Safety Persists as ASA Meeting Theme

Robert S. Holzman, M.D.; Susan L. Polk, M.D.; Donald R. Webre, M.D.; Gerald L. Zeitlin, M.D.

Anesthesia patient safety and related topics were featured in 59 scientific presentations in four sessions at the October American Society of Anesthesiologists’ Annual Meeting in San Francisco. In the Patient Safety and Education Section of the meeting, there mere two half-* oral paper presentations, one poster session, and one session of the new poster-discussion format, which was moderated by Ellison C. Pierce, Jr., M.D., President of the APSF, with discussion by Frederick Cheney, M.D., Chairman of the ASA Committee on Professional Liability and principal investigator of the ASA closed claims study.

Summarized here are selected papers from these four sessions. Several other presentations, including some from other sessions, also were relevant to patient safety. In all, patient safety persists as one of the major themes of the submitted presentations at this meeting.

Prospective Pulse Oximeter Study

Dr. J. Moller and associates from Denmark reported further findings from what will probably be the only prospective outcome study comparing patients given anesthesia with or without monitoring of oxygen saturation. This study is supported by the Anesthesia Patient Safety Foundation and included as co-authors Drs. J. Cooper and J. Gravenstein of the APSF Executive Committee This was a randomized study involving more than 20,802 patients, observed in both the operating room and in the PACU. As might be expected, there was a 19-fold increase in diagnosed hypoxemia in the group in which oximeters were used as opposed to the group in which they were not supporting Comroe’s observations many years ago that observation of ‘cyanosis’ is a virtually useless method of trying to estimate hypoxemia. In the OR, cardiovascular events-were observed in 7. 8 percent of both groups, but myocardial ischemia defined as angina or ST depression was noted in 12 of the patients with oximetry and 26 of those without. This unexpected observation suggests an association between pulse oximetry, diagnosis of hypoxemia, and related myocardial ischemia that requires more study. In all, 15 predefined complications in the period from the day of surgery until the seventh postoperative day were studied. One or more postoperative complications were found in 5.2 percent of the population. Pneumonia, pulmonary edema and atelectasis were seen in 1.3 percent, 0.2 want and 0.7 percent of patients respectively. Cardiac failure and myocardial infarction were seen in 0.5 percent and 0.1 percent respectively. Cerebral stroke and pulmonary embolism were seen in 0.1 percent of patients. Most of the cardiac and pulmonary complications occurred in the first three days, peaking on the second postoperative day.

In a second paper, also presented by Dr. Moller, early postanesthesia outcome was examined in the same patient population, continuing the random assignment to pulse oximetry or nonpulse oximetry monitoring groups. Specially trained individuals recorded the presence of predefined complications from the first to the seventh postoperative day without knowledge of pulse oximetry use during anesthetic care. One or more postoperative complications were found in 10 percent of the pulse oximetry and 9.4 percent of the nonpulse oximetry patients. There was no statistically significant difference between respiratory, cardiovascular, neurologic, or infectious complications between groups. This finding of no difference in postoperative complications and perioperative outcome stands in interesting contrast to pulse oximetry’s contribution to increased speed of intervention in adverse intraoperative events.

Choice of Technique

Dr. R. Sorensen and colleagues from the University of Utah used the technique of meta-analysis to try to decide an old but important question in anesthesia whether regional or general anesthesia affects morbidity and mortality. Meta-analysis uses statistical methods to systematically combine evidence from existing studies. They searched Medline for all the relevant studies since 1966 and only included randomized controlled clinical trials. The effect of the method of anesthesia used was summarized as the risk difference, which is the absolute difference in frequency of outcome between control and treatment groups. The authors found little difference in outcome when they compared death rates, cardiovascular, pulmonary and neurological complications, pulmonary embolus and nausea and vomiting. However there was a 2 3 percent greater incidence of deep vein thrombosis in patients who received general anesthesia.

The authors stated that this study was only intended to give a provisional answer to the question until a similar prospective study can be done.

