Scientific Papers Highlight Patient Safety at the 2012 ASA Annual Meeting

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

Over 1,500 abstracts were presented at the 2012 American Society of Anesthesiologists Annual Meeting in Washington, DC. As in previous years, a number of these abstracts examined issues directly related to patient safety. This brief review will highlight several abstracts discussed at the meeting.

Database Studies and Perioperative Complications

Several investigator groups utilized database information to determine factors associated with perioperative morbidity and mortality. Bauer et al. from the University of Michigan attempted to define the incidence and risk factors associated with severe maternal sepsis by reviewing a cohort of 2,758 deliveries complicated by maternal sepsis (BOCO8). Over the 10-year study period, the incidence of severe sepsis increased significantly from 17.1% to 30.3%. In addition, mortality increased during the same time period from 2.2% to 4.9%. Factors associated with severe sepsis included chronic renal insufficiency, chronic liver disease, stillbirth, retained products of conception, cesarean delivery, hypertensive diseases of pregnancy, chronic heart failure, cerclage during pregnancy, and preterm delivery (<37 weeks) (BOCO8). Investigators from the Cleveland Clinic (A015) reviewed electronic records from 31,148 ASA 3-4 adults undergoing non-cardiac surgery under general anesthesia to investigate the effect of etomidate induction on mortality and hospital length of stay. Propensity scoring was used to account for some confounding factors (i.e., ASA status, Charlson comorbidity score, and emergency surgery) between 2,143 patients given etomidate and 5,231 patients given propofol. Using multivariable logistic or Cox proportional hazard regression, the authors reported a 2.3-fold increase in 30-day mortality postoperatively in patients given etomidate compared with those administered propofol. Furthermore, patients who received etomidate were 18% less likely to be discharged at anytime in the postoperative period (A015). Sampat et al. from the University of Chicago (A1173) used the Nationwide Inpatient Sample (NIS) to examine the incidences of postoperative visual loss (POVL) and corneal abrasions in patients undergoing either robotic assisted (RAP) or open radical prostatectomy (OP). A total of 136,711 surgical cases were reviewed over a 10-year period; the overall incidences of POVL and corneal abrasions were 0.22% and 0.15%, respectively. Rates of POVL and corneal abrasions increased nearly 10-fold during the years of 2000-2009, which corresponded to the period of time when the robotic approach became the predominant surgical technique. The rate of total POVL and corneal abrasions increased with RAP when compared to OP (A1173).

Two database studies focused on outcomes associated with duration of red blood cell (RBC) storage. Gazmuri et al. from the Cleveland Clinic reviewed data from 86,483 patients undergoing a variety of general surgery cases. These investigators found no increased risk of postoperative mortality due to increased mean storage duration of RBCs (A073). Contrary to this study, other investigators from the same institution examined the effect of prolonged RBC storage on clinical outcomes in patients undergoing orthotopic liver transplantation (OLT) (A505). In a sample of 915 patients, those OLT recipients who received older blood (≥15 days) had an increased risk of either graft failure or mortality (HR (95% CI): 1.47 (1.02, 2.11), when compared to those patients receiving younger blood (<15 days) (A505). The age of blood may affect different populations in different manners.

