication, intraoperative blood pressure monitoring can clearly contribute to the overall safety of the surgical patient. Thus, the scope of intraoperative morbidity targeted by anesthetic monitoring practices is much broader than the set of possible complications attributable solely to the administration of anesthesia.7-9
Opportunities for Impact
In the United States, there are no mandatory regulations for monitoring practices. However, virtually all anesthesiologists abide by the monitoring standards set forth by the 1986 ASA standards, last modified in 1998 (Figure 24.1). Although these standards were implemented with only speculative evidence of their benefit,4 few clinicians doubt their merit.
Using a structured MEDLINE search, we identified articles presenting data related to the impact of perioperative monitoring. Many of these studies11-17 involved Level 4 designs (e.g., observational studies without a control group). For instance, several of the articles11-13,15 reported data from the Australian Incident Monitoring Study and involved analysis of a case series of 2000 incident reports without accompanying controls. Other studies only indirectly pertained to intraoperative monitoring. One study surveyed anesthesiologists regarding their views on the appropriate alarm settings for intraoperative blood pressure monitoring.18 Another focused on the personnel performing intraoperative monitoring—physician anesthesiologists versus certified nurse anesthetists.19 We chose not to purse this contentious and intensely political comparison, as few studies have compared the outcomes achieved by these two groups. Moreover, our reviewer team did not include a nurse anesthetist, making any conclusions drawn more susceptible to bias. Of the 3 remaining studies, one involved a non-randomized clinical trial (Level 2), but a Level 3 outcome.20
The remaining 2 studies met our inclusion criteria (Chapter 3). One was a retrospective analysis of anesthesia accidents before and after the implementation of monitoring standards (Level 3),2 and the other used a prospective, randomized, controlled trial design (Level 1) to assess the impact of pulse oximetry on postoperative complications (Level 1 outcome).21
The 2 studies2,21 meeting the methodologic inclusion criterion reported morbidity and mortality (Level 1) attributable to anesthesia, i.e., a major complication or death occurring in the immediate postoperative period not obviously explained by the patient's underlying condition or the operation itself.
Evidence for Effectiveness of the Practice
Through a review of cases reported to a liability insurer, Eichhorn identified 11 major intraoperative accidents solely attributable to anesthesia among over 1,000,000 cases performed at the nine Harvard hospitals from 1976-1988.2 Eight of these accidents were judged to be preventable as they were caused by failure to ventilate or to deliver adequate oxygen to the patient. Only one of these accidents occurred after the adoption of monitoring standards in mid-1985, supporting the safety benefit of intraoperative monitoring standards, although the difference in accident frequency was not statistically significant.
In a multicenter, randomized, controlled trial of 20,802 surgical patients, Moller et al21 studied the impact of perioperative pulse oximetry on patient outcome. Despite the large sample, the authors were unable to show a difference in in-hospital mortality or postoperative complications. During anesthesia and in the post-anesthesia care unit (PACU), more episodes of hypoxemia and myocardial ischemia were detected in patients monitored with pulse oximetry.21
Potential for Harm
Routine noninvasive monitoring carries minimal (although not zero) additional risk for iatrogenic complications from the devices themselves. Current standard of practice requires that they be used in all cases of general or regional anesthesia. However the number of monitors and their concomitant alarms raises the possibility of additional harm. A study of monitor alarms in the intensive care unit (ICU) suggested that monitor alarms might actually reduce quality of care because of their high frequency and low specificity. In this study, an alarm occurred every 37 minutes, and in the majority of cases (72%) no change in management was indicated as a result.22
Costs and Implementation
The costs of intraoperative monitors are largely fixed in the acquisition cost of the monitoring device. Incremental patient costs for disposables are minimal.
The inability of a very large multicenter study21 to detect a benefit in morbidity and mortality from pulse oximetry—by all accounts the most useful monitor—suggests that the magnitude of benefit may be so small that an adequate study to detect this difference may not be feasible. Along with capnography (carbon dioxide monitoring), pulse oximetry is often cited as the monitoring method most able to detect potential critical incidents early enough to prevent adverse outcomes.2,6 This conjecture is supported by the ASA Closed Claims Study. Analyzing 1175 claims, the study concluded that the combination of pulse oximetry and capnography "could be expected" to help prevent anesthetic-related morbidity and mortality.23
Despite a lack of randomized trial data, the practice of noninvasive intraoperative monitoring has become standard of care. This has resulted from the ASA Monitoring Standards and physicians' faith in the practice based on its face value, along with some confirmatory evidence drawn from incident reporting systems.11,16,17 As such, it seems likely that future research into intraoperative monitoring will be unable to study approaches that do not include standard, noninvasive monitoring. Future investigation might seek to determine which monitoring methods detect "near misses" more effectively.
Moving beyond non-invasive techniques, there is a great need to identify which specialized monitors provide a safety benefit in selected patient populations. The use of pulmonary artery catheters for monitoring critically ill patients represents a well-known example of a practice with substantial face validity but unclear impact on patient outcomes.24,25 In addition, new, noninvasive alternatives to invasive monitors (e.g., esophageal or spirometry-based cardiac output monitors) may ultimately allow us to obtain the same information at less risk to the patient.
