Chinese Medical Journal 2011;124(2):227-232
Intraoperative cardiac arrest during anesthesia: a retrospective study of 218 274 anesthetics undergoing non-cardiac surgery
AN Jian-xiong, Li-Ming Zhang, Erin A. Sullivan, , GUO Qu-lian , John P. Williams
AN Jian-xiong (Department of Anesthesiology, University of Pittsburgh Medical Center and School of Medicine, Pittsburgh PA, 15213, USA;Department of Anesthesiology, Xiangya Hospital of Central South University, Changsha, Hunan 410078, China)
Li-Ming Zhang (Department of Anesthesiology, University of Pittsburgh Medical Center and School of Medicine, Pittsburgh PA, 15213, USA)
Erin A. Sullivan, (Department of Anesthesiology, University of Pittsburgh Medical Center and School of Medicine, Pittsburgh PA, 15213, USA)
GUO Qu-lian (Department of Anesthesiology, Xiangya Hospital of Central South University, Changsha, Hunan 410078, China)
John P. Williams (Department of Anesthesiology, University of Pittsburgh Medical Center and School of Medicine, Pittsburgh PA, 15213, USA)Correspondence to:GUO Qu-lian,Department of Anesthesiology, Xiangya Hospital, Changsha, Hunan 410078, China (Tel: 86-731-84327413. Fax:86-731-84327413. E-mail:anjianxiong@ yahoo.com;email@example.com)
The protocol was approved by the University of Pittsburgh Institutional Review Board (IRB). We searched cases of cardiac arrest compiled over a period of 13 years from the University of Pittsburgh Medical Center (UPMC) between January 1, 1989 to December 31, 2001 using the MARS electronic searching system. Search keywords used were “anesthesia” and “cardiac arrest.” All searches were conducted by an expert in the MARS system at UPMC and a research fellow in the Department of Anesthesiology at University of Pittsburgh School of Medicine.
Cardiac arrest was defined as an event requiring cardiopulmonary resuscitation with either closed-chest cardiac compressions or open cardiac massage. Only cardiac arrests that occurred after anesthesia had been initiated in the operative suites were included. Events that occurred outside operating rooms, such as in trauma patients in the emergency room or postoperative patients in transport to the recovery room or intensive care unit, were not included. Cardiac arrests that occurred during cardiac procedures such as cardiopulmonary bypass, cardiac catheterization, or cardiac-related manipulation such as cardioversion or electroconvulsive therapy were also excluded. There were 23 patients with intraoperative cardiac arrests that met the criteria according to documented medical records, intraoperative anesthesia records, surgical postoperative notes, and pathological reports. The probable cause of arrest in most circumstances was noted in the surgical report. In cases when the cause was not clear from these records, the authors made a consensual decision after evaluating the original medical records including intraoperative anesthesia records, postoperative surgical notes, and sometimes pathological reports and patients’ outcome. The patients’ preoperative general information such as age, sex, American Society of Anesthesiologists (ASA) physical status (PS), and significant past medical histories such as cardiovascular or cerebrovascular disease, diabetes, end-stage organ failure, hypertension, sepsis, and systemic disease were evaluated in our study. Surgical procedures, type of anesthesia (general vs. regional), period of arrest, and methods and duration of resuscitation were collected as well. The anesthesia care team in each case was supervised by at least one board-certified attending anesthesiologist. Resuscitation was followed by the current guidelines for advanced cardiac life support (ACLS) led by the anesthesia team with assistance from surgical and nursing teams. The cardiac arrest time, the primary EKG rhythm at the time of arrest, and the duration from cardiac arrest to cardiopulmonary resuscitation (CPR) were evaluated. All the resuscitations lasted more than 30 minutes. Another CPR was undertaken if the arrest reoccurred after success in first resuscitation. Survival rate after CPR in the operating room was recorded to evaluate the correlation of success of CPR and relevant independent factors. The data was analyzed by the authors of this study, including a research fellow and anesthesiology faculty.
