Volume 8, No. 4 • Winter 1993

ASA Scientific Presentations Showcase Safety

David M. Gaba, M.D.; Gerald L. Zeitlin, M.D.; Arthur I.L. Schneider, M.D.; Jeffrey B. Cooper, Ph.D.

Patient safety again figured prominently as a theme of the scientific sessions of the Annual Meeting of the American Society of Anesthesiologists October 9-13 in Washington, D.C. In six scientific sessions, there were 83 individual scientific presentations.

One session featured a frequent topic of discussion in the APSF Newsletter, pulmonary edema, which can occur as a complication of anesthesia and surgery without mismanagement of fluid administration. Dr. M. Maroof and colleagues from Riyadh, Saudi Arabia, reported on postoperative pulmonary edema in healthy (ASA I or 11), young patients having elective surgery. The cases were identified by the retrospective review of records. Excessive fluid administration was not found. The authors suggested that upper airway obstruction and negative intrathoracic pressure might have been involved in development of pulmonary edema.

Opinions, Pressures

Two papers discussed the practices of anesthesiologists in the complex world of medicine. Dr. K. Poterack from Wilford Hall Medical Center presented case vignettes involving controversial management issues such as the patient with an .open-eye’ but a full stomach. Respondents were asked to state whether the actions described were .acceptable’ or not, and whether they would ‘do this in your own practice.” They found that for six out of 11 vignettes there was a marked difference between what respondents felt was acceptable versus what they believed they would do.

Dr. S. Howard and associates from Stanford University and the Palo Alto VAMC surveyed California anesthesiologists concerning their attitudes about and experiences with production pressure in the work environment (a topic also discussed in several recent articles in the APSF Newsletter). Their survey controlled for selection bias of respondents. They found that many anesthesiologists reported internal pressures they put on themselves to please surgeons and avoid delays. They also reported overt external pressure from surgeons not to cancel cases, and to avoid delays. Nearly half of respondents had witnessed unsafe actions resulting from such pressures, although the effect on patient outcome is unknown. The authors suggested that the anesthesiology profession needs to confront these issues openly and find ways to prevent such pressures from causing preventable adverse outcomes for patients.

Lipid Vehicle Risk?

The underlying science related to potential risks of nosocomial infection from bacterial contamination in the lipid vehicle of propofol was studied by Dr. P. Langevin and associates from the University of Florida, Gainesville. They studied the distribution of bacteria in body tissues of rabbits inoculated with S. aureus suspended either in intralipid or buffer. The time course of bacterial distributor in blood did not differ between the groups, but distribution to other tissues was markedly different. The authors suggest that bacterial growth in Intralipid at room temperature is unlikely to explain nosocomial infection, but they propose that suspension in Intralipid micelles may allow bacteria to escape destruction in the rediculo endothelial system.

Pain Service Dangers

In a report from the Brigham and Women’s Hospital in Boston, Dr. H. Flanagan and his colleagues from an active acute pain service illustrated the concept of ‘total quality management’ while improving patient safety. They describe 10 critical incidents (Cls) resulting from the use of an ordinary intravenous infusion pump for continuous postoperative narcotic/local anesthetic infusions. Their poster illustrated dramatically how the physical features of the pump allowed the Cls to occur. A few examples include:

* family member increased the infusion rate;

* patient increased the flow rate to 10 times the infusion rate ordered;

* two instances when the pump was mistaken for an IV and an excessive infusion rate was set.

Other fascinating mishaps were: after a patient had undergone an MRI examination an epidural cassette was snapped into an IV pump -and vice versa; aminophylline was piggybacked into the epidural infusate tubing in another patient.

Modifications to the pump did not help. Detente pins to limit maximum flow rate were added and the pumps were painted bright orange. Human ingenuity and muscle power overcame these prophylactic maneuvers and there were two more critical incidents.

Flanagan et al. conclude by listing key design specifications in designing purpose-built epidural pumps:

* the drug reservoir must be enclosed in a locked transparent receptacle;

*software must control the maximum possible flow rate;

* software must prevent tampering with the rate setting;

* the pump cassette must be de-aired easily but not allow gravity flow;

* the infusion tubing must have no tempting injection ports;

* the infusion tubing must be colored or striped.