Dr. N. Sharrock from the Hospital for Special Surgery in New York reviewed a ten-year experience in changing over from general to regional anesthesia for total joint arthroplasties. From 1981-1985, during which all total joint arthroplasties underwent general anesthesia, 28 of 5,394 patients (0.52 percent) died. From 1987-1990, the era of increased regional techniques 10 of 6,635 (0.15 percent) died. While it is possible that anesthetic technique made a contribution to the decrease in mortality, the author acknowledged that concurrent improvements in patient monitoring technology, increased use of invasive monitoring, the introduction of postoperative epidural analgesia, and changes in perioperative care acuity (postoperative intensive care, staffing changes and an enhanced quality assurance program) may have also contributed.

Modifiers and Outcome

Dr. D. Cullen from the Massachusetts General Hospital and Harvard Medical School presented a paper on the relationship of the physical status (ASAPS) and/or patient age to length of hospital stay, postoperative complications, and post discharge physician visits following total hip replacement, transurethral prostate resection, and cholecystectomy in 1,099 patients. For total hip replacement, both age and ASA-PS correlated with increasing length of stay, postoperative complications, and post discharge physician visits, with age the most important factor. For transurethral prostate resection, length of stay increased with age, but even more so with physical status. Postoperative complications and physician visits increased significantly with physical status, but not with age. For cholecystectomy, length of stay increased with both age and physical status, while complications and physician visits increased significantly only with physical status. It was suggested by Dr. Cullen that surveillance of this type of data is of particular importance when the Resource-Based Relative Value Scale will be implemented, as care for sicker and older patients may be under compensated through the averaging system, and that age and physical status are pragmatic means of stratifying patients for more equitable reimbursement.

Outcomes of Clinical Practice

In another installment of the ASA Closed Claim Project database, patterns of eye injury associated with anesthesia were reported by Dr. W. Gild of the University of Washington and his collaborators. Seventy-one claims for eye injury (three percent of the total database) consisted of eye injuries, with most of these cases (61 percent) resulting in blindness. Half of these claims were for anterior chamber injuries, and half for posterior chamber. Patient movement during anesthesia was the mechanism of injury in 30 percent of the cases, with 16 of the claims occurring during general anesthesia, and five during monitored anesthesia cam AU of the movement claims resulted in permanent injury, and payments were made to the plaintiff in 88 percent. The median payment for movement-related claims was $108,000 (compared to $9,000 for nonmovement claims). It is of particular interest for anesthetic plan formulation that muscle relaxants were not used in more than half of the claims for movement under anesthesia, and peripheral nerve stimulators were not used in any. It was concluded that patient movement during general anesthesia for ophthalmic surgery is a significant source of eye injury claims, that these claims incorporate a high severity of injury, are generally determined to follow substandard anesthetic care (by peer review), and are characterized by high payments. It was suggested that complication-related movements could be prevented through the use of muscle relaxants and monitoring of neuromuscular blockade.

With increasing success in medical and surgical interventions for children with complex congenital heart disease, many are surviving to develop normal childhood diseases requiring ordinarily “routine” surgical care. Dr. M. Strafford of the Children’s Hospital and Harvard Medical School, Boston, presented her work with Dr. K. Henderson on anesthetic morbidity in congenital heart disease patients undergoing outpatient surgery. In one year, they identified 25 consecutive patients with congenital heart disease who underwent 27 anesthetics. Multiple diagnoses were found, and physical status category was two to three. All patients had preoperative room air saturations 90 percent by pulse oximetry. Two patients were noted to have an adverse event during the 2 7 procedures: one patient with unrepaired Tetralogy of Fallot had severe nausea and vomiting in the post anesthesia care unit, and one patient with Down’s Syndrome and an unrepaired ventricular septal defect had airway obstruction and bradycardia on induction, releived with an oropharyngeal airway and atropine. They concluded that patients with congenital heart disease can be successful candidates for day surgery, but suggest that patients with moderate cyanosis and uncompensated congestive head failure should be excluded.

