2008 CAEP/ACMU Scientific Abstracts - Poster Presentations: 104-131

2008 Scientific Abstracts

CJEM 2008;10(3):255-295

Abstracts: 1-4, 5-24, 25-43, 44-73, 74-103, 104-131

104 A review in management of minor thoracic injuries (MTIs) in the emergency department

Shields J-F, Emond M, Guimont C, Pigeon D. Laval University, Québec City, QC

Introduction: MTIs with or without ribs fractures are a very common presentation in ED. About 75% of MTIs are treated on an ambulatory setting. Recently, it has been demonstrated that nearly 10% of the patients with MTIs may develop important delayed complications, such as hemothoraces and pneumothoraces. Our objective was to review the management and follow-up of patient with MTIs.Methods: A multicenter retrospective cohort study in 3 universityaffiliated Canadian EDs was conducted from January 2004 to January 2006. Patients older than 16 years with a suspected or proven rib fracture following a traumatic event were included. Chart of patient with any intra or extrathoracic complication at initial visit in ED were excluded. Univariate analyses were used to compare patient management profiles between hospitals. Results: There were 447 charts analyzed; only 21 (4.6%) patients were admitted during the study period. The proportion of admissions was significantly different between the three hospitals. No follow-up recommendations were identified in the majority (53.5%) of charts and there were no differences after stratifying for hospital. A planned follow-up visit was schedule for 5.7%. Neither age greater than 65 years nor the number of ribs fractured influenced the follow-up recommendations. Inadequate pain relief was the primary reason for an unplanned follow-up visit in ED in 23.7% of patients. There was not difference after stratifying for age and type of analgesia. Conclusion: Admission proportions are lower than expected from previous publications and varies across surveyed hospitals. The follow-up offer to patients with MTI seems insufficient in view of possible delayed complications and disabilities. Further study is suggested to derive predictors of delayed complication of MTI and orient appropriate use of follow-up resources. Keywords: minor thoracic injury, rib fractures, admission criteria

105 DO rate control drugs increase the likelihood of successful cardioversion in acute atrial fibrillation?

Stiell IG, Vaillancourt C, Clement C, Perry JJ, Symington C, Dickinson G. University of Ottawa, Ottawa, ON

Introduction: In the context of an ED protocol of acute rhythm control for acute atrial fibrillation (AAF), some clinicians believe that pretreatment with IV rate control (RC) drugs enhances safety and the chance of successful cardioversion. We sought to test this belief. Methods: We conducted a health records review of consecutive patient visits over a 5-year period for adults who presented to a university hospital ED with acute-onset AAF ( < 48 hr) and were managed with rhythm control. All patients received IV procainamide followed by electrical cardioversion if necessary. At the discretion of the attending ED physician, some patients were pretreated with IV RC drugs, usually 1 or 2 doses of IV metoprolol (5 mg) or diltiazem (15–25 mg) given over 5–10 minutes. Outcomes included conversion, adverse events and length of stay. We conducted Student t test, χ2 and odds ratio analyses. Results: We enrolled 628 patients, of whom 96.8% were discharged home from the ED in sinus rhythm. Patients who received RC drugs (39.2%) were similar to those who did not (60.8%) but had higher arrival heart rates (HR) (mean 129.5 v. 101.8 beats/min). RC drugs had no effect on HR (129.5 beats/min before v. 133.1 beats/min after) despite being given a median 30 minutes prior to rhythm control. Patients receiving RC drugs were less likely to convert to sinus rhythm after procainamide than those who did not (53.3% v. 64.1%, p < 0.01, OR 0.64, 95% CI 0.46–0.88). This lower conversion rate was also observed in the subgroups with HR >120 beats/min (n = 278; 58.6% v. 71.6%, p < 0.05) and HR >140 beats/min (n = 135; 59.6% v. 70.7%, p = 0.22). Metoprolol and diltiazem were followed by similar conversion rates (51.2% v. 57.1%, p = NS). RC patients had similar have adverse event rates (7.7% v. 7.1%, p = NS) but had longer times in the ED prior to discharge (mean 377.0 v. 308.7 min, p < 0.01).Conclusion: We found no benefit from use of RC drugs prior to ED rhythm control for AAF and that their use was associated with lower conversion rates and longer lengths of stay. Keywords: atrial fibrillation, rate control, cardioversion