Figure 24.1. ASA standards for basic anesthetic monitoring*
Standard 1: Qualified anesthesia personnel shall be present in the room throughout the conduct of all general anesthetics, regional anesthetics and monitored anesthesia care
Standard 2: During all anesthetics, the patient's oxygenation, ventilation, circulation, and temperature shall be continually* evaluated
Oxygen analyzer for inspired gases
Auscultation of breath sounds
Continuous* ECG display
Monitor temperature when changes are intended, anticipated, or suspected
* The term "continuous" means prolonged without interruption; "continually" means repeated regularly and frequently. ECG indicates electrocardiography; BP, blood pressure.
1. American Society of Anesthesiologists. Standards of the American Society of Anesthesiologists: Standards for Basic Anesthetic Monitoring. Available at: http://www.asahq.org/Standards/02.html. Accessed May 23, 2001.
2. Eichhorn JH. Prevention of intraoperative anesthesia accidents and related severe injury through safety monitoring. Anesthesiology 1989;70:572-577.
3. Eichhorn JH. Effect of monitoring standards on anesthesia outcome. Int Anesthesiol Clin 1993;31:181-196.
4. Keenan RL, Boyan CP. Cardiac arrest due to anesthesia. A study of incidence and causes. JAMA 1985;253:2373-2377.
5. Orkin FK. Practice standards: the Midas touch or the emperor's new clothes? Anesthesiology 1989;70:567-571.
6. Brodsky JB. What intraoperative monitoring makes sense? Chest 1999;115:101S-105S.
7. Cohen MM, Duncan PG, Pope WD, Biehl D, Tweed WA, MacWilliam L, et al. The Canadian four-centre study of anaesthetic outcomes: II. Can outcomes be used to assess the quality of anaesthesia care? Can J Anaesth 1992;39:430-439.
8. Phillips OC, Capizzi LS. Anesthesia mortality. Clin Anesth 1974;10:220-244.
9. Deaths during general anesthesia: technology-related, due to human error, or unavoidable? An ECRI technology assessment. J Health Care Technol 1985;1:155-175.
10. National Transportation Safety Board. Aviation Accident Statistics. Available at: http://www.ntsb.gov/aviation/Stats.htm. Accessed May 28, 2001.
11. Webb RK, van der Walt JH, Runciman WB, Williamson JA, Cockings J, Russell WJ, et al. The Australian Incident Monitoring Study. Which monitor? An analysis of 2000 incident reports. Anaesth Intensive Care 1993;21:529-542.
12. Klepper ID, Webb RK, Van der Walt JH, Ludbrook GL, Cockings J. The Australian Incident Monitoring Study. The stethoscope: applications and limitations-an analysis of 2000 incident reports. Anaesth Intensive Care 1993;21:575-578.
13. Cockings JG, Webb RK, Klepper ID, Currie M, Morgan C. The Australian Incident Monitoring Study. Blood pressure monitoring-applications and limitations: an analysis of 2000 incident reports. Anaesth Intensive Care 1993;21:565-569.
14. Hewer I, Drew B, Karp K, Stotts N. The utilization of automated ST segment analysis in the determination of myocardial ischemia. AANA J 1997;65:351-356.
15. Williamson JA, Webb RK, Cockings J, Morgan C. The Australian Incident Monitoring Study. The capnograph: applications and limitations-an analysis of 2000 incident reports. Anaesth Intensive Care 1993;21:551-557.
16. Spittal MJ, Findlay GP, Spencer I. A prospective analysis of critical incidents attributable to anaesthesia. Int J Qual Health Care 1995;7:363-371.
17. Findlay GP, Spittal MJ, Radcliffe JJ. The recognition of critical incidents: quantification of monitor effectiveness. Anaesthesia 1998;53:595-598.
18. Asbury AJ, Rolly G. Theatre monitor alarm settings: a pilot survey in Scotland and Belgium. Anaesthesia 1999;54:176-180.
19. Silber JH, Kennedy SK, Even-Shoshan O, Chen W, Koziol LF, Showan AM, et al. Anesthesiologist direction and patient outcomes. Anesthesiology 2000;93:152-163.
20. Kay J, Neal M. Effect of automatic blood pressure devices on vigilance of anesthesia residents. J Clin Monit 1986;2:148-150.
21. Moller JT, Johannessen NW, Espersen K, Ravlo O, Pedersen BD, Jensen PF, et al. Randomized evaluation of pulse oximetry in 20,802 patients: II. Perioperative events and postoperative complications. Anesthesiology 1993;78:445-453.
22. Chambrin MC, Ravaux P, Calvelo-Aros D, Jaborska A, Chopin C, Boniface B. Multicentric study of monitoring alarms in the adult intensive care unit (ICU): a descriptive analysis. Intensive Care Med 1999;25:1360-1366.
23. Tinker JH, Dull DL, Caplan RA, Ward RJ, Cheney FW. Role of monitoring devices in prevention of anesthetic mishaps: a closed claims analysis. Anesthesiology 1989;71:541-546.
24. Connors AF, Speroff T, Dawson NV, Thomas C, Harrell FE, Wagner D, et al. The effectiveness of right heart catheterization in the initial care of critically ill patients. SUPPORT Investigators. JAMA 1996;276:889-897.
25. Ivanov R, Allen J, Calvin JE. The incidence of major morbidity in critically ill patients managed with pulmonary artery catheters: a meta-analysis. Crit Care Med 2000;28:615-619.