All characteristics were treated as categorical variables. A chi-square test was performed for the comparison of two values expressed as a percentage. SigmaPlot statistical software ver. 10.0 (Systat Software, Inc., San Jose, CA, USA) was used for statistical analysis and P values less than 0.05 were considered statistically significant.
General information regarding 23 patients experiencing intraoperative cardiac arrest
In the 13-year duration of the study, 218 274 anesthesia cases, including 203 382 general anesthesia cases, 12 282 spinal anesthesia cases and 2610 epidural anesthesia cases were administered. Intraoperative cardiac arrest was confirmed in 23 patients according to the exclusion criteria in the above methods. All 23 patients had been given general anesthesia, and there were 15 males and 8 females ranging in age from 20 to 89 years old (mean of 58 years). Overall incidence of intraoperative cardiac arrest from all causes was 1.05 per 10 000 anesthetics. Of the 23 patients who experienced an intraoperative cardiac arrest, 14 (60.8%) died in the operating room, leading to a mortality rate from all causes of 1.0 per 15 625 anesthetics. Among the 14 fatal cases, two patients experienced arrest twice; both of these patients responded well to initial resuscitation, but resuscitation efforts were unsuccessful upon a second cardiac arrest. Nine of the 23 patients (39%) who experienced an intraoperative cardiac arrest survived after successful resuscitation. Nearly all intraoperative cardiac arrests occurred during maintenance of anesthesia, except in one patient who experienced cardiac arrest during the period of anesthesia induction. This led to an intraoperative cardiac arrest rate due to anesthesia of 0.45 per 100 000 of all anesthetics. This female patient became markedly tachycardic (150 beats/min) after an induction dose of thiopental and enflurane. She was found in asystolic arrest shortly after giving 10 mg of esmolol intravenously. She recovered after successful resuscitation and her surgery was delayed to two months later, without any complications.
Characteristics of intraoperative cardiac arrest
Due to insufficient numbers of arrest cases, we could not perform multivariate analysis to identify independent factors influencing survival outcome. However, a remarkable percentage of patients (86.9%) who had intraoperative cardiac arrest were classified as ASA PS IV or V. Survival was significantly improved in patients with ASA PS I–III (66.7%) compared with those with ASA PS IV–V (35%). Of the 23 cardiac arrests, the survival rate was lower in patients older than 65 years (20%) compared to patients younger than 64 years old (53.5%). There was no difference between males and females in the survival rate. In addition, there were no significant differences in patients’ survival rates with trauma (50%) or emergency surgery (42%) compared with non-trauma (36%) or elective surgery (36%). Patients tended to have higher survival rates after cardiac arrests if CPR was performed in less than 5 minutes compared to more than 5 minutes. Asystole was the most common (52%) on initial EKG rhythm, and those patients seemed less likely to survive after resuscitation compared to other rhythms such as PEA and VF; however, due to limited case numbers this data did not show significant difference. Detailed data were shown in Table 1.
|Table 1. Characteristic profiles of intraoperative cardiac arrest|
Causes of intraoperative arrests
The causes of intraoperative arrests are listed in Table 2. Of 23 cardiac arrests, intraoperative hemorrhage was the most common cause (52.2%), with an overall survival rate of 50%. Other probable related factors were end-stage organ disease (21%), thromboembolic events (17%), cardiac origin (13%), sepsis (8.6%), and anesthesia (4.3%). Anesthesia was the least likely factor to cause intraoperative arrests and was associated with the highest survival rate after resuscitation. Surgical bleeding, thromboembolic events, and sepsis were among the worst in terms of survival outcome.