106 ALCOHOL-related visits in youth: pediatric emergency department resource utilization

Clarke M, Lin L, Saewyc E, Whitehouse S. Division of Emergency Medicine, Department of Pediatrics, British Columbia’s Children’s Hospital, Vancouver, BC

Introduction: The incidence of binge drinking is increasing in Canadian youth. Its impact on emergency department resource utilization is unknown. This study characterizes the demographics, clinical features and management of youth who present to the pediatric emergency department (PED) with alcohol ingestion. Methods: A retrospective study in an academic, urban, tertiary care PED with an annual census of 38 000 visits was conducted. We reviewed the emergency records of all patients aged 10–17 presenting to the PED between Jan. 1 and Dec. 31, 2006. Youth with a documented history of alcohol ingestion and/or a positive blood alcohol level were included. Demographics, clinical presentation, management and disposition were collected using a standardized data collection form.Results: A total of 8290 visits were made by youth aged 10 to 17 years. There were 119 visits (1.4%) by 100 youth for alcohol ingestion. The average patient age was 14.8 years (SD 1.1 yr). The majority was female (58%). Most patients arrived by ambulance (79%). The average GCS on arrival was 12.6 (range 5–15). Ten (9%) patients had a GCS of less than 8. The average blood alcohol level was 46.4 mmol/L. Twenty-one (19%) patients had body temperatures less than 35°C. No patients were hypoglycemic. Injuries were reported in 47 visits (40%). One patient required intubation and 3 required ICU admission. Health care personnel were assaulted during 7 visits. Most patients were discharged directly from the PED (94%). The average length of stay was 7.2 hours. Conclusion: Although binge drinking may be considered a common adolescent risk taking behaviour, it is associated with high prehospital and emergency department resource utilization. There is a need to develop protocols to standardize management of this population and ensure health care provider safety.Keywords: alcohol ingestion, pediatrics, chart review

107 HIGH effectiveness of a new developed miniaturized extracorporeal assist device for mobile cardiopulmonary bypass in emergency medical services

Arlt M, Voelkel S, Philipp A, Hilker M, Schmid C, Zimmermann M. Department of Anaesthesiology, Airmedical Service University Hospital, Regensburg, Germany

Introduction: Severe cardiopulmonary failure resistant to advanced treatment is a major reason for exceedingly high mortality of patients located in outlying medical facilities. We retrospectively reviewed our experiences with a new developed miniaturized extracorporeal assist device for mobile closed chest cardiopulmonary bypass support in patients with severe cardiopulmonary or pulmonary failure.Methods: Between March 2006 and October 2007, the authors used mobile cardiopulmonary bypass support in 10 adult patients with severe cardiopulmonary failure (n = 7) and pulmonary failure (n = 3). We started closed chest cardiopulmonary bypass in the outlying medical facility using percutaneous femoro–femoral veno– arterial (n = 7) and femoro–jugular veno–venous (n = 3) vessel access. The new developed cardiopulmonary bypass system is a closed-loop extracorporeal circulation system, consisting of a centrifugal pump and a membrane oxygenator. The total weight is 27 kg. The system is capable for hand-held use and suitable for storing on a standard gurney for additional emergency medical service. Results: Bedside cannulation procedure was uneventful. After start of the circulatory support, pharmacological support could be marked down and systemic blood flow and gas exchange were restored. During extracorporeal assistance, including air (n = 8) and ground (n = 2) ambulance transport, no technical complication occurred. Limb ischemia was observed in 2 cases. Hospital survival rate was 70%. Conclusion:The use of this new miniaturized extracorporeal assist device is safe and highly effective in management of cardiopulmonary failure. Extracorporeal assistance for patients in need of emergency medical service is first time possible without extended technical or personnel support. Keywords: extracorporeal assist device, cardiopulmonary bypass, EMS

108 NORMAL renal ultrasound predicts low risk of urologic intervention for emergency department patients with suspected renal colic

Yan JW, Edmonds ML, McLeod SL, Theakston KD, Sedran RJ. Division of Emergency Medicine, Faculty of Medicine, University of Western Ontario, London, ON