The data that Flanagan et al. gathered and analyzed was necessary to persuade their hospital’s administrators to purchase new equipment; this type of activity is clearly important given the current fiscal climate.

CVP Complications

Drs. R. Blackshear and N. Gravenstein of the University of Florida neatly demonstrated the necessary condition leading to the excess of superior vena cava perforations previously reported in the literature with left-sided central vein cannulation when compared to the use of the right side.

They reviewed retrospectively the records and chest x-rays of 200 patients. There were 50 in each of the following groups: right-sided internal jugular (IJ) and subclavian, and left-sided internal jugular and subclavian. They also limited the cases studied to those in which the tip of the catheter could be seen to he in the superior vena cava (SVC) on the chest x-ray. Most importantly, they measured the angle of incidence of the catheter to the wall of the SVC using the vertebral column as a zero degree reference point. In summary, they found significantly greater angles of incidence associated with left-sided insertions and in particular with left sided IJ insertions. Their observations led them to recommend avoidance of left-sided insertions whenever possible and the use of chest x-rays to assure that the catheter tip lies at an angle of incidence less than 40 degrees to the wall of the SVC.

Nerve Injuries

Drs. M. Werner and J. Martin of the Mayo Clinic performed a retrospective chart review of nearly 200,000 patients who underwent operations historically performed with the patient in the lithotomy position. They identified 55 patients with a persistent neuropathy of the lower extremities. They defined this as a motor neuropathy lasting more than three months. They identified three major risk factors: thin body habitus, smoking within 30 days of the procedure and lithotomy position maintained for more than four hours. Regional anesthesia was not associated with an increased risk of neuropathy. These observations might explain the findings of Kroll et al. from the ASA Closed Claim Study database, that only 28% of lumbosacral nerve root injuries leading to claims had an identifiable mechanism that was anesthesia-related.

In a return to last year’s theme of the safety of the anesthesia provider, Dr. M. Swanson and colleagues demonstrated that latex allergens can take off and become airborne in O.R. suites. The potential for greatest exposure occurs during glove manipulations and since anesthesia personnel change their gloves frequently (or are supposed to), their exposure is greater than that for other personnel.

Airway Ideas

Several papers relating to patient safety were presented at the session on Airways Complication and Management. Four papers related to laryngeal mask airways. Dr. D. Reinhart of the University of Utah began the session by comparing the effects of laryngeal mask airways (LMAS) and endotracheal tubes in a series of 36 propofol and thiopental anesthetized patients. Significantly less increase in systolic, mean and diastolic blood pressure, and heart rate were seen during airway manipulation with the LMA series. Also, significantly fewer patients in the LMA group had post anesthetic coughs or sore throats. No regurgitation was noted in either group although one LMA group patient developed laryngospasm. Dr. Reinhart suggested that LMAs may be useful in patients in whom cardiovascular stimulation should be avoided.

In a second report, Dr. Reinhart described a comparison of endotracheal tube and LMA placement by a group of paramedical personnel. A group of 10 paramedics, respiratory therapists, and nurses, all with some experience in endotracheal intubation, watched a 20 minute training film on LMA placement. Each recruit then attempted to place both an ETT and a LMA in an anesthetized patient. Three attempts were allowed before failure was declared. The LMA was placed successfully in each case on the first attempt in an average of 38 seconds. ETT was placed successfully in three (of 10) cases in an average of 237 seconds. Subjects scored LMA placement as being less difficult. Dr. Reinhart concluded that LMA may be an acceptable alternative to endotracheal intubation by paramedical personnel in difficult circumstances.

Dr. T. Owens and colleagues from St. Vincent’s Hospital in Dublin and the University of Texas at Galveston described their study in which 55 anesthetized patients were studied for evidence of gastroesophageal reflux during either face mask or LMA airway management. Reflux was assessed with small pH electrodes placed in the esophagus before induction and maintained in place during anesthesia. A significantly higher percentage of LMA patients was found to have episodes of pH < 4.0 in the distal esophagus, 57 percent with LMA and 22 percent with mask airways. Continuous monitoring of pH allowed the time of reflux to be recorded as well; LMA group reflux was found intraoperatively, during patient position change, and during recovery more often than was face mask group reflux.

