Circulation 83,045 • Volume 22, No. 4 • Winter 2007   Issue PDF

Safety Abstracts Stand Out at ASA

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

Over 2,200 abstracts were presented at the 2007 American Society of Anesthesiologist Annual Meeting in San Francisco, CA. As in previous years, a number of these abstracts examined issues directly related to patient safety. This brief review will highlight a few of the important abstracts discussed at the meeting.

Sedation Outside of the Operating Room

Propofol is commonly administered for sedation during upper and lower endoscopy. Two studies monitored sedation levels by utilizing 2 technologies (Bispectral [BIS] and Patient State Index [PSI]). In a study of 98 colonoscopy patients (A1014) monitored with a PSI, at least one airway intervention was required in 66.3% of patients, and 80% of interventions occurred at a Patient State Index (PSI) of <70. In a larger endoscopy study enrolling 202 patients (A44), the average BIS value during the procedure was 49.3. These studies suggest that propofol sedation during these procedures was associated with BIS/PSI levels commonly associated with general anesthesia and not moderate sedation, and that clinicians skilled in airway management should be immediately available if propofol is used. When monitored anesthesia care was delivered for endoscopic procedures by anesthesiologists, it appeared that no serious complications were noted among 2,766 high acuity cases (A497).

Risks of Non-Cardiac Surgery Following Coronary Stents

Previous small studies have observed a high incidence of adverse events following surgery after bare-metal (BMS) or drug-eluting stent (DES) placement. Two abstracts reported a relatively low frequency of adverse events in this patient population. A retrospective investigation from the Mayo Clinic (A798) examined data from 349 patients who underwent non-cardiac surgery within 2 years after placement of a DES. Perioperative ischemic events occurred in only 4.6% of patients, and 13% of patients required a red blood cell transfusion. Preliminary data were also presented from large multi-center international registry (POSTENT) established to examine the incidence of morbidity and mortality in surgical patients with previous stent placement (A193). In-stent thrombosis developed in 3.3% of 215 patients, and 4.7% of patients died within 60 days of surgery.

Adverse Events Associated with Aprotinin

The safety of aprotinin has been questioned in recent investigations. Several abstracts examined this issue. A review of a large registry of patients undergoing cardiac surgery at Duke (A240) revealed a greater rise in postoperative serum creatinine in patients receiving aprotinin, although the need for dialysis was not increased. A survival analysis of this same registry revealed a reduction in long-term survival in patients receiving aprotinin (A242). An investigation from the University of Tennessee compared 150 patients who received aprotinin to 150 historical controls who did not receive the drug (A243). No differences in any outcome measures were observed, with the exception of a higher incidence of renal dysfunction in the aprotinin group. These databases highlight the continued need for large randomized trials to assess the safety of aprotinin.

Anesthesia Workspace Contamination & Hand Hygiene

Several abstracts discussed the disappointing hand washing compliance among health care providers. First, it was identified that despite using standard disinfectants to clean in the morning, before the start of cases, and at the end of the operating day, there existed a remainder of >10 colony forming units (CFU) in approximately 52% of the inanimate sites tested (A1788). Types of organisms recovered include coagulase negative staphylococcus, beta and alpha hemolytic streptococcus, corynebacterium, and staphylococcus aureus. The author emphasized the need for sterile barrier techniques, aseptic medication administration, and frequent hand washing to mitigate the amount of contaminated areas. Three abstracts addressed the poor hand hygiene compliance of health care staff at varying points in their careers. Abstract A2140 examined the hand washing practices of 131 new interns when examining a standardized patient. Approximately 35% of the interns did not wash their hands prior to, and 95% did not wash their hands after, examination of the patient. Another abstract (A2141) examined hand-washing adherence among anesthesiology residents during their obstetric anesthesia rotation. During the first 2 weeks of the rotation, only 6.7% of observed epidural catheter placements were associated with proper hand hygiene prior to the procedures. This rate increased to almost 81% when the residents were given both explicit instructions on how to properly engage in hand hygiene and a handheld bottle of alcohol-based handrub. Similarly, A2139 developed a protocol for 90 second-year medical students to observe hand washing practices in several different ICUs. Hand washing compliance was found to be 30-35% among physicians and slightly better among nurses. Only 37% of the medical students involved reported that they would stop someone who had not washed their hands for fear of a poor grade by their superiors. This indicates that a cultural change is needed for improvement in quality care and hand hygiene compliance to curb the escalating amount of preventable nosocomial infections nationwide.

Diabetes and Insulin

Several posters examined the effect of diabetes and insulin therapy on outcomes following surgery. Investigators from Duke observed that preoperative hemoglobin A1c levels were predictive of postoperative acute kidney injury in both diabetic and non-diabetic cardiac surgery patients (A969). The same investigators noted that higher preoperative hemoglobin A1c levels were independently associated with increased mortality after primary cardiac surgery (A972). In a retrospective study from Belgium, cardiac surgical patients treated with a tight glucose control protocol (blood glucose 80-110 mg/dL) were compared to subjects in whom blood glucoses were maintained <150 mg/dL (A1209). Significant reductions in renal dysfunction, renal failure requiring dialysis, and in-hospital mortality were noted in the tight glucose control group. In other abstracts, intraoperative use of insulin was associated with a lower incidence of atrial fibrillation following cardiac surgery (A970) and trends toward reductions in troponin release in vascular surgical patients (A973).