Introduction: Renal colic is a common emergency department (ED) diagnosis. Computed tomography (CT) is a frequently employed imaging modality for patients with suspected renal colic because of its high diagnostic accuracy. However, there is increasing concern about the lifetime cumulative radiation exposure attributed to CT. Ultrasound (US) is an alternative imaging modality that may also be used to diagnose renal colic without exposing the patient to radiation. The objective of this study was to determine the ability of US to identify renal colic patients with a low risk of requiring urologic intervention within 90 days of their initial ED visit. Methods: A retrospective chart review was completed for all adult patients who had an ED-ordered US for suspected renal colic. Data was gathered from 2 tertiary care EDs with a combined annual census of 95 000 during a 1-year period (Jan. 1 to Dec. 31, 2006). Independent, double data extraction was performed for all imaging reports and results were categorized as normal, suggestive, stone seen or nonrenal disease. The charts of all patients with a normal US were reviewed to determine if they required any urologic intervention within 90 days after their initial ED visit. Results: There were 857 ED-ordered renal US during the study period. The study patients had a mean age of 44 years (range 18–95 yr) and 53% were male. Of the 857 renal US ordered during the study period, 373 (43.5%) were classified as normal. Of these, 49 (13%) underwent additional imaging identifying 6 (1.6%) stones, only 2 (< 1%) of which required urologic intervention with lithotripsy. Conclusion: A normal renal US predicts a very low likelihood for urologic intervention within 90 days for adult ED patients with suspected renal colic. The use of US may avoid the risks of radiation for many patients with suspected renal colic without adversely affecting their clinical outcomes. Further prospective research is needed to better define the role of US in the emergency management of renal colic. Keywords: renal ultrasound, urologic intervention, clinical prediction guide

109 BUILDING a student-run medical quiz website for medical students using Ruby on Rails an open-source programming framework

Kozan D, Graham T, Bullard MJ, Parks P, Rowe BH. Department of Emergency Medicine, Faculty of Medicine, University of Alberta, Edmonton, AB

Introduction: Rapidly developing Internet technologies allow development of websites that contain functionality previously only seen in desktop applications. A website created by a medical student called QuizMD (http://quiz.md) takes advantage of these so-called Web 2.0 technologies to allow students and teachers to publish quiz questions instantly online. QuizMD allows users to sort and refine the questions based on relevance and quality through discussion, voting, and automatically recorded statistics. Quizzes are generated on demand, containing randomly selected questions on desired topics.Methods: Using the Ruby on Rails (RoR) web-development framework, QuizMD was developed over a 6-month period based on user feedback and an iterative design approach. The flexibility of RoR allowed for multiple revisions to be created in a relatively short time period. Results: At the end of the 6-month period there were a total of 242 users in 7 Canadian provinces. 87% were students enrolled in the University of Alberta where the website was created. There were a total of 713 questions in 15 subject collections. Use of the highly flexible RoR framework allowed the website to quickly grow and adapt to the needs of its users. Efforts to establish an emergency medicine component are underway, and will be demonstrated at the exhibit. Conclusion: RoR is a robust and flexible webdevelopment framework that is well suited to database-backed applications. It fully incorporates modern web techniques such as AJAX (Asynchronous JavaScript and XML) that enable a better user experience. Its use enabled rapid development and deployment of a medical quiz website for medical students by a single developer. More iterative feedback is needed to examine the potential for such a site to affect the learning behaviors of medical students and residents (especially in Emergency Medicine) and to explore the feasibility of using the RoR framework for development of future medical applications.Keywords: Internet, Ruby on Rails, online quiz

110 CEFTRIAXONE for persistent acute otitis media: impact of a clinical practice guideline

Gauthier M, Chevalier I, Gouin S, Lamarre V, Abela A. Departments of Pediatrics and Otolaryngology, CHU Ste-Justine, Université de Montréal, Montréal, QC