A study on the resistance of the LMA to laser damage was presented by Drs. J. Pennant, N. Gajraj, and J. Miller from the University of Texas Southwestern Medical Center in Dallas. Because of previous reports of resistance of LMAs to low power tunable dye lasers, they investigated LMA resistance to the higher power carbon dioxide laser.

Instantaneous LMA shaft perforation and ignition were achieved with the laser set at 10 and 50 W. A one W beam required 60 seconds to perforate and ignite the LMA shaft. LMA cuff perforation was also achieved at all three settings, although it required 10 seconds at one W power. The silicone rubber LMA tubes produced less smoke and flaming than polyvinyl chloride tubes similarly tested. The authors felt that LMA could not be recommended for use in an unmodified form for laser treatment of the upper airway and that further studies on protecting LMAs from laser energy should be undertaken.

An analysis of airway management by anesthesiologists was presented by Drs. M. Mascia and M. Matjasko from the University of Maryland. They used a specially designed data collection form to collect information about the 613 emergency airway consultations done by their department from July 1991 until June 1992. Most patients, 61%, were males which might be explained by the fact that 23% of encounters took place in the emergency room, while 45% occurred in the ICU, and 6% in step down units. Acute respiratory failure was the most common reason for consultation while failed extubation, trauma, codes, and neurologic compromise were also frequently encountered. Complications were recorded in 18% of intubations. The complication rate of intubations done using drugs was significantly higher than the rate when intubation was accomplished in awake patients. Study data were useful in determining when junior residents might be given responsibility to attend codes alone.

Balloons and Bulbs

Two new devices were described. The first, a nasogastric balloon tube designed to prevent regurgitation and aspiration, was described by Drs. N. Roewer and J. Schulte am Esch from University Hospital Eppendorf, Hamburg, Germany. The tube, which resembles a nasogastric tube with a large balloon at the distal end, is inserted into the stomach prior to anesthesia, the balloon is inflated, and traction is applied to the tube. Three series of tests with the tube were described. In the first, anesthetized pigs were fitted with the tube and balloon while vigorous mechanical and chemical efforts were made to force the pigs to regurgitate. In the second series, volunteers ingested mineral water and were then fitted with tubes and inflated balloons and efforts were again made to induce regurgitation. Finally, a series of emergency patients was fitted with tubes and balloons prior to induction of anesthesia. In all cases, regurgitation was prevented.

The use of a self-inflating bulb in detecting esophageal intubation was further investigated and reported by Dr. Y. Waifai, and associates from the Illinois Masonic Medical Center in Chicago. The bulb, which now must be home-made, consists of an irrigating syringe bulb fitted to an adapter which allows it to be attached to the connector of an endotracheal tube. It has been shown that the bulb, if deflated and attached to a properly positioned endotracheal tube with the cuff inflated, will nearly always quickly reinflate when released and that it will nearly always not reinflate if the endotracheal tube is in the esophagus. The question was whether this was also true if the tube cuff was deflated. Endotracheal tubes were placed in both the esophagus and trachea in 30 anesthetized patients. All cuffs were deflated. In all cases the bulb instantaneously reinflated when attached to the tracheal tube and did not reinflate when attached to the esophageal tube. The authors conclude that the selfinflating bulb reliably distinguishes tracheal from esophageal intubations even when the tube cuff is deflated.

Dirty Handles

Drs. R. Morrell, D. Crews, D. Ririe, and R. James from Wake Forest University Medical Center reported on a study in which they looked for blood contamination of 38 laryngoscope handles and blades that had been routinely cleaned and returned to service. The handles and blades were rinsed with 10 ml of saline and two drops of the rinse were tested using a Hemoccult Sensa occult blood slide. Nineteen handles (50%) and four blades (10%) tested positively for occult blood. The authors did no bacterial or viral cultures but do suggest that disposable laryngoscope covers, disposable laryngoscopes, or improved cleaning protocols may become necessary to address this apparent problem of equipment contamination.