In another study, Drs. Strafford and Henderson studied anesthetic morbidity in patients with congenital heart disease undergoing noncardiac surgery. They gathered data from 110 consecutive patients undergoing 135 anesthetics in one recent year. Sixty of these patients had previously undergone either definitive or palliative surgery. Twenty-four suffered congestive heart failure, 17 of whom were compensated at the time of this additional surgery. The median age was 2.9 years. Remarkably, 47 percent of the group suffered a perioperative event. The majority of these perioperative events were described as airway emergencies, dysrhythmias, or incidents involving circulatory instability. The authors pointed out that what might be considered a minor adverse event in a healthy patient is potentially much more serious in these children.

Special Anesthetic Problems

Dr. C. Haberkern of the Children’s Hospital Medical Center and University of Washington in Seattle presented a preliminary report of the National Preoperative Transfusion Study Group on preparation of sickle cell anemia patients for surgery. Preoperative transfusion is routinely carried out to suppress sickle cell production, dilute circulating sickle erythrocytes, and improve blood viscosity, although no prospective studies support this therapy, 565 patients have been enrolled from 35 institutions so far. They were randomized to two treatment arms: aggressive treatment to lower Hb S 30 percent and correct anemia to 10 gm/dl or simple treatment to correct anemia to 10 gm/dl only. Those patients who did not meet entry criteria were studied prospectively according to two further groups: no preoperative treatment, or nonrandomized treatment. Although the patient groupings have not yet been identified, data from the first 319 patients have been examined, including 169 in the randomized treatment groups. Initial results indicate an intraoperative complication rate of ten percent (hypothermia, hypoxia, hypertension, hypotension, airway obstruction and acidosis). Significant perioperative complications were acute chest syndrome or pain crisis (I 2 percent each), fever or infection in ten percent, new red blood cell antibody formation in nine percent, transfusion reactions in five percent, neurological events in two percent, death in two percent, and renal failure in one percent. The preliminary results outline an intra and perioperative care and complication profile for sickle cell anemia patients, and may suggest preferable methods of treatment once the study is completed.

Two abstracts addressed safety issues about malignant hyperthermia. The first, by Drs. C. Greenberg and H. Rosenberg of the Departments of Anesthesiology at Columbia University and Hahnemann University, reported the “hotline” experience of the Malignant Hyperthermia Association of the United States (MHAUS) in 1990, when over 534 calls were handled by 21 MH consultants. Seventy-three percent of the calls were from anesthesiologists or CRNA’s. Three hundred and thirty-five calls were to request information about MH; 60 percent concerned preoperative evaluation or perioperative management of known susceptibles or those with a suspicious family history. Ninety-six were requests for assistance with case management, 71 of which concerned masseter muscle rigidity (MMR) after succinylcholine, and 65 were inquiries about acute or suspected MH. Sixty-two percent of MMR cases were associated with otolaryngology procedures, and 12 cases of MMR progressed to clinical MH. Four cardiac arrests were reported, but no deaths.

A related study was reported by Dr. S. Yentis of the Hospital for Sick Children in Toronto. Many patients are labeled as “malignant hyperthermia susceptible” (MHS) because of positive family history or a previous MH reaction, without confirmation by a muscle biopsy. He reported a review of 268 patients (ages one day to 19 years) labeled as MHS who underwent 365 trigger-free general anesthetics during a ten-year period. Twenty patients (5.5 percent) were biopsy positive, 103 (28.2 percent) received prophylactic dantrolene (which was routinely administered prior to 1986), and 12 (3.3 percent) were pyrexial (38.5 degrees C) postoperatively, most episodes of which were explained by other causes. The authors concluded that intra and postoperative pyrexia and complications related to MH are rare in children labeled as MHS undergoing trigger-free anesthesia.

Do Not Resuscitate Orders and the Anesthesiologist

Three reports of surveys addressed a most interesting evolution in patient safety concerns the care of the patient with “do not resuscitate” (DNR) orders coming to the operating room for surgical procedures. Dr. M. Clemency of Emory University reported on a survey of 187 anesthesiologists, 59 percent of whom assumed that a DNR order is suspended when a patient is scheduled for surgery. Half of this group discussed with the patient/family this assumption or their intention to intubate and resuscitate during a cardiopulmonary arrest. Responses differed for palliative or elective procedures, elective procedures eliciting a greater tendency to provide resuscitative efforts than palliative procedures, but did not differ according to type of anesthetic technique (MAC, regional, or general anesthesia). While the respondents agreed that after discussions with the patient about DNR status and anesthetic risk, patients would retain DNR status, they nevertheless acknowledged that if the patient sustained a cardiopulmonary arrest, they would institute positive pressure ventilation with a mask or endotrachial tube, and use vasoactive drugs or defibrillation in a large number of cases.