|Table 2. Causes of cardiac arrests|
Incidence of cardiac arrest
Intraoperative cardiac arrest is uncommon and it is difficult to design case-controlled prospective studies to investigate its etiologies. The overall incidence of cardiac arrest has been reported to range between 0.5 and 9.6 per 10 000 anesthetics.10 The most recent single center study3 reported that the incidence of cardiac arrest decreased over period of time (from 7.8 per 10 000 during 1990–1992 to 3.2 per 10 000 during 1992–2000, and to 2.5 per 10 000 in the year 2000), implying a significant improvement in patient care over this period. In our study, the incidence of overall intraoperative cardiac arrest was 1.1 per 10 000 anesthetics during 1989–2001. The incidence of cardiac arrest primarily attributed to anesthesia seemed to decline over time as well. In the 1950s, most intraoperative cardiac arrests and deaths were primarily related to anesthetic related complications and could be as high as 7.1 per 10 000 anesthetics.11 It has been reported that anesthetic-related cardiac arrest was 2.1 per 10 000 anesthetics between 1969 and 1978;10 1.0 per 10 000 anesthetics between 1979 and 1988;7 1.1 per 10 000 anesthetics from 1989–1995 in France,12 0.69 per 10 000 anesthetics from 1989 to 19992 and 0.5 per 10 000 anesthetics from 1990 to 2000.3 In our study, the incidence of anesthetic-related cardiac arrest was 0.05 per 10 000 anesthetics from 1989 to 2001 — the lowest yet reported. In addition, anesthesia-related cardiac arrests represented 4.3% (1/23) of all cardiac arrests in non-cardiac procedures in our study, which is similar to the finding of Girardi et al13 (4.2%, 1/24) but lower than the results of Newland et al2 (10.4%) and Sprung et al3 (10.8%). The causes of lower incidence of both intraoperative and anesthesia-related cardiac arrest in this study are unclear but may include the following: (1) we only reported cardiac arrests within the operative suites while other studies included the perioperative period, such as 24 hours of surgery and the intensive care unit after surgery.2,3,12-15 (2) we only reported the incidence of cardiac arrests in adult patients from 20 to 89 years old, while both Newland and Sprung’s studies2,3 involved patients of all ages, including pediatric patients. Zuercher and Ummenhofer1 reported that pediatric patients had higher incidence of cardiac arrest (1.4–4.6 per 10 000 anesthetics). Morray et al15 showed infants younger than 1 year of age accounted for 55% of anesthesia-related cardiac arrests in pediatric patients, indicating that pediatric patients, especially infants younger than 1 year of age, have an increased incidence of cardiac arrest.
Anesthesia-related mortality was defined as perioperative death to which human error on the part of the anesthesia provider had contributed. Lagasse et al16 collected data regarding perioperative mortality from 1966 to 2000 and identified 23 anesthesia-related mortality rates with wide variation ranging from 7.2 per 10 000 anesthetics to 0.12 per 10 000 anesthetics within the 35-year period. Such a wide variation may be attributed to different reporting sources and operational definitions as well as a lack of appropriate risk stratification, as suggested by Lagasse et al.16 However, several recent studies have shown much less variation in anesthesia-related mortality. For example, Arbous et al14 showed anesthesia-related deaths (within 24 hours period) of 1.4 per 10 000 anesthetics; Newland et al2 reported anesthesia-related mortality (within 24 hours) to be 0.55 per 10 000 anesthetics; Lagasse et al16 showed anesthesia-related mortality (within 48 hours) as 0.79 and 0.75 per 10 000 anesthetics in his studies; and finally Sprung et al3 estimated anesthesia-related mortality at approximately 1.0 per 10 000 anesthetics in their large population study. A study from French national mortality database showed the death rates related to anesthesia in 1999 were 0.69 in 100 000 anesthetic procedures. Taken together, it seems reasonable to estimate anesthesia-related mortality over the past ten years at approximately 0.8 per 10 000 anesthetics.17-19 In our study, no anesthesia-related deaths were found in 218 274 anesthetics, suggesting that anesthesia-related mortality was less than 0.05 per 10 000 anesthetics — the lowest anesthesia-related mortality that has been ever reported. However, as our data only limited to operative suites rather than within 24 hours or 48 hours after surgery, it is difficult to compare our data with other reported data. Interestingly, a latest study showed that although delayed defibrillation was associated with lower rates of survival after cardiac arrests in periprocedural areas, there was no association with survival for cardiac arrests in the operating room.20,21 Perhaps immediate simple CPR e.g. good quality of chest Compressions followed with Airway and Breathing (C-A-B) will increase survival rate as shown in the recent CPR guidelines by the American Heart Association.