Introduction: IV ceftriaxone (CEF) is recommended as secondline treatment for persistent acute otitis media (AOM). We aim to determine the effect of a clinical practice guideline (CPG) on the use of CEF for persistent AOM. Methods: Charts of all patients aged 3–60 months referred from a pediatric ED to a day treatment center (DTC) for persistent AOM for IV CEF were reviewed. Data were collected for two 18-months periods before (September 2002 to February 2004) and after (May 2004 to October 2005) implementation of a CPG. CEF was indicated for children with AOM who remained symptomatic despite 48 hours of high-dose amoxicillin or amoxicillin–clavulanate, or in whom symptomatic AOM recurred despite treatment with 1 of these 2 antibiotics over the previous month. Therapy was considered adequate if patients met these indications for CEF, if at least 3 daily doses had been prescribed and if all doses were of 40–60 mg/kg range. Results:Thirty-two ED physicians referred 127 patients to the DTC for treatment of persistent AOM with CEF. In the pre-CPG group (n = 60), patients had received a median of 2 (range 0–5) different antibiotics over the previous month, including amoxicillin or amoxicillin– clavulanate in 53% of cases. In the post-CPG group (n = 67), patients had received a median of 1 (range 0–5) different antibiotics, including amoxicillin or amoxicillin–clavulanate in 60% of cases. Indications for prescription of CEF were adequate in 17% of the pre- and 22% of the post-CPG groups (p = 0.4). Dose of CEF was adequate in 45% and 82% of the pre- and post-CPG groups, respectively (p < 0.001). Number of doses was sufficient in 62% and 58% of the pre- and post-CPG groups (p = 0.7). Before implementation of the CPG, overall treatment was considered adequate in 3 patients (5%); it was adequate in 7 patients (10%) after CPG implementation (p = 0.3). Conclusion: Implementation of a CPG for the treatment of persistent AOM with CEF has significantly improved prescription of adequate doses of CEF, but has not led to better indications for its use. Keywords: acute otitis media, ceftriaxone, clinical practice guideline

111 BARN door-to-needle time: thrombolysis in rural emergency departments

Vlahaki D, Fiaani M, Milne WK. South Huron Hospital and University of Western Ontario, London, ON

Introduction: The American Heart Association (AHA) has established a benchmark for the administration of thrombolytic therapy for acute myocardial infarction (AMI) patients as a door-to-needle (DTN) time of less than or equal to 30 minutes. Previous research at the national, provincial and rural levels suggests that this goal was not being achieved. The purpose of this study is to determine if the AHA benchmark of a DTN time of < 30 minutes for thrombolysis could be met in 2 rural emergency departments. Methods: A retrospective chart review was conducted at 2 rural hospitals in Southwestern Ontario. Descriptive data was obtained for each patient, including demographic information, Canadian Emergency Department Triage and Acuity Scale (CTAS) score and disposition. Visit timeline data was also collected and included the time the patient first experienced pain, was triaged, saw the emergency physician, had an ECG, received thrombolytic therapy and was discharged. Visit time intervals such as the median DTN times were then calculated. Summary descriptive statistics were calculated for each of these data points. Results: A total of 454 charts with a diagnosis of acute myocardial infarction (AMI) were pulled and checked for the administration of thrombolytic therapy. The final data pool consisted of 101 patients. The median age was 67, CTAS score level II (emergent), 63% male, 83% admitted to hospital, median door-to-ECG (DTE) time was 6 minutes, door-to-physician assessment (DTP) time was 8 minutes, DTN time 27 minutes, 58% of patients received thrombolysis in < 30 minutes and length of stay 2:20. Conclusion: A median DTN time of < 30 minutes is possible in a rural ED and suggests high quality care of AMI patients in these two rural hospitals.Keywords: door-to-needle time, thrombolysis, chart review

112 CARDIAC resuscitation skills of first-year internal medicine residents

Froats M, McGraw B, Pickett W. Queen’s University, Kingston, ON

Introduction: New physicians often provide the initial response to in-hospital cardiac emergencies and should be proficient at resuscitation skills. Several studies have raised concerns that undergraduate medical programs may not be adequately preparing graduates in this respect. We conducted a descriptive study to evaluate the hypothesis that new medical graduates lack the skills required to effectively manage cardiac emergencies. Methods: Using an objective structured clinical examination format, we assessed the cardiac resuscitation skills of 16 first-year internal medicine residents at our tertiary hospital-based center. All participants had completed an Advanced Cardiac Life Support course. The study was conducted in our patient simulation lab. Residents attended four skills stations that evaluated for proficiency at 1) basic airway management; 2) operation of a cardiac defibrillator/monitor; 3) cardiopulmonary resuscitation (CPR) and directing a team to manage a cardiac arrest; and 4) external cardiac pacing. Performance was assessed objectively using checklists of items and timed events, and subjectively via a global proficiency rating. Results: Performance varied widely within and across stations. Mean checklist scores ranged from 44% (3.1/7.0) for external cardiac pacing to 83% (5.0/6.0) for operation of the cardiac defibrillator/monitor. Mean global proficiency scores ranged from 2.1/5.0 for external cardiac pacing to 3.2/5.0 for CPR and leadership. The mean time to begin bag–mask ventilation of an apneic patient was 57 seconds (range 30–95). The mean time to demonstrate a cardiac rhythm on the monitor of a patient in cardiac arrest was 84 seconds (range 29–169). A total of 31 errors (mean 1.9/resident, range 0–4) that jeopardized the health of the patient or safety of health care staff were committed. Conclusion: New internal medicine residents at out center lack proficiency with cardiac resuscitation skills, resulting in delays in patient assessment and proper management during critical events. Keywords: ACLS, medical students, resuscitation skills

