by Glenn S. Murphy, MD, and Jeffery S. Vender, MD

Nearly 1,500 scientific papers were presented at the 2005 American Society of Anesthesiologists Annual Meeting in Atlanta, Georgia. Important studies relating to patient safety were discussed in many of the 94 separate sessions. This review will highlight a few of these important presentations.

Intraoperative Awareness

Several abstracts examined the incidences and consequences of intraoperative awareness. The Thai Anesthesia Incidence Study (A-1283) prospectively collected data from 20 hospitals over a 1-year period. Details of intraoperative awareness were recorded and analyzed to identify contributing factors and preventive strategies. Among over 150,000 anesthetics, 99 cases of awareness were observed. Awareness was noted more frequently in certain patient populations (female gender, ASA I and II patients, patients undergoing cardiac, obstetric, and lower abdominal surgery). Although 50% of patients reported experiencing pain during the awareness episode, only 13% described postoperative emotional stress or anxiety. Pollard and colleagues (A-8) conducted a modified Brice interview in postoperative patients (within 24-48 hours) to determine the incidence of intraoperative awareness. A total of 161,824 general anesthesia patients were interviewed over a 4-year period. Only 12 cases of intraoperative awareness (0.007% incidence) were identified in this investigation, which suggests that the risk of awareness may be lower in certain practice settings. Researchers in Sweden (A-7) examined the severity of immediate and delayed suffering due to intraoperative awareness. After interviewing 2,681 consecutive patients scheduled for general anesthesia, 98 patients were identified who considered themselves to have experienced awareness. Detailed interviews were conducted in the 46 patients who appeared to have actually been aware during a general anesthetic. Thirty patients described an acute emotional reaction, and 15 patients experienced late symptoms with a median severity score of 4 (on a scale of 12). Four patients contacted medical personnel due to mental symptoms relating to awareness. However, only 1 patient was diagnosed with post-traumatic stress disorder. Leslie et al. (A-9) presented details of awareness cases in the B-Aware Trial. Patients with confirmed awareness in this trial were more likely to have preoperative impaired cardiovascular status and intraoperative hypotension requiring vasopressor treatment than patients without awareness. In addition, these patients received lower concentrations of inhaled volatile anesthetics (MAC equivalent of 0.3%). Six of 13 patients reported adverse consequences resulting from the awareness episode. These findings clearly demonstrate that hemodynamically unstable patients are at greatest risk of awareness, and measures to reduce the risk of this complication in this high-risk population must be considered.

Pediatrics

Three abstracts from the Pediatric Sedation Research Consortium (PSRC) were presented (A-1312, A-1312, A-1314). The PSRC is a collaborative group of 24 institutions organized to examine the safety of pediatric sedation practices and practitioners. A web-based data collection tool was used to collect data on all sedation procedures performed at each institution. Data were received on 10,552 procedures. Complications were defined as any of the following; apnea, desaturation, unplanned mask ventilation or intubation, prolonged sedation or unplanned deep sedation, emesis, use of reversal agents, or change in vital signs >30%. The incidence of complications was lowest when sedation was provided by an anesthesiologist (2.6%) when compared to other clinicians (nurse/physician assistant 4.7%; ER physician 7.0%; intensivist 7.0%; pediatrician 8.7%; radiologist 5.7%). Adjusting complication rates for age, ASA status, and emergency status increased these differences further. The results from the PSRC suggest that serious complications related to pediatric sedation are rare, and that the risk of complications may be influenced by the type of provider administering sedative agents.

The Pediatric Perioperative Cardiac Arrest (POCA) Registry was formed in 1994 to investigate causes and outcomes associated with perioperative cardiac arrest in children. Researchers from the University of Washington School of Medicine examined the causes of cardiac arrest in pediatric patients over 2 time periods: 1994-1997 and 1998-2003 (A-1310). There was a decrease in the proportion of infants and an increase in the proportion of older children (6-18 years) suffering a perioperative cardiac arrest over time. The severity of injury during the 2 time periods did not differ, with more than one-quarter of cardiac arrests resulting in death. The proportion of medication-related deaths was significantly lower in the 1998-2003 period (20%) compared to the 1994-1997 interval (32%). The authors attribute this difference to the declining use of halothane in favor of sevoflurane in pediatric patients.