Factors that decrease incidence of anesthesia-related cardiac arrest
The results from the study by Sprung et al3 demonstrated that the frequency of cardiac arrests for patients receiving general anesthesia decreased over time, although the frequency of arrest during regional anesthesia and monitored anesthesia care remained consistent. The only anesthesia-related cardiac arrest occurred early in our 13-year study period, which also suggests a trend of decreasing frequency. Several changes in anesthesia practice could explain the observed decline of the incidence of anesthesia-related cardiac arrest that may include: (1) the introduction of new methods and drugs effective in decreasing anesthetic cardiac arrests. For example, Zuercher and Ummenhofer1 found that the enhanced respiratory monitoring accounted for most of the observed decrease in anesthetic cardiac arrest rate due to respiratory causes. They further showed that there were no respiratory cardiac arrests occurred since 1984, when the pulse oximetry was used at clinical anesthesia. Eichhorn et al17 claimed that with the use of the same monitoring in more than one million anesthetics, the anesthesia-related arrests decreased almost 69% with the rate from 0.13/10 000 to 0.04/10 000 anesthetics. (2) the availability of trained anesthesiologists. The study by Olsson and Hallen9 demonstrated the inverse relationship between the incidence of cardiac arrest and the number of qualified anesthesiologists employed. (3) Human error. Human error seems one of major factors leading to anesthesia related cardiac arrest.2 Continued education of anesthesia practitioner and widely application of medical simulations could be useful for preventing human error in training of anesthesiology. At the UPMC, anesthesiology has implemented widely accepted guidelines on basic monitoring, conducted long-term analyses of closed malpractice claims, addressed fatigue of residents serving in-house call, developed patient simulators as meaningful training tools, and tackled problems of human error. Most importantly the profession has institutionalized safety in its scientific and governing bodies, creating the ASA’s Patient Safety and Risk Management Committee and the Anesthesia Patient Safety Foundation. We created the largest WISER Simulation Center in the United States with numerous simulation courses toward patient safety issues such as difficulty airway management, anesthesia crisis management, central venous cannulation training, advanced cardiac life support, etc. UPMC is the first major academic center in the United States adopting medical simulation competency evaluation and training as a component of hospital credentialing to address a significant patient safety and medical liability problem within its self-insured system. All these strategies and measures make UPMC as one of top medical institutes in the United States with its highest standards in quality patient care and patient safety.
Causes of cardiac arrest
Non-anesthetic intraoperative cardiac arrests
The incidence of intra- or perioperative cardiac arrest from non-anesthetic attributable causes was much higher than that of anesthesia-related causes in almost all previous studies, and most cardiac arrests were due to patient disease/condition or surgical factors.2,3,7-9 Girardi et al7 reported that cause of cardiac arrest among 162 661 non-cardiothoracic surgical procedures were due to cardiac (50%), hypoxic (25%), hemorrhagic events (16.7%) and sepsis (8.3%), respectively. Newland et al2 found that trauma including motor vehicle accidents, gunshot wounds, end stage liver disease, and its complications with liver transplantation were the most common causes for cardiac arrests accounting for 21.7% and 19.4%, respectively. Sprung et al3 found that 43.9% of cardiac arrests were related to cardiac events and 35.0% related to bleeding. In our study, the most common cause of intraoperative cardiac arrests was related to intraoperative bleeding (50%). The other causes found in our study were related to end-stage liver disease and its associated complications with liver transplantation (21.7%), a similar result to Newland study, thromboembolic events (17.4%), cardiac events (13%) and sepsis (15.4%) (Table 1). Our results may reflect hospital practice population. Since intraoperative hemorrhage accounted to 50% of intraoperative cardiac arrests, controlling intraoperative bleeding with improved new techniques sounds reasonable to reduce the incidence of intraoperative cardiac arrests. Using intraortic balloon occulusion technique20 has proved very effective in controlling intraoperative bleeding in certain surgical procedures although it has not been used clinically in the United States due to its potential complications such as end-organ ischemia and thrombosis.