113 POTENTIAL vitamin interactions in children

Goldman RD, Rogovik AL, Vohra S. Pediatric Research in Emergency Therapeutics (PRETx) Program, Division of Pediatric Emergency Medicine, BC Children’s Hospital, Department of Pediatrics, University of British Columbia, Vancouver, BC; CARE program, Department of Pediatrics, University of Alberta, Edmonton, AB

Introduction: Significant growth in vitamin use has been documented in the literature in recent years both in children and adults. All vitamins have significant pharmacological activity and can interact with prescribed or over-the-counter medications. The objective of the study was to determine the frequency and types of potential interactions between vitamins and conventional medications in children arriving at the pediatric emergency department. Methods:A survey of parents and/or patients 0–18 years arriving at a large tertiary pediatric ED in Canada in the preceding three months.Results: A total of 1804 families were interviewed in this study. A considerable number of patients (11% of our cohort) had possible vitamin-medication interactions in the preceding 3 months which could theoretically result in adverse events, and over one-third of these children had more than one interaction. Patients with potential interactions and their parents were significantly older (p < 0.001 for the child and mother, p = 0.02 for father), they were much more likely to have a chronic illness (p < 0.001) and concurrently receive prescribed or over-the-counter medication (p < 0.001), and more children with interactions were completely immunized (p = 0.02). Child’s sex, parental education, employment status, family income and primary language spoken at home were not associated with interactions.Conclusion: Taking into account the high rate of potential vitamin–drug interactions, especially among older children and patients with chronic illness, parents and health care providers need to balance the potential benefit of concurrent vitamin–medication use with its potential harms. Keywords: vitamins, drug interactions, pediatrics

114 BILL 110: an evaluation of the impact of Canada’s first mandatory gunshot wound reporting law

Ovens HJ, Borgundvaag B, Park H. Mount Sinai Hospital and University of Toronto, Toronto, ON

Introduction: On Sept. 1, 2005, Ontario proclaimed into law Bill 110, Canada’s first law mandating GSW reporting. We assessed awareness and impact of the law through surveys of emergency physicians, police and the public. Methods: Electronic surveys were sent to 1080 members of the OMA Section on Emergency Medicine in June 2007. The public survey of 1000 Ontario residents was conducted by Ipsos–Reid in July 2007. The Ontario Government assisted in obtaining opinions by email from selected members/chiefs of police forces in Ontario. Results: The physician response rate was 24%; respondents were representative of the EM community. The great majority (89% and 93%, respectively) was aware of the law and willing to comply. Eighty percent reported no problems with either the police or the bill, and 86% perceived no change in relations with patients. Six incidents of patients delaying care were reported. Two-thirds of the public respondents were aware of the law; after being informed almost all (96%) expressed support, the majority (80%) felt it would not affect relations with physicians. The Ontario Association of Chiefs of Police informed us that there was strong consensus among the police leaders that Bill 110 has been beneficial. All 47 Bureau Commanders of the Ontario Provincial Police who received our survey replied. All felt Bill 110 was helpful to investigations, 70% felt it had improved public safety and 60% felt it had improved relations with health care workers. Eight (17%) reported being personally involved in an investigation initiated by a report under the Bill; leading to between 6 and 14 incidents of charges being laid and/or guns confiscated. Data on actual reports and results of investigations are not available. Conclusion: Bill 110 has been broadly accepted in Ontario. Anecdotes of patients delaying care to avoid reporting and of investigations based on reports leading to charges/gun confiscations were identified. Actual impact on injury prevention and safety cannot be assessed. Keywords:mandatory reporting, gunshot wound, Bill 110

115 EXPERIENCE, training not a predictor of high, average or low admitting in emergency physicians

Mutrie D, Bailey SK, Malik S. Northern Ontario School of Medicine, Thunder Bay Regional Health Sciences Centre, Thunder Bay, ON