Lost Tubes

A paper entitled “Anesthesia and Unplanned Extubation” was presented by Drs. S. Shields, B. Harrison and R. MacKenzie of the Mayo Clinic. These authors reviewed their hospital data files and identified 76 patients (or 227,699) in whom an unplanned extubation had occurred during surgery from 1985 to 1992. Forty-three of these extubations occurred in pediatric patients (1:411), and 33 in adults (1:6,364). Twelve percent of extubations occurred during tonsillectomies and a total of 46% during head and neck surgery. A variety of methods had been used to secure the tubes. The study points out the need for extra vigilance in securing and monitoring endotracheal tubes in head and neck surgery, particularly when the patient is a child.

Dr. A. Ovassapian, et al. of the Northwestern University Medical School and the Lakeside VA Medical Center in Chicago presented a report on the length of the upper airway in adult male patients. Fiberoptic bronchoscopy was used to measure the distance from the incisors to the vocal cords and from the incisors to the carina in 30 anesthetized patients. Attempts were made to correlate these measurements with patient age, weight, and external body measurements including distance from the tip of xiphoid to cricoid ring, length of the middle finger, and mandibular circumference. Although there was statistical correlation between age and the bronchoscope measurements, the authors concluded that true airway lengths were poorly predicted by the external measurements and that placement of endotracheal tubes to a fixed depth should be discouraged.

Intubation Predictions

The problem of predicting difficulty of endotracheal intubation was addressed in a study by Drs. S. Keramati, M. Lewis, and J. Benumof of the University of California in San Diego. Oropharyngeal class (I pillars and uvula, II uvula, III base of uvula and IV soft palate alone) was determined preoperatively in a variety of neck positions with and without phonation in 213 adult patients. The distance between the larynx and mandible was also measured in these patients in a similar variety of conditions in order to estimate mandibular space. The best prediction of subsequent difficult intubation using oropharyngeal class was found when patients were examined sitting, phonating and with neck and tongue extended. The best mandibular space prediction came from measuring from the inside of the mandible to either the hyoid bone or the thyroid cartilage in patients in the sniffing or neck extended position. These measurements allowed predicting difficult intubation with about the sensitivity of other published studies. The authors emphasize that other factors such as neck flexibility and soft tissue mobility also markedly affect the difficulty of intubation.

Drs. S. Samra and F. Guinto from the University of Michigan gave a progress report on an APSF sponsored study in which MRI scans and x-rays of 20 patients who had been difficult intubations were examined in an effort to find physical identifiers of difficult intubation. All subjects, and 20 matched controls, had lateral MRI scans of the neck and two lateral x-rays of the neck, one exposed for bone, the other for soft tissue. None of the skeletal measurements revealed a significant difference between the two groups. Dr. Samra reported that a seemingly positive correlation between MRI soft tissue measurements and difficult laryngoscopy had not been up to repeated examination. While examination of the films continues, a reliable soft tissue or skeletal indicator of the difficulty of intubation has not yet been identified in this study.

A general theme emerged from a set of presentations in the patient safety poster/discussion session: Clinicians have behaviors that may be less than optimally safe; changing those behaviors isn’t easy, but it can be done. Patient behavior can also be altered. Several posters demonstrated the many ways in which realistic simulators are now being used to measure and alter behavior.

Dr. R. Botney and colleagues from Stanford studied ‘fixation error,” using a set of videotapes taken during a course in Anesthesia Crisis Resource Management at another institution. Subjects had been presented two different, simultaneous events during the training, which used the realistic anesthesia simulator developed by the Stanford group. One of the problems was a simple failure of a capnometer; the other the more serious case of a fully opened vaporizer, concealed from view by printout from the NIBP monitor. Although most of the 18 subjects eventually discovered the more serious problem, five did not. The study demonstrates that fixation error is a real phenomena that can be measured.