Anesthesia-related cardiac arrests
Before 1984, the most common cause of anesthetic cardiac arrest was inadequate ventilation,8,9,18 followed by medication-related problems including overdose of an inhalation agent16 and post-succinylcholine asystole.9 Arrests due to ventilation problems decreased dramatically since 1984 because of the application of advanced respiratory monitoring.7 Studies have shown that medication-related problems have become the major cause of anesthetic cardiac arrests since 1989.2,11,14 Both Newland et al2 and Sprung et al3 found that perioperative anesthesia-related cardiac arrests were 10.4% and 10.8% of all cardiac arrests respectively, whereas 40%–54% were due to medication-related events and 20%–21% of cardiac arrests were caused by airway-related complications. Our study showed that anesthetic-related cardiac arrest was uncommon. Similar to our surgical colleagues’ study6 showing an intraoperative anesthetic- related cardiac arrest of 4.2% of all cardiac arrests, our investigation also found that only 4.3% of all cardiac arrests were a direct complication of an anesthesia- medication related event. We cannot definitively explain the difference of ours and Girardi’s with Sprung and Newland, but the study period and power of cardiac arrests numbers could account for the difference. Our study did agree with results from previous reports2,9,12 that most medication-related arrests were successfully resuscitated and went to complete recovery. The patient who experienced anesthesia-related arrest in our study was resuscitated without difficulty and she later underwent successful anesthesia and surgery.
Risk factors of cardiac arrest
In the current study, patients who experienced lower survival rates after cardiac arrest were of advanced age, had greater ASA physical status, and had emergency trauma surgical procedures. Although several factors such as age, sex, ASA physical status, emergency, and operation length and time have been proposed to associate with survival after cardiac arrest in a univariate analysis,2,9,12,15,18 no single factor was defined as a significant independent risk factor in multivariate analysis.3 Given such a small number of cardiac arrests in our current study, it was not possible to evaluate independent risk factors using multivariate analysis. We found that there was no significant difference in patient survival after cardiac arrests with patient age, sex, ASA physical status and emergency in a univariate analysis. Perhaps a combination of factors, such as in a patient with advanced age and high ASA physical status undergoing emergency abdominal and thoracic surgical procedures, may predict worse outcome after the arrests.
Disadvantage of this study
Though we attempted to study a large population of patients, the number of cases of intraoperative cardiac arrests was still not sufficient to perform statistical analysis to identify independent factors predicting survival outcome. In addition, data collected only within the operative suites limited comparison with other studies that collected data in a 24- or 48-hour period after surgery.
In summary, our study from a large US academic medical center has shown most intraoperative cardiac arrests were not due to anesthesia-related causes, and anesthesia- related cardiac arrest might have a higher survival rate compared to other probable causes of arrest. There was no single death primarily attributable to anesthesia between 1989 and 2001 (inclusive) in this study suggesting the mortality attributable to anesthesia was less than 1 in 200 000 anesthetics.
1. Zuercher M, Ummenhofer W. Cardiac arrest during anesthesia. Curr Opin Crit Care 2008; 14: 269-274.
2. Newland MC, Ellis SJ, Lydiatt CA, Peters KR, Tinker JH, Romberger DJ, et al. Anesthetic-related cardiac arrest and its mortality: a report covering 72 959 anesthetics over 10 years from a US teaching hospital. Anesthesiology 2002; 97: 108-115.
