Anesthesia Safety: Is It a Model or a Myth?
Anesthesia has been touted as being one of the safest specialties in medicine. In 1999, the Institute of Medicine (IOM) published a report on medical errors in U.S. hospitals, which noted that anesthesiology had made substantial improvements in patient safety. One impetus for reducing medical errors in the 1970s and the 1980s was the soaring cost of medical malpractice. Anesthesiologists responded by establishing national practice standards for patient monitoring, deliberately analyzing adverse events, improving the safety of anesthesia machines, fostering the widespread adoption of new technologies (e.g., pulse oximetry), improving provider training in crisis event management, and creating an independent foundation whose sole purpose was to advance anesthesia patient safety (the Anesthesia Patient Safety Foundation or APSF). Deaths in anesthesia have decreased from 2 deaths per 10,000 anesthetic procedures in the 1980s to about 1 death per 200,000 to 300,000 in 2000. These numbers have been validated by surveys conducted in the Netherlands, France, and Australia. A 2003 report from the Center for Quality Improvement and Patient Safety of the Agency for Healthcare Research and Quality (AHRQ) found 1369 complications from anesthesia in 1,933,085 patients at risk of 0.71 per 1,000 discharges. This rate compares favorably to other rates of hospital complications as shown in Table 1–1 . Not all anesthesiologists believe that anesthesiology is as safe as these data suggest—even in young, healthy patients. For instance, in 2002, Lagasse published an extensive review of the literature in which he concluded that anesthesiology mortality was still in the range of 1 in 10,000. Similar contention has been made by other authors. It can be very difficult to separate errors or mishaps in anesthesia from surgical mishaps or patient disease. In studies reviewed by Lagasse, the definitions for death in which anesthesia was “associated,“ “related,” “contributory,” or “preventable” varied widely as did the time windows for defining the perioperative period (24 hours to 30 days). Many of these studies had small numbers and involved single healthcare sites. An alternative way of understanding patient safety in anesthesia is to study “opportunities for error.” This more probabilistic approach focuses on events and their likelihood of causing patient harm. A key advantage of an event, rather than injury focus, is that data can be collected prospectively and the analysis is less likely to be affected by hindsight or outcome bias. For example, Weinger and colleagues introduced the concept of “nonroutine events” (or NRE) and showed that NRE, which represent any deviation in optimal care, occurred in 25% to 35% of all anesthetics in three different academic medical centers. Moreover, an NRE data collection system captured seven times more patient injuries than a traditional anesthesia quality assurance reporting system. They concluded that anesthesiology is complex and errors still occur, resulting in poor patient outcomes.
| Patient Safety Indicators | No. of Events | Risk Pool | Rate per 1000 Discharges at Risk | Match Rate % |
|---|---|---|---|---|
| Accidental puncture or laceration | 11,810 | 5628 112 | 3.32 | 75 |
| Birth trauma, injury to neonate | 4740 | 720 021 | 6.53 | 96 |
| Complications of anesthesia | 1369 | 1933 085 | 0.71 | 74 |
| Decubitus ulcer | 41,440 | 1932 676 | 21.51 | 56 |
| Foreign body left during procedure | 536 | 6572 845 | 0.09 | 69 |
| Iatrogenic pneumothorax | 3919 | 5861 689 | 0.67 | 66 |
| Obstetric trauma, cesarean birth | 1138 | 191 227 | 6.97 | 99 |
| Obstetric trauma, vaginal birth with instrumentation | 12,518 | 51,225 | 224.21 | 95 |
| Obstetric trauma, vaginal birth without instrumentation | 51,223 | 591 752 | 86.61 | 99 |
| Postoperative hemorrhage or hematoma | 3494 | 1695 495 | 2.06 | 69 |
| Postoperative hip fracture | 1068 | 1397 898 | 0.77 | 51 |
| Postoperative physiological and metabolic derangement | 799 | 801 702 | 1 | 44 |
| Postoperative pulmonary embolism or deep vein thrombosis | 15,704 | 1689 662 | 9.34 | 61 |
| Postoperative respiratory failure | 2275 | 633 855 | 3.58 | 37 |
| Postoperative sepsis | 2592 | 229 853 | 11.25 | 33 |
| Postoperative wound dehiscence | 843 | 411 099 | 2.05 | 55 |
| Selected infection resulting from medical care | 11,449 | 5752 102 | 1.99 | 63 |
| Transfusion reaction | 30 | 6572 845 | 0.004 | 80 |
Role of Equipment in Anesthesia Safety
In 1978 Cooper et al applied the critical incident technique first described by Flanagan to understand anesthesia incidents. A critical incident was defined as: a human error or equipment failure that could have led (if not discovered or corrected in time) or did lead to an undesirable outcome, ranging from increased length of stay to death. In this study, 139 anesthesiologists were interviewed and 1089 preventable incidents were reported. Seventy incidents were deemed a critical event with a substantial negative outcome. They reported that 30% of critical incidents reported by clinicians were related to equipment problems. Nineteen percent were reported instantly, while 11% were reported retrospectively. Twenty-eight percent of these demonstrated inadequate knowledge or familiarity with specific equipment or use of a relatively new technique or device.
