Safety Abstracts Abound at 2006 ASA

Steven B. Greenberg, MD, FCCM; Glenn S. Murphy, MD; Jeffery S. Vender, MD, FCCM

Over 1000 abstracts were presented at the 2006 American Society of Anesthesiologist Annual Meeting in Chicago, IL. This brief review will summarize a few of the key abstracts related to patient safety.

Sedation in Non-Operating Room Settings

The safety of administering sedation (primarily propofol) outside of the operating room was one focal point addressed in this year’s abstracts. One retrospective trial (A1695) examined 74 pediatric patients undergoing sedation (propofol or ketamine) for a variety of procedures in the emergency room. Wide variability in dosing of sedatives was recognized due to different physician practices. Hallucinations, hypoxemia, and inappropriate arousal were found in 13.5% of patients. The author concluded that complications associated with providing sedation in the emergency room are not inconsequential, and that standardization of administration needs to take place. Another study (A1584) looked at 40 patients undergoing colonoscopies with anesthesiologists delivering propofol. Airway interventions (>2 interventions per patient) were required in 70% of patients. This study suggests that, even in experienced hands, airway interventions are still required. A retrospective study (A1691) compared nurse administered sedation with physician-administered sedation in pediatrics. Although physicians managed more critically ill patients, nurse administered sedation was associated with increased PACU length of stay and prolonged, deeper sedation times.

Prediction of sedation failure was examined in at least 2 abstracts. In a review of over 39,000 pediatric sedation encounters (A354), the presence of an anesthesiologist substantially reduced sedation failures. Abstract A1395 was an observational study of sedation in 50 non-randomized adults undergoing upper endoscopic ultrasound under videotape scrutiny. Standard care (opioid and benzodiazepine) given by an endoscopist resulted in less effective and efficient sedation compared to when an anesthesiologist administered propofol. These studies mention the need for anesthesiologists providing sedation in these environments.

A new automated propofol delivery system was assessed during GI endoscopy cases (A1586). Forty-eight patients undergoing colonoscopies or upper endoscopies had propofol titrated to a desired clinical effect. An electronic data acquiring system checked vital signs, alarms, respirations, and altered propofol dosing. Three patients experienced significant desaturation (<90 %) and 18 subjects had apnea spells>30 seconds. Despite the limitations of the technology, the automated system’s response resulted in recovery of all subjects to normal respiratory parameters without an anesthesiologist’s intervention. Proponents of this technology argue that it may mitigate the need for high physician costs, while reducing complications and improving safety in these non-standardized settings. However, the relatively small number of subjects studied, and the significant incidence of desaturation and apnea raise valid concerns.


The U.S. is experiencing an obesity epidemic with an associated high incidence of undiagnosed patients with obstructive sleep apnea. Several abstracts focused on addressing this crucial topic facing anesthesiologists. In a retrospective trial (A989) of 116,035 subjects, patients who were overweight or obese were at a significantly increased risk of airway difficulty, failed intubation, and reintubation (P<0.05). Another large retrospective study (A992) involving approximately 19,000 patients noted an association between an increased occurrence of perioperative events when comparing obese patients to non-obese patients (P=0.003). Obese patients may indeed suffer from a higher incidence of perioperative complications and, therefore, require ICU admission. Another retrospective study (A987) involving 248 patients undergoing gastric bypass devised a scoring system to predict which obese patients would need an ICU admission. The following categories: BMI, comorbidities, performance status, pulmonary status, and age were assessed and given a score of 0, 1, 2, or 3 based on increasing risk. Although sensitivity was low (41%), a score of >=10 gave an adequate specificity of 88% to predict ICU admission. This scoring device may aid anesthesiologists in predicting which patients may need ICU care, thereby, minimizing inadequate or overuse of limited resources.

In addition to obesity, the diagnosis of obstructive sleep apnea (OSA) and the correlation with difficult intubation was also examined. A prospective trial (A988) was performed involving 1,898 patients over 4 months. Twenty-four percent of these patients were found to be at high risk for OSA based on screening. These patients took an ARES Unicorder home (a device that detects RDI or the number of abnormal breathing events per valid hour of recording time and indicates the severity of OSA). Seventy-six percent of the patients who were in this group and were not already diagnosed with OSA were newly diagnosed with OSA with this model. Study A986 required anesthesiologists to identify 68 patients with difficult intubations and refer them for a sleep study to rule out OSA. Of the 20 patients who agreed to take the sleep study test, 55% of the patients were diagnosed with OSA. This suggests that a difficult intubation may be another harbinger for the diagnosis of obstructive sleep apnea.