In another study using the same group and simulator environment, Dr. Botney described how subjects reacted to a simulated, unexpected loss of oxygen pipeline pressure. Review of the taped sessions revealed the kinds of deficiencies that occur when reacting to a critical event. The responses of the subjects varied considerably; in general they did not have a well-formulated plan. Fortunately, most managed to ventilate the ‘patient’ adequately, although sub-optimally. Few instituted the most appropriate measures. The difficulty of urgently replacing an oxygen cylinder was apparent.

Dr. V. Chopra and colleagues at the University Hospital Leiden in the Netherlands, using their realistic simulator environment, demonstrated that practicing in a simulator seems to be effective at improving performance. Using a 3-phase experimental design that incorporated incidents of either malignant hyperthermia(MH) or anaphylactic shock(AS), the study measured changes in performance according to a standardized scale. Both groups had simulator training in AS, but only one group had simulator training in MH. Here again, the value of simulation was illustrated for measuring and, in this case, improving performance. The trainees with simulator training in MH appeared to perform better on that event than did the control group.

Regarding the design of simulators, Dr. W. van Meurs, representing the group that developed the Gainesville Anesthesia Simulator, described the physiologic model that has now been incorporated into the system and which drives the responses of the manikin. The model is capable of mimicking most aspects of the manikin’s physiology, thus allowing for a large assortment of event simulations. He explained how such model-driven simulators have great advantage over ‘script” driven systems. That ” is the trend in realistic simulator design was clear: the two other realistic simulators represented in this group of abstracts also have a model-controlled design philosophy.

Those who experience training in a realistic simulator generally seem to perceive that the experience is valuable and that most clinicians should be provided such training periodically. These findings were from post-course evaluations administered by a group from the Harvard Medical School Anesthesia Departments who instructed subjects in ACRM using the CASE 2.0 simulator, which had been transported to their institution. Dr. R. Holzman presented the study, which also demonstrated that the simulator environment is transferable to a venue outside of the one in which it was developed.

Several abstracts addressed aspects of resident training using techniques other than realistic simulation. Drs. M. Olympio and M. Goldstein of Wake Forest University illustrated how feedback of videotaped performance can be effective in improving performance and altering practice. Using the FDA recommended guidelines for anesthesia machine check-out and other ASA-developed practice guidelines, the study compared performance of two groups of residents-one with and one without feedback about their performance in the form of observation and critique of a videotape of their activities during an actual anesthetic. The group with feedback showed substantial improvement over the control group.

Dr. T. Cutter and others at the University of Chicago measured the factors that affect residents’ perceptions of the value of their educational experience. Interactions between attendings and residents were observed during 130 procedures and residents were surveyed for their perception of the value of various characteristics of their teachers and the educational environment. It appears that what an attending ‘says’ is perceived as being more valuable than what an attending ‘does’. Dr. Cutter remarked during the discussion presentation that residents most appreciated being challenged by their teachers to consider ‘what if’ situations.

Dr. J. Tome, presenting for herself and Dr. 1. Cohen of the Children’s Hospital of Pittsburgh, reported success at teaching third-year medical students skills in mask airway technique. This was despite surprising deficiencies in ability and understanding of airway management prior to the course even though all had taken Basic Life Support and introductory airway skills courses. The study also demonstrated that clinical skills can be measured via OSCES, Objective Structured Clinical Examinations.

Dr. W. Foster of the University of Texas Medical Center, Dallas, reported that patients remember more when instructed by a videotape of preoperative information versus having the same kinds of preoperative information given by an anesthesiologist and supplemented by written material. The study suggests that an audiovisual program may be effective in improving compliance with preoperative instructions, although specific study of that question was not attempted.

Finally, Dr. R. Prielipp reported on the remarkably low utilization of the esophageal stethoscope by most anesthetists who were observed in four different medical centers. fin 520 recorded observations, overall listening frequency was 28%. Only CRNA students seemed to listen to the stethoscope with regularity (75% of observations). Dr. Prielipp and his colleagues shared their concern over the substitution of other forms of monitoring for the esophageal stethoscope.

Dr. Gaba is from Stanford University; Dr. Zeitlin is from the Brigham and Women’s Hospital, Harvard Medical School; Dr. Schneider is at Hershey Medical Center, Hershey, PA; and Dr. Cooper is at the Massachusetts General Hospital and Harvard.