Transfusions and Adverse Outcomes

The administration of PRBCs has been associated with an increase in morbidity and mortality in cardiac surgical patients. Two abstracts examined this important topic in other patient populations. Perioperative data from all patients undergoing hip fracture surgery over a 3-year period were analyzed (A1441). In propensity-matched patients, transfusion was a significant predictor of death (relative risk = 3.76). In contrast, transfusion of PRBCs was not independently associated with increased postoperative morbidity or mortality in patients following endovascular aortic repair (A1673). Another abstract discussed the changing tide of transfusion practices. Abstract A285 identified transfusion practices among 1000 transfusions in 2004 in Tunisia. Transfusion thresholds depended upon indication and included 6.16 g/dl (± 2.03) for urgent medical pathologies, 6.22 g/dl (± 1.6) for chronic medical pathologies, 7.74 g/dl (±2.49) for urgent surgical pathologies, 10.38 g/dl (± 2.2) for elective surgery, and 6.15 g/dl (± 2) for urgent obstetrical pathologies. These thresholds certainly suggest a turn toward a restrictive pattern of transfusion given the rising acknowledgment of complications associated with them.

Transfusion Requirements and Normothermia

As previously mentioned, blood transfusions might be associated with increased morbidity and mortality. Two abstracts performed meta-analyses examining 10 (A201) and 14 (A196) studies respectively involving hypothermia as it relates to transfusion requirements. Both studies demonstrated an inc

rease in transfusion requirements with hypothermia. Normothermia was associated with a 16% (A196) and 22% (A201) reduction in blood loss. One abstract studied risk factors for generating postoperative hypothermia in a retrospective fashion. In evaluating over 10,000 cases, abstract A186, found that the starting OR temperature, age, length of anesthesia care, BMI, and gender all played a role in temperature change perioperatively. Inexperienced trainees had a small effect on occurrence of post-operative hypothermia as well. Another abstract utilized Six Sigma methodology to evaluate rates of perioperative hypothermia (A496). Temperature evaluation and operating room temperature were identified as contributors to hypothermia in this abstract. By utilizing a temporal artery thermometer (TAT; Exergen Corp., MA) in the PACU as well as decreasing the range of temperatures that were possible in the laminar flow air handler in the operating room, this group improved their postoperative normothermia rates from 55% to 88% (A496). Perioperative normothermia continues to be an important quality improvement measure nationally.

Transfer of Care in the Operating Room (OR)

Two abstracts addressed the transfer of care of anesthesia staff during prolonged cases. Abstract A1782 examined 243,832 anesthesia cases and concluded that the incidence of adverse events increased with the number of anesthesia providers involved regardless of ASA physical status or case length. In fact, when comparing 1 attending/1 assistant vs. 2 attendings/2 assistants, the relative risk of adverse events was 0.53 (p<0.001). In another abstract (A1785), transfer of care to another provider was already identified as a potential patient hazard. Therefore, this group devised a 1.5-hour training session with 12 first-year anesthesia residents to educate them on effective handoff of care in the OR. The important characteristics of effective communication included up-to-date information exchange, eliminating distractions, and 2-way interactive dialogue that allowed for questions and verification of information. Video scenarios were also included in this training session. All 12 residents reported that this program met or exceeded their expectations and was helpful in improving patient safety. Outcome data on the utility of educational programs to curb miscommunication during handoff of care needs to be collected.

Obstructive Sleep Apnea

As the obesity epidemic in the United States continues to escalate, obstructive sleep apnea (OSA) remains a serious health concern that will affect the delivery of anesthesia worldwide. One abstract (A935) examined OSA and its relation to increased postoperative complications. Among 181 patients who tested positive for OSA by polysomnography, there was a statistically significant increase in postoperative complications (30% OSA vs. 15% Non-OSA). Respiratory complications were most common (23% OSA vs. 9% Non-OSA). Furthermore, OSA patients tended to require more therapy, including prolonged supplemental oxygen, additional monitoring, admission to intensive care unit, and re-admission within 30 days. Another abstract (A920) examined OSA patients and difficult mask ventilation and intubation. Adult patients scheduled for surgery completed a screening questionnaire from the Apnea Risk Evaluation System (ARES™). Patients who were identified as high-risk were asked to use the ARES™ Unicorder, a validated portable OSA diagnostic device. Patients were stratified into 3 OSA severity groups by their apnea-hypopnea indices (AHI): none, mild (AHI 0-20 events/hr), moderate (AHI 21-40), and severe (AHI >40). Information was collected by review of anesthetic records and included mask ventilation grades, laryngoscopic views, and ease of tracheal intubation. Those patients with severe OSA (AHI>40 events/hr) had an increased incidence of both difficult mask ventilation and intubation compared to patients with mild OSA. This indicates the potential for the severity of OSA to affect important airway practices executed by anesthesiologists.

This brief review summarized only a small number of the important abstracts on patient safety presented at the 2007 Annual Meeting. The abstracts referenced do not necessarily reflect the opinions of the authors or the APSF. 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 Evanston Northwestern Healthcare Department of Anesthesiology.