Antibiotic Administration/ Contamination of Objects in the OR

Prevention of perioperative infection is a critical patient safety issue. Analysis of 3,623 procedures in relation to antibiotic administration was performed (A140). Appropriate administration within 1 hour of incision occurred in 92% of procedures in which the antibiotics were given in the OR as opposed to 59.8% of the time prior to entering the operating room. This suggests that the optimal administration of antibiotics for patients is most appropriately performed in the operating room.

Two abstracts investigate microbiological growth on objects handled by an anesthesiologist in the operating room. The first study (A944), cultured 5 different objects (oxygen flowmeter on the anesthesia machine, OR bed control, ventilator valve, telephone handle, and the right hand door handle inside the OR) in three different OR settings (trauma, general, and cardiac surgery). Although the most prevalent organism was coagulase negative staphylococcus, alpha streptococcus and E. coli were also found. These more pathogenic bugs were most commonly identified in the cardiac surgical operating rooms. This study also noted that more organisms were found in the morning, suggesting that inexperienced or inadequate cleaning personnel may be a factor in promoting pathogenic bacterial overgrowth during the preceding night. Another trial (A948) examined bacterial contamination of computer keyboards and mouse pads in the OR and ICU. By utilizing ultraviolet lamps, it was evident that bacterial growth on these objects existed. Most cultures identified coagulase negative staphylococcus. However, 3 keyboards in the OR as well as 1 keyboard in the ICU were contaminated with MRSA. Ethyl alcohol swabs significantly reduced the overall bacterial load detected. This abstract is a reminder that anesthesiologists should remove their gloves before touching objects and the importance of washing their hands to reduce the risk of spreading infectious organisms to patients.


Airway management is one subject that defines our profession. A large prospective (A1687) trial involving 2,837 adult patients scheduled for general anesthesia with tracheal intubation attempted to identify risk factors for difficult mask ventilation. Following logistic regression, contributing factors included Mallampati >II, age, presence of beard, and history of neck radiation. Impossible mask ventilation was associated with oxygen desaturation in 22% of cases and an increase in difficult intubation. Another study (A555) attempted to establish the incidence of difficul
t mask ventilation (MV) in 5,434 pediatric general anesthesia cases. The incidence of difficult or impossible mask ventilation was 0.2%. Although BMI was not shown to be a predictor of difficult mask ventilation, a grade 2-3 (requiring oral airway, inadequate MV, or requiring 2 providers) MV was associated with difficult intubation in ages 2-5 years. This abstract is the first step in identifying the incidence of difficult mask ventilation in a pediatric population. Further studies may look at predictable risk factors for difficult pediatric mask ventilation.

Aprotinin’s safety profile has recently been called into question. A multi-center prospective double blind randomized trial (A383) involving 352 patients attempted to address adverse events associated with aprotinin’s use during total hip arthroplasty. The study reported no differences in postoperative laboratory values (serum creatinine or BUN values) or adverse events (such as deaths, cardiac failure, renal failure, TIA, DVT, or PE) when comparing the aprotinin (APR) and placebo groups. This small study suggests a safe profile for aprotinin during total hip arthroplasty. Another prospective study (A384) evaluated the presence and severity of allergic reactions associated with the use of aprotinin in 1,307 orthopedic surgical patients. The overall allergic reaction rate was 1.3%. However, there was a significantly higher allergic reaction rate in those patients with repeated administration of aprotinin within 3 months of previous aprotinin treatment (P<0.0001).

This brief review highlighted only a small number of the many important abstracts on patient safety presented at the 2006 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

Drs. Greenberg, Murphy, and Vender are affiliated with the Department of Anesthesiology at Evanston Northwestern Healthcare, Evanston, IL. Drs. Murphy and Vender are also members of the APSF Editorial Board.