The ambiguity of this dilemma continued in the report of Dr. G. Boyd from the University of Alabama at Birmingham, who surveyed the anesthesiology faculty as well as participants at an Annual Review course. Seventy-two percent of 125 respondents felt that DNR patients should never or rarely be candidates for surgical procedures. Fifty-nine percent felt that such orders should be routinely discontinued for the perianesthetic period and 68 percent would perform a full resuscitation for an iatrogenically induced arrest, while 66 percent would oblige the DNR order if the patient died from their intrinsic disease in the operating room. Eighty-seven percent felt the surgeon should inform the anesthesiologist in a timely fashion preoperatively. The point was further made that patients should be informed of the differences in causes and outcome for resuscitation in the operating room rather than elsewhere in the hospital..

The third poster, presented by Dr. D. M. Rothenberg of Rush Presbyterian-St. Luke’s Medical Center, Chicago, reported a survey of 106 hospitals. Fifty-four percent haw an existing DNR policy for surgical patients; 79 percent of those with a policy suspend it at the time of surgery. In addition, 31 percent still administered CPR to a patient despite maintaining DNR status. Only one out of three hospitals without a DNR policy were planning to formulate one.

Internal inconsistencies and ambiguities ran through all three of these reports. As was pointed out in the presentations, the Patient Self Determination Act just now taking effect will provide strong impetus for anesthesiologists to become informed and involved with their institutional ethics committees in the clarification of their response to DNR status and the administration of anesthesia.

Anesthesia Machine Safety Issues

Dr. J. Brooks of the Ohio State University revisited microbial contamination of anesthesia machines since general replacement of reusable equipment with disposable endotracheal tubes, masks, breathing tubes, and rebreathing bags over the past 20 years. Twenty sets of anesthesia machine cultures were obtained at the inspiratory valve, the expiratory valve, and the C02 canister. Anesthesia ventilators were cultured at the bellows and/or ventilator expiratory valve. Five of 20 sets of circle system cultures were positive for bacterial growth (Staphylococcus and Pseudomonas-like organisms). Anesthesia ventilators were 44 percent (8/1 8) contaminated with K. pneumoniae, Ps. aeruginosa, Xanthomonas maltophilia, Comomonas acidovorans, and Aspergillus sp. It is of particular interest that some of the most common surgical wound infections are from the same organisms identified in the contaminated anesthesia machines. This is an area that deserves attention because of the potential for safer care through lowering of surgical infection rates and also avoiding patient cross contamination.

Carbon monoxide (CO) may be generated in C02 absorbent following exposure to fluorinated anesthetics, according to an abstract presented by Dr. R. Moon from Duke University in collaboration with colleagues at Emory University and Northwestern University. Several cases of carbon monoxide poisoning have been reported previously, uncovered during routine co-oximetry, when either Baralyme® and Sodasorb® were used as the absorbent. The investigators measured CO levels inside Sodasorb cannisters in idle anesthesia machines 3 20 times. Two hundred seventy-one of the 3 20 samples (84.7 percent) showed CO concentration less than 20 ppm. Sixteen of the 320 samples (5 percent) showed CO concentrations greater than 100 ppm, with six of the 320 samples (1.9 percent) exceeding 1,000 ppm. (EPA standards are 9 ppm/8 hrs., or 35 ppm/1 hr. exposure). Cannisters which have been in place for longer periods of time were more likely to contain high CO concentrations, but one cannister which had been used for only nine anesthetics over four days also had a concentration greater than 1,000 ppm. The used Sodasorb contained formate, which, when gently heated, may have produced CO. The authors provided convincing evidence that formate may be an intermediate that may contribute to CO poisoning in C02 absorbent following exposure to fluorinated anesthetics.