Jimenez and colleagues analyzed 525 pediatric claims from the ASA Closed Claims database to identify trends in types of patient injury and outcomes over the last 3 decades (A-1309). The proportion of claims relating to respiratory events decreased over time (1970s - 57%; 1990-2000 - 25%; P < 0.001), as well as the proportion of claims for death or permanent brain damage (1970s - 78%; 1990-2000 - 61%; P=0.03). Although the reasons for these changes in pediatric claims over time are not established, the authors hypothesize that improvements in monitoring, drugs, or training (subspecialization) may have influenced patient outcomes.

Airway

The incidence and predictors of difficult mask ventilation have been poorly understood. Kheterpal et al. (A-1415) examined the level of ease or difficulty of mask ventilation during 15,923 general anesthetics over a 1-year period. The ability to mask ventilate was graded on a 4-point scale (1: ventilation without the need for an oral airway, 2: ventilation requiring an oral airway, 3: ventilation that was difficult, inadequate, or required 2 providers, or 4: impossible ventilation). Grade 3 ventilation was observed in 214 (1.3%) patients and Grade 4 occurred in 24 (0.16%) cases. Independent predictors of Grade 3 ventilation included a history of snoring, a BMI >25, limited jaw protrusion, the presence of a beard, and an ASA status of 3-5. Predictors of Grade 4 ventilation were the presence of a neck mass or sleep apnea. These data suggest the predictors for difficult intubation and ventilation may differ.
Two studies by Mort and colleagues investigated emergency airway management of the obese patient outside the operating room setting. An emergency intubation database was reviewed to examine the incidence of difficult intubation (A-1145), and the role of the LMA as a rescue device (A-288), in this patient population. When compared to a cohort of patients with a BMI <25, morbidly obese (MO) patients (BMI >40) had a higher incidence of poor glottic visualization (Lehane-Cormack grade 4: 28%-MO vs. 8%), unsuccessful intubation on first attempt (48%-MO vs. 28%), and requiring more than 3 intubation attempts (18% vs. 10%). The incidence of hypoxemia was also higher in the MO group. The authors also reported on the use of the LMA as a rescue device in obese patients (BMI >30) outside the OR. A total of 97% of obese patients were successfully ventilated with the LMA within 3 attempts at placement. Overall, 91% of patients were successfully intubated via the LMA. The authors note that obese patients pose challenges to the airway manager outside the OR, and that the LMA may be a useful device to establish both ventilation and intubation in these situations.

Miscellaneous

The impact of type of anesthesia (inhalational vs. intravenous) on outcomes remains controversial. Investigators from the Netherlands studied the association between method of general anesthesia (propofol or volatile agents) and 1-year mortality (A-1271). Adult patients (n = 1,508) undergoing general or vascular surgery were examined. Multivariate logistic regression analysis was used to adjust for potential confounding variables. Overall 1-year mortality was 5.2%. Mortality was associated with comorbidities, age, and surgical procedure, but not with type of anesthesia. The authors note that inhalational agents were more frequently used in older patients with comorbidities, which may explain previous observations of higher mortalities in patients administered volatile agents.

Ischemic optic neuropathy (ION) is the most common cause of postoperative visual loss (POVL) after spine surgery. Lee and colleagues examined the ASA POVL Registry to identify potential risk factors for POVL (A-1). Seventy-one cases of ION following spine surgery were reviewed and compared to 9 cases of central retinal artery occlusion. Of the 71 ION cases, the median age was 50, mean anesthesia duration was 10 hours, the mean estimated blood loss was 3.8 liters, and 79% of cases had ≥15 minutes of a systolic blood pressure <100 mmHg. Cases of central retinal artery occlusion were associated with a shorter anesthesia duration (6.4 hours) and a lower estimated blood loss compared to ION cases. Although ION may occur during prolonged spine surgery with significant blood loss and hypotension, the wide ranges in these reported variables suggest a multi-factorial etiology of ION.
This brief review summarized only a small number of the important abstracts on patient safety presented at the 2005 Annual Meeting. To view other abstracts on patient safety, or to obtain further information on the abstracts discussed in this review, please visit the Anesthesiology website at www.anesthesiology.org.

Dr. Murphy is Director of Cardiac Anesthesia for Evanston Northwestern Healthcare and an Associate Professor at Northwestern University Medical School in Chicago. Dr. Vender is Chairman of the Department of Anesthesia at Evanston Northwestern Healthcare and Professor at Northwestern University Medical School in Chicago.