Closed Claims Database

Several studies used the Closed Claims database to investigate the incidence and risk factors for perioperative complications. Mehta et al. from the University of Washington reviewed 9,536 closed claims to examine patient injuries from anesthesia gas delivery equipment (A1072). Anesthesia gas delivery claims decreased from 4% of claims in the 1970s to 1% from 2000-2010. The specific claims in the later era included vaporizer problems (N-13), breathing circuit problems (N-10), anesthesia machine problems (N-7), ventilator problems (N-5), and supplemental oxygen line events (N-4). Payments in the later era also went down substantially (approximately a $600,000 reduction in median payment). The same group used the closed claims database (N-9,536) to compare burn injuries from warming devices during 1995-2010 vs. those that occurred during 1970-1994 (A1079). Both periods had the same 1% incidence of burn injuries from warming devices. The most common cause of burn injury from forced air warming devices was use of the hose without the appropriate blanket attachment (A1079). The buttocks, legs, and axilla were the most common areas burned by warming devices or materials. Payments for new burns were higher in the newer era when compared to the older era (p<0.01). Esmail et al. from the University of Washington reviewed the closed claims for airway injuries that occurred in 1995 or later (A1081). The most common airway injury was esophageal perforation, which accounted for 24% of airway injury claims. Other sites reported included vocal cord or laryngeal injury, tracheal tear, and pharyngeal injury. The primary causes of esophageal injury were due to difficult intubation or esophageal equipment (e.g., transesophageal echocardiography probe or esophageal dilators/anvils for gastric surgery). Forty-three percent of patients with esophageal perforation had preexisting esophageal pathology. Similar to previous analyses, death from esophageal perforation occurred in 19% of cases. Esophageal injury related payments were significantly higher than payments for other airway injury claims (median payment-$117,900). The same institution used the closed claims database to compare central venous catheter (CVC) injuries between 1995-2009 and 1970-1994 (A1075). Fifty-nine percent of CVC injury related claims resulted in either death or permanent brain damage and this incidence did not differ between time periods. Complications related to access increased significantly over time (63% in 1970-1994 vs. 87% in 1995-2009). Carotid cannulation/puncture was the most common complication in both time periods and increased in the later time period to 24% from 14% in the earlier period. During 1995-2009, 50% of CVC claims were evaluated to be potentially preventable by using ultrasound and 41% of the claims could have been prevented by using pressure wave form monitoring. Sixty-one percent of CVC claims resulted in payment, but there was no difference in payment amounts between the 2 time periods (A1075).

Anesthesia Type and Outcomes

Two abstracts focused on whether anesthesia type affects perioperative outcomes. Fisicaro et al. from Jefferson Medical College investigated the decision making process involved in selecting general anesthesia (GA) or monitored anesthesia care (MAC) for patients undergoing endoscopic retrograde cholangiopancreatography (ERCP) (A502). Of the 4,727 ERCP cases evaluated, 38% of patients received GA. Factors associated with patients undergoing GA for ERCP were increasing BMI (>30 kg/m2), emergency case status, higher ASA physical status, increasing anesthesia experience for ERCPs by the supervising anesthesiologist, and cases started after 12 pm. There was a 1.7% conversion rate from MAC to GA. Airway events accounted for 50% of the changes in anesthetic technique, and full stomach considerations accounted for 30% of the changes in anesthetic technique (A502). Case duration was significantly longer with the GA technique when compared to MAC. Cata et al. from MD Anderson Cancer Center investigated anesthesia factors that may contribute to longer term survival and recurrence free survival (RFS) in patients undergoing surgery for non-small cell lung cancer (A743). Among the 204-patient cohort, the long-term survival was 63% and the RFS was 53%. Anesthesia time longer than 270 minutes was significantly associated with a lower RFS. Other variables associated with poor long term survival or RFS included age >65 years, BMI <25 kg/m2, ASA 3 & 4, Stage II-III cancer, current smoking status, and COPD. Perioperative blood transfusion was not associated with poor oncological outcomes (A743). Further studies are warranted to validate this association of prolonged anesthesia time and poor RFS.

DeLiT Trial Results

The DeLiT trial is a clinical investigation examining the impact of glycemic control, steroid use, and depth of anesthesia on major outcomes after non-cardiac surgery. Abdelmalak et al. at the Cleveland Clinic enrolled 381 patients in a study utilizing a 3-way factorial design to investigate 3 important interventions (intravenous dexamethasone vs. placebo, intensive (80-110 mg/dl) vs. conventional glucose (180-200 mg/dl) control, and lighter vs. deeper anesthesia) on major morbidity (BOC05). In an abstract presenting data on glycemic control, the authors reported that patients randomized to receive intensive control did not have a reduction in major morbidity. In addition, no severe hypoglycemic episodes (<40 mg/dl) were noted. The authors also reported their findings from the study regarding the effect of corticosteroids and depth of anesthesia on acute postoperative pain (A510). In this trial, neither light anesthesia nor dexamethasone was associated with improved pain scores or a reduction in opioid consumption (A510). Data from the DeLiT trial examining the effect of steroids on perioperative inflammation (assessed by measuring plasma hsCRP levels) and clinical outcomes after non-cardiac surgery were also presented (A745). Both the mean hsCRP and change in hsCRP levels were significantly lower in patients receiving dexamethasone versus placebo on postoperative days 1 and 2 (A745). While steroid administration did not have a direct effect on major morbidity, changes in hsCRP from baseline to the maximum value on postoperative days 1 and 2 were associated with an increase in major morbidity. Another arm of the DeLiT trial reported the effect of depth of anesthesia on outcomes following non-cardiac surgery (A1200). The median BIS values were greater in the lighter anesthesia group when compared to the deeper anesthesia group (51 vs. 43, respectively). The anesthetic depth had no effect on major morbidity. There was also no association between the incidence of any major morbidity and median patient BIS or percent of time spent under deep anesthesia. Overall, the DeLiT trial did not show significant outcome benefits with the use of tight intraoperative glycemic control, steroid use, and lighter anesthesia.