Vaporizer Contamination

Dr. P. Popic and associates from the University of Wisconsin studied the incidence of contamination of vaporizers with an incorrect volatile agent. Fifteen ml aliquots from a total of 926 vaporizers at 32 institutions were obtained and examined for contamination with a gas chromatograph. Results showed that 29 of the 926 vaporizers (3 percent) were contaminated. Both high(50 percent halothane in an isoflurane vaporizer) and low degree contamination existed. There were few differences among various anesthesia practice settings (private practice vs. teaching) and the incidence of contamination.

Contamination increased significantly with: 1) more personnel available to fill vaporizers; 2) larger hospitals; and 3) more variety of type of vaporizers requiring filling. Agent specific monitoring did not decrease the incidence of contamination. Pin indexing decreased but did not prevent contamination (one vaporizer with pin indexing was contaminated).

Misfiring Vaporizers With Desflurane

Dr. J. Andrews and colleagues at the University of Texas, Galveston developed a computer model to demonstrate the effects of misfilling contemporary vaporizers currently in common use with desflurane. Desflurane has a vapor pressure of 743 mmHg at 22 degrees C.

At a one percent setting, desflurane output would be 40 percent from halothane and isoflurane vaporizers and a dramatic 58 percent from an enflurane vaporizer. MAC of desflurane is approximately six percent, indicating that a massive overdose is possible from this type of accidental misfilling. Further, a hypoxic mixture occurs with all misfillings at dial settings exceeding four percent even if the vaporizer is supplied with 100 percent oxygen.

Vaporizer Output During Patient Transport

In contrast to the misfilling issue, bouncing a vaporizer around appears less dangerous to the patient. The steady state output from vaporizers attached to transport carts encountering obstacles and jostling was studied by Drs. R. Friedhoff and J. Abenstein from the Mayo Clinic. A Fluotec vaporizer was welded to the frame of a gurney stretcher. While traveling down hallways, over floor/carpet strips, over electrical cords, accelerating, decelerating, shaking, and running into walls, output of halothane from the vaporizer was measured using an anesthetic analyzer.

Results showed the Fluotec vaporizer can he expected to deliver accurate and predictable concentrations of halothane under these conditions. There was never a change of more than 0. I percent anesthetic concentration during any of these potentially disruptive events.

Methodology for Incident Analysis

Dr. R. Cook of the Cognitive Systems Engineering Laboratory and the Department of Anesthesiology of Ohio State University presented a preliminary but fascinating abstract on the application of cognitive science techniques for analyzing anesthesia critical incidents, a study partly supported by a grant from the Anesthesia Patient Safety Foundation. While a reader new to these concepts presented in the printed abstract may be confused by the broad and nonspecific references to cognitive strategies, a handbook (“Human Performance in Anesthesia: A Corpus of Cases”) published by his laboratory was distributed by Dr. Cook at the session. Much of the effort and principles of cognitive science involve attempts to distinguish between bad outcomes and flawed performance and then focus on the causes of performance error. Human cognitive performance research is a dynamically evolving field that has already proven valuable in other event-driven professions often cited in analogies to the practice of anesthesiology such as nuclear power plant operation and commercial aviation, and is of fundamental interest in anesthesia patient safety.

OR Safety Behavior

In directly related work, Dr. D. Gaba and his group from Stanford presented two abstracts from their continuing studies of the behavior of anesthesiologists in critical situations in the operating room. In the first, they studied novice residents all within the first eight weeks of training. During each case, an observer noted which tasks the resident was performing; every five minutes the resident was asked to rate his or her workload on a numerical scale and the observer also noted whether the resident was noticing a red light placed near the EKG monitor and illuminated at random times during the procedure. The activities associated with preparation, induction, and emergence account for a large portion of the workload but occupy only a small fraction of the residents total time. Manual tasks diverted the residents attention from visual stimuli but it remained uncertain whether they would have been able to detect patient abnormalities appearing on monitors under the same circumstances. Comfortingly, Gaba did not find that interaction with the attending impaired visual vigilance.