3. Sprung J, Warner ME, Contreras MG, Schroeder DR, Beighley CM, Wilson GA, et al. Predictors of survival following cardiac arrest in patients undergoing noncardiac surgery: a study of 518 294 patients at a tertiary referral center. Anesthesiology 2003; 99: 259-269.
4. Beecher HK, Todd DP. A study of the deaths associated with anesthesia and surgery: based on a study of 599548 anesthesias in ten institutions 1948–1952, inclusive. Ann Surg 1954; 140: 32-35.
5. Braz LG, Modolo NS, do Nascimento P Jr, Bruschi BA, Castiglia YM, Ganem EM, et al. Perioperative cardiac arrest: a study of 53 718 anaesthetics over 9 years from a Brazilian teaching hospital. Br J Anaesth 2006; 96: 569-575.
6. Kopp SL, Horlocker TT, Warner ME, Hebl JR, Vachon CA, Schroeder DR, et al. Cardiac arrest during neuraxial anesthesia: frequency and predisposing factors associated with survival. Anesth Analg 2005; 100: 855-865.
7. Girardi LN, Barie PS. Improved survival after intraoperative cardiac arrest in noncardiac surgical patients. Arch Surg 1995; 130: 15-18.
8. Keenan RL, Boyan CP. Decreasing frequency of anesthetic cardiac arrests. J Clin Anesth 1991; 3: 354-357.
9. Olsson GL, Hallen B. Cardiac arrest during anaesthesia. A computer-aided study in 250 543 anaesthetics. Acta Anaesthesiol Scand 1988; 32: 653-664.
10.Fleisher LA. Risk of anesthesia. In: Miller RD, ed. Miller’s Anesthesia. 5th ed. Philadelphia, PA: Elsevier; 2000: 806.
11.Briggs BD, Sheldon DB, Beecher HK. Cardiac arrest; study of a thirty-year period of operating room deaths at Massachusetts General Hospital, 1925–1954. J Am Med Assoc 1956; 160: 1439-1444.
12.Biboulet P, Aubas P, Dubourdieu J, Rubenovitch J, Capdevila X, d’Athis F. Fatal and non fatal cardiac arrests related to anesthesia. Can J Anaesth 2001; 48: 326-332.
13.Girardi LN, Barie PS. Improved survival after intraoperative cardiac arrest in noncardiac surgical patients. Arch Surg 1995; 130: 15-18.
14.Arbous MS, Grobbee DE, van Kleef JW, de Lange JJ, Spoormans HH, Touw P, et al. Mortality associated with anaesthesia: a qualitative analysis to identify risk factors. Anaesthesia 2001; 56: 1141-1153.
15.Morray JP, Geiduschek JM, Ramamoorthy C, Haberkern CM, Hackel A, Caplan RA, et al. Anesthesia-related cardiac arrest in children: initial findings of the pediatric perioperative cardiac arrest (POCA) registry. Anesthesiology 2000; 93: 6-14.
16.Lagasse RS. Anesthesia safety: model or myth? A review of the published literature and analysis of current original data. Anesthesiology 2002; 97: 1609-1617.
17.Eichhorn JH. Prevention of intraoperative anesthesia accidents and related severe injury through safety monitoring. Anesthesiology 1989; 70: 572-577.
18.Keenan RL, Boyan CP. Cardiac arrest due to anesthesia. A study of incidence and causes. JAMA 1985; 253: 2373-2377.
19.Lienhart A, Auroy Y, Pequignot F, Benhamou D, Warszawski J, Bovet M, et al. Survey of anesthesia-related mortality in France. Anesthesiology 2006; 105: 1087-1097.
20.Zhang L, Gong Q, Xiao H, Tu C, Liu J. Control of blood loss during sacral surgery by aortic ballon occlusion. Anesth Analg 2007; 105: 700-703.
21.Mhyre JM, Ramachandran SK, Kheterpal S, Morris M, Chan PS. Delayed time to defibrillation after intraoperative and periprocedural cardiac arrest. Anesthesiology 2010; 113: 782-793.