Monitoring For Consciousness and Respiratory Function

Several abstracts this year focused on appropriate monitoring strategies for patients undergoing sedation procedures or recovering in the PACU. Authors from the University Medical Center, Utrecht, Netherlands, performed an open, stratified, randomized controlled trial in 427 healthy adult women during outpatient gynecological procedures (A584). Patients undergoing propofol sedation were randomized to standard respiratory monitoring or standard monitoring plus capnography. Results suggested that there was no difference in the incidence of hypoxemic episodes between the 2 groups. However, the number of airway interventions (authors did not specify the types of interventions) performed was significantly increased in the capnography cohort (A584). Another study (A768) examined whether a TSE mask (face tent mask) was more efficient than high nasal cannula oxygen flow in reducing severe desaturation events in patients undergoing deep propofol sedation during upper GI endoscopy. Two cohorts of patients were evaluated, those with a TSE mask (N=171) and those with a nasal cannula only (N=64). The data demonstrated that the TSE mask was more effective than nasal cannula high oxygen flow in reducing severe desaturation events requiring bag-mask ventilation (A768). Mestek et al. from Boulder, Colorado, developed an algorithm (RRoxi) to derive respiratory rate from the photoplehthysmogram signal (i.e., pulse oximetry) and compared the accuracy of this modality with a reference measurement of respiratory rate from capnography in 12 patients in the PACU that underwent a variety of elective surgeries (A094). The agreement between RRoxi and the reference was R2=0.89. Pulse oximetry derived respiratory rate provided continuous measurements during 97% of the monitoring period. Further studies are required to investigate alternative methods to reduce critical desaturation events during and after anesthesia sedation cases.

Factors Related to Nosocomial Infections

Punj et al. from the All India Institute of Medical Sciences screened a total of 325 mobile phones of health care workers (HCWs) for bacteria (A593). This study showed that 94.5% of the mobile phones screened had evidence of bacterial contamination. Gram negative strains were isolated from 31.3% of the mobile phones, while 52% of the mobile phones were found to have staph aureus. This study also reported that only 6% of HCWs disinfected their phones. Another study implemented a program to improve the compliance of hand washing among HCWs in a pediatric operating room (A592). Preliminary data were collected and a subsequent education presentation was provided on recommendations from the Centers for Disease Control (CDC) and the World Health Organization (WHO). There was a significant increase in hand washing compliance in the post-education era from 61% to 73%. The authors suggest that further steps are required in order to achieve a targeted compliance of 90%, such as increasing the number of alcohol dispensers in the operating room and auditing the functionality of the dispensers on a regular basis.

Vasopressors and Cerebral Oxygenation

Two abstracts investigated the effect of using vasopressors on cerebral oxygenation measured by cerebral oximetry. Kalmer et al. from the University Medical Centre Groningen, Groningen, Netherlands, performed a double-blinded, randomized controlled trial of 60 patients undergoing general anesthesia for ophthalmic surgeries to investigate the effect of atropine, norepinephrine, and phenylephrine on mean arterial pressure (MAP), heart rate (HR), cardiac output (CO), and cerebral tissue oxygenation (SctO2) (A840). When the MAP dropped below 90 mmHg, either norepinephrine or phenylephrine was administered randomly. When the HR dropped below 60 bpm, patients were treated with atropine. Results suggested that while both phenylephrine and norepinephrine increased MAP, phenylephrine decreased both CO and SctO2. Norepinephrine preserved CO, but did decrease SctO2 to a lesser degree than phenylephrine. Lastly, atropine administration was associated with an increase in MAP, CO, and SctO2 (A840). Allen et al. from Duke University performed an observational study in 14 parturients undergoing spinal anesthesia for cesarean delivery (A134). The authors compared the effects of a phenylephrine bolus versus a prophylactic infusion of phenylephrine for the treatment of hypotension (to maintain MAP within 20% of baseline values) on cerebral tissue oxygenation (SctO2). Cerebral tissue oxygenation decreased significantly in both groups over time. The phenylephrine infusion group experienced a greater reduction in SctO2 measurements than the bolus group, presumably due to the higher doses used and the longer administration time (A134). These findings suggest that, although phenylephrine increases systemic blood pressure, cerebral oxygenation may be compromised by this therapy.