In their second paper Gaba and colleagues describe how, using the principles developed in the aviation industry for Cockpit Resource Management, they have developed a course for anesthesiologists entitled Anesthesia Crisis Resource Management (ACRM). After an elaborate didactic course, trainees spend time in a very realistic simulator in which they have to manage crises of varying intensity without the possibility of harm to a patient. Although Gaba states that this training will not necessarily improve the outcome of patients of the physicians who participated, the attendees believed that it would. In these days when serious adverse outcomes are declining in frequency and residents will complete their training having very little experience with critical incidents, the use of such simulators seems to merit serious attention and support from the specialty.

Quality of Care and Peer Review

Dr. K. Posner and her colleagues who administer the ASA Closed Claims Study have extended their investigations into the scientific reliability of peer review. In this newly reported study they asked pairs of independently practicing anesthesiologists to review anesthesia malpractice claim files and judge the appropriateness of anesthesia cam “Standard of Care” judgments were based on the implicit criterion of Treasonable and prudent” practice. In a prior study this group has demonstrated that severity of injury influences the judgment of reviewers as to the appropriateness of cam So, in this study they measured levels of agreement in two subsets of claims of differing severity (namely, temporary and permanent injuries) as well as for the entire group of claims. Using sophisticated statistical methods of analysis for peer-reviewer agreement they found fair to good agreement for the whole group of claims, fair to good agreement for the permanent injuries but no agreement greater than chance for the temporary injuries.

Their results suggest that in a malpractice proceeding it may not be difficult for an attorney to find conflicting opinions among experts and that this may be the result of the absence of explicit guidelines in judging the quality of anesthesia care

Algorithms Ignored

In a study from the mountains of Utah and Switzerland, Dr. I East and associates asked why physicians taking care of mechanically ventilated patients in an ICU did not always follow the treatment instructions given them by a computerized protocol. The computer software also asked the physicians to give a reason why they did not feel that a particular treatment was valid and then the investigators looked at the antecedents when the reason given was ‘hemodynamic instability’. In a retrospective analysis, the authors found that there was no central tendency among the clinicians in any of the variables that would indicate they were following rules that defined ‘instability’. In fact, such variables as mixed venous oxygen saturation and wedge pressures were not even measured most of the time. The authors concluded that the decision making process used by most physicians has great variation, both internally one physician at different times, and externally among different physicians. Therefore, they now have protocols developed by a consensus group which they state are associated with a significant increase in patient survival perhaps due to better application and consequent reduction in the number of mistakes.

Picking Safe Residents

Two papers addressed the process of resident selection, adding credence to the almost universal impression that personality characteristics are most important in the eventual success and safety of an anesthesiologist. Dc 1. S. McDonald and colleagues from Ohio State University reported a five-institution study which attempted to develop a predictive profile for screening resident candidates to determine them fitness for anesthesiology. Ninety-five residents were tested using the California Psychological Inventory (CPI) before beginning their clinical training, and the results were compared with faculty evaluations at & end of the first and second years of residency. Four personality facets were positively correlated with performance: Empathy, Socialization, Achievement via Conformance (in structured situations), and Achievement via Independence (in unstructured situations). In two of the programs picked at random, the CPI was able to identify the weakest residents, and the authors concluded that the CPI could be used to screen candidates for their ability to succeed in anesthesiology.

Drs. D. Frankville and J. Benumof from the University of California at San Diego reported results of a national survey of factors used in selection of residents. They found that, by far, the inter-view was the most important component of the resident file, followed distantly by the Dean’s letter, the transcript, letters of recommendation, NBME scores and the CV. Little weight was afforded to the medical school or the personal statement.

Teaching Safe Residents

Three papers addressed how residents learn. Drs. S. Polk and S. Wirtes from the University of Chicago reported the correlation between conference attendance and examination scores to he weak but present in a group of first year residents. Dr. D. Reinhart from Southwestern Medical Center in Dallas found that resident participation in an independent study program with quizzes correlated positively with improvement in entraining examination scores, but that the residents were nonetheless poorly motivated to continue to participate in the program. Drs. 1. Downs and M. R. Hodges from the University of South Florida found that residents’ behavior (such as moderating the pressure in endotracheal tube cuffs) was positively, but not optimally, influenced by the presence of an independent observer in the operating room checking cuff pressure over a period of time. A lecture by the chairman specifically addressing appropriate inflation pressure did nothing to further optimize resident behavior in this respect. Dr. Downs is continuing this field of research, examining whether operant conditioning techniques can be used to bring about appropriate attention to details by residents.