The American Society of Anesthesiologists (ASA) Closed Claims Project (CCP) was initiated in 1985 to collect information about anesthesia-related adverse outcomes. A total of 8496 closed insurance claims have been collected and analyzed. An analysis in 1997 found that only 2% of the claims were related to equipment issues ( Figure 1–1 ). Death or brain damage occurred in 76% of these cases. Misuse of equipment was judged to have occurred in 75% of the cases and true equipment failure in 24%. Overall, 78% of claims were deemed preventable by appropriate use of monitoring. Subsequent studies have called into question the validity of this finding since the reviewers were not blinded to case outcome.
Subsequent studies in anesthesia showing similar results are summarized in ( Table 1–2 ).
| Authors | Year | Type of Study Design | No. of Cases | % Equipment-Related |
|---|---|---|---|---|
| Cooper JB et al | 1978 | Retrospective critical incident reporting | 359 | 14% as a result of failure |
| Craig and Wilson | 1981 | Retrospective critical incident reporting | 81 | 12% related to failure |
| Cooper JB et al | 1984 | Retrospective critical incident reporting | 1089 | 11% as a result of failure |
| Also instant reporting | 239 | 19% | ||
| Keenan and Boyan | 1985 | All anesthesia-related cardiac arrests | 27 | |
| Utting JE | 1987 | Deaths or cerebral injury cases reported to medical defense unit of UK | 1501 | 28% of technique errors |
| Kumar V et al | 1988 | Voluntary QA reporting | 129 | 19% related to failure |
| Cheney FW et al | 1989 | ASA closed claims study analysis | 869 lawsuits | ? |
| Chopra V et al | 1992 | Voluntary QA reporting | 549 | 21% related to failure |
| Caplan RA et al | 1997 | ASA closed claims study of gas delivery equipment claims | 72 | 24% |
| Weinger et al | 2007 | Prospective videotaped anesthetics | 407 | 45% as judged by at least two expert reviewers of the actual videotape; usability or failure was considered to be a contributory factor in a nonroutine anesthesia event |
Similar studies have been conducted in other countries. The Australian Incident Monitoring Study issued results on 2000 critical incidents in 1993. One hundred and seventy-seven (9%) were due to equipment problems and of these 107 (60%) were due to failures of the anesthesia gas delivery system. The National Reporting and Learning System database from the United Kingdom reported similar results. Of 12,606 incidents reported to the National Patient Safety Agency, 13% were related to equipment failures. Of those incidents, 81% caused little or no harm, 18% produced moderate harm, and only 1.2% resulted in severe harm or death.
As of May 2009, the ASA Closed Claims Database had 71 claims of a total 2945 claims from 1995 to 2003 due to problems with anesthesiology equipment. Eighteen of these claims were for gas delivery equipment. These included one anesthesia machine problem (unspecified), five vaporizer problems, three ventilator problems, and four breathing circuit problems. There were another five claims involving supplemental oxygen equipment or other devices attached to the patient’s endotracheal tube. There were an additional two claims involving malfunctioning Ambu bags. Most equipment problem claims resulted in temporary or nondisabling injuries (66%). There were 8 (11%) permanent and disabling injuries and 16 (23%) deaths. Payment was made in 52 (73%) of these claims, with a median payment (in 2007 inflation adjusted dollars) of $137,525 (range $2720 to $2,825,750). Over time, gas delivery problems appear to be decreasing as a proportion of total claims. These types of incidents represented 3% of all claims in the 1970s, 2% in the 1980s, and 1% in the period 1990 to 2003.
In Canada, medical device problems reported to the Health Protection Branch were studied to determine the problems associated with anesthesia devices. While only 2.3% of new devices were classified as anesthesia devices, these devices produced 8.6% of problem reports and 37.5% of alerts. The percentages of recalls and problem reports were also higher in the anesthesia (10.2%) than in other (4.9%) devices.
Why Is It Important to Know Your Equipment If the Mortality Has Decreased Tenfold in the Last Decade?
Although the most recent gas delivery system closed claim was for an event in 2003, equipment failures have been reported in the literature after that and even in 2008. New equipment continues to be introduced and errors reported. If the goal of the APSF, “no patient will be harmed during anesthesia,” is to be realized, then any error occurring due to equipment misuse or malfunction is unacceptable. Safety can plateau or even diminish without constant effort at improvement. Included in Table 1–3 are recent examples in the literature of equipment problems.
|
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