Miscellaneous

Several abstracts addressed issues related to thrombosis and embolism. Glick et al. from the University of Chicago investigated the incidence of postoperative deep venous thrombosis (DVT) in the lower extremities following elective surgery and the associated perioperative risk factors (A597). The DVT incidence was 1.3% among all 231 patients included and 4.5% among patients undergoing orthopedic surgeries. The authors suggested that the following risk factors may increase DVT incidence: longer hospital stays, periods of immobility after surgery and orthopedic procedures. The low overall DVT incidence may reflect the implementation of perioperative measures to reduce DVTs, although DVT prophylaxis was given to only 54% of patients (A597). Ono et al. from Fukuyama City Hospital, Fukuyama, Japan, investigated the effects of continuing and discontinuing aspirin on hemorrhagic and thrombotic risks in 498 patients who underwent open urological cancer surgery (A495). The main finding was that the risk of perioperative transfusion was not increased by preoperative aspirin continuation. However, the incidence of thromboembolic events increased by more than 10-fold in those patients who discontinued aspirin preoperatively. Jun Kim et al. examined the incidence of venous air embolism (VAE) in 100 ASA I parturients undergoing cesarean section under general anesthesia (A130). The presence of air emboli was evaluated by 2 cardiac anesthesiologists who used intraoperative transesophageal echocardiography. The observed incidence of VAE was 94%. Those women who received uterus externalization had a higher grade VAE than those without uterine manipulation.

Another study investigated the relationship between the length of oral intake restriction and circulating blood volume by using the stroke volume variation (SVV) method (A367). Ninety-seven patients undergoing either otolaryngological or breast surgery were enrolled. Patients were randomly assigned to either restriction of clear liquid intake for 2 (short period group, SPG) or 4 (long period group, LPG) hours prior to surgery. There was no significant difference in the amount of clear liquid intake between groups. Both groups had dehydration as indicated by the estimated SVV. However, the degree of dehydration was significantly greater in the LPG when compared to the SPG. This study suggests that even a 2-hour clear liquid restriction is associated with dehydration.

As in previous years, a series of abstracts discussed trends in techniques for intubation and extubation. Bellmore et al. from the Mayo Clinic investigated the impact of the introduction of video laryngoscopy (VL) on the utilization of fiberoptic bronchoscopy (FOI) to secure airways (A306). The authors utilized the Mayo Clinic database to compare pre-VL airway management techniques (N-10,176) with post-VL ones (N-12,617). During the 5-year study period, VL use increased significantly from 0% to 8.65%, while FOI decreased from 2.79% to 0.97%. The use of direct laryngoscopy also decreased significantly (from 82.24% to 72.88%), while the use of LMAs increased significantly from 9.59% to 11.52%. Another observational study of 44 patients with difficult airways described experiences with extubating this patient population with an airway exchange catheter (AEC) combined with the cuff leak test (A770). With this technique, there were no reintubations required. Oxygen saturations were ≥95% in all patients post-extubation. The AEC remained in all patients <4 hours post-extubation. Thirty-eight out of 48 patients tolerated the AEC without cough or discomfort, while the other 10 patients experienced a mild cough or discomfort. Of the 44 patients who received a cuff leak test, the test was positive in all patients. The authors concluded that the AEC coupled with the cuff leak test may enhance the safety of extubation of the difficult airway (A770).

This brief review summarized only a small number of abstracts on patient safety presented at the 2012 Annual Meeting. This is not an endorsement of the methods, results, or conclusions of any particular abstract. 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. Greenberg is Director of Critical Care Services, Evanston Hospital and Co-Director for Resident Education Department of Anesthesia NorthShore University HealthSystem and Clinical Assistant Professor, Department of Anesthesiology Critical Care University of Chicago, Pritzker School of Medicine.

Dr. Vender is the Harris Family Foundation Chairman Department of Anesthesia / Critical Care Services and Vice President, Physician & Programmatic Development at NorthShore University HealthSystem and Clinical Professor Anesthesiology University of Chicago Pritzker School of Medicine.

Dr. Murphy is the Director of Cardiovascular Anesthesia at NorthShore University HealthSystem and Clinical Associate Professor University of Chicago Pritzker School of Medicine.