Safety Issues for Anesthesiologists

It is now generally acknowledged and supported by data in the literature that anesthesiologists are at increased risk for chemical dependency, and two ASA abstracts addressed substance abuse. Dr. P. Frasco and associates from the Duke University Medical Center examined substance abuse problems in anesthesia residency training programs through the use of surveys obtained at the 1989 and 1990 Annual Symposia for anesthesiology chief residents m the United States and Canada. Survey results were analyzed according to four regions in the United States (Northeast, Southeast, Central, and West) and Canada. From 47 percent to 89 percent of training programs, depending on regional distribution, have a substance abuse rectum and from 0 percent to a maximum of 33 percent have a substance abuse committee.

Seventy clinicians (residents, fellows, faculty and CRNAs) were identified as substance dependent in 1989, and 73 in 1990, with at least nine deaths, but no attempt was made to correlate the presence of educational programs with the incidence of the disease. In discussion of this paper, Dr. B. Arnold presented some preliminary findings of the longitudinal study being conducted by the ASA Committee on Occupational Health of Operating Room Personnel, which indicate that training programs which address the issue of chemical dependency in anesthesia personnel do have a higher incidence of identification of affected colleagues. It seems possible that educational programs do not prevent the disease, but rather heighten awareness of it so that it may be more readily identified.

In a second paper on substance abuse, Dr. R. Klein and associates of the Oregon Health Sciences University attempted to look at the relationship between controlled substance accountability and controlled substance abuse. They received responses from 102 anesthesiology residency program directors, representing 137 hospitals. Ninety-five of the 137 hospitals (69 percent) reported at least one controlled substance dependent individual between 1985 and 1989. Seventy-six of the 102 training programs has at least one such individual. It was not possible to establish a correlation between the controlled substance accounting method and the occurrence of controlled substance abuse due to the low frequency of occurrence at an individual institution, with many additional variables. However, 45 percent to 48 percent of the responders indicated that they used the anesthetic record on a daily or random basis to audit the amount of controlled substance administered, and 31 percent performed random audits of individual anesthesia providers’ controlled substance usage pattern. Seventy-one of the 102 hospitals made changes in their accountability systems over the five years of the survey. Most program directors felt that their accountability programs were not aiding in the detection of dependent individuals.

Finally, in a further evaluation of anesthesiologist safety, Dr. K. Henderson and colleagues of Children’s Hospital and Harvard Medical School compared the radiation exposure of pediatric anesthesia fellows in the operating room and the cardiac catheterization laboratory (CL). The typical day in the CL would consist of two to three catheterizations, with fluoroscopy time ranging from 14-85 minutes, with an average of 30 minutes. The badge readings ranged from 20-180 mrem/month. Operating room fellows received undetectable ( 10 mrem/month) levels. AD fellows wore lead aprons, 50 percent wore a thyroid shield, and one stepped at least ten feet away from the source during every exposure (resulting in a reading of 30 mrem/month, despite spending 26 hours in the CL!). As the maximum permissible dose for non-radiation workers (including anesthesiologists) is 42 mrem/month, many of the anesthesiologists in the CL were exceeding the levels for many of the fluoroscopy technicians (40-150 mrem/month). Anesthesiologists, for their own well-being, should be aware of the principles of radiation safety (including duration of exposure) and the inverse square law (which governs the density of the dose per distance from the source), and should incorporate these principles, as patient safety permits, in their care of patients.

Dr. Holzman, Children’s Hospital, Boston is Chairman of the Harvard Department of Anesthesia Risk Management Committee; Dr. Folk moderated the ASA session on education and is from the University of Chicago; Dr. Webre is in charge of didactic education in the Department of Anesthesiology at the University of Mississippi; and Dr. Zeitlin of the Brigham and Women’s Hospital in Boston, is an Associate Editor of the APSF Newsletter.