Introduction: Empirical evidence for emergency physician admission rate variability is sparse to nonexistent. The few occasions this phenomenon is mentioned in the literature are in the context of anecdotal evidence or physicians’ adherence to some protocol or guideline. To the authors knowledge this is the first independent study of physician admission rate variability. Methods: A survey measuring attitudes and demographics was distributed to all emergency physicians (n = 30) at a large regional hospital (ED census ~ 90 000 patients/ yr). Admissions data from the previous year, including CTAS scores, was drawn from existing hospital databases and matched to individual physician survey results. The resultant dataset was analyzed for correlations and trends. Researchers were blinded to physician identity at all stages of the study. Results: Survey response rate was 97% (n = 29). Overall admission rates ranged from 8.69% to 17.00%, (mean 12.53, SD 2.21). Admissions data showed the greatest variability in the CTAS I (highest acuity) category, contrary to the researchers expectations, with an admission rate range of 69.23, (mean 70.63%, SD 13.78). More years of EM experience was significantly correlated with more admissions in the CTAS 2 category, r = 0.4, p < 0.05; however, in all other CTAS categories the correlation was negative, though not significant. Whether a physician worked full time, part time, less than part time or as a locum did not predict a pattern of high or low admitting, nor did having any particular post graduate certification (CCFP, CCFP[EM], FRCP). Conclusion: The absence of a significant relationship between rates of admission and educational factors or years of emergency medicine experience is notable and suggests that the variability measured is due principally to inherent physician characteristics. This study should be replicated to test generalizability of results. Broad variability in admitting practices in the ED might be lessened with treatment guidelines specifically targeting higher-acuity patients. Keywords: admission thresholds, practice variation, training

116 A comparison of survival probabilities according to the transfer status of trauma victims

Emond M, Lavoie A, Moore L. Laval University, Québec City, QC

Introduction: Little is known about the injury profiles of patients transferred from a lower to a higher level of care hospital compared to patients that are not transferred. Our objectives were to compare injury profiles and survival probabilities of transferred and nontransferred adult trauma victims in a regionalized system including 4 levels of care. Methods: The study population was based on the provincial Trauma Registry, which contains data observations from all trauma centres between 1998 and 2005. The population comprised 36 118 adults trauma patients, of these 9281 (25.7%) were transferred to a higher level of care. Deaths on arrival and in less than 2 hours after arrival at the initial hospital were excluded. Multiple logistic regression was used to compare the mortality experience of patients according to their transfer status and level of trauma care, while adjusting for injury severity (ISS), body region of the worst injury and age. Results: Comparison of adjusted mortality of patients transported directly to a trauma centre and not transferred revealed decreasing mortality for increasing expertise (odds ratios [OR] of 1.00, 1.08, 1.23 and 1.41 for levels I to IV, respectively). However, adjusted mortality of patients transferred to a level I centre was lower than that of patients sent directly to a level I trauma center: level IV OR 0.84, level III OR 0.907 and level II OR 0.48. Adding a transfer factor to the regression analysis model also revealed a protective effect of being transferred compared to direct transport (OR 0.86, 95% confidence interval 0.76–0.97). Conclusion: Lower mortality for patients transferred to a level I centre over patients arriving directly is illogical. This could be caused by the fact only patients with a higher survival probability according to clinical judgment are in fact transferred. Our results suggest that the results of studies evaluating the benefits of transferring trauma patients to higher levels of care could be misleading. Keywords: trauma, survival analysis, inter- institution transfer

117 SUGGESTIVE renal ultrasound is associated with significant rates of urologic intervention in emergency department patients with suspected renal colic

Sedran RJ, Yan JW, McLeod SL, Theakston KD, Edmonds ML. Division of Emergency Medicine, Faculty of Medicine, University of Western Ontario, London, ON

Introduction: Computed tomography (CT) is a common imaging modality for suspected renal colic because of its diagnostic accuracy; however, ultrasound (US) is an alternative that does not expose the patient to radiation. Clinical decisions are straightforward when the US is normal or demonstrates a ureteric stone. Frequently the US fails to visualize a ureteric stone but does demonstrate sonographic features suggestive of renal colic (hydronephrosis, perinephric fluid, absent urinary jets or intrarenal stones). The purpose of this study was to determine the urologic outcome of patients with a suggestive renal US. Methods: A retrospective chart review was completed for all adult patients who had an emergency department (ED)–ordered US for suspected renal colic. Data was gathered from 2 tertiary care EDs with a combined annual census of 95 000 during a 1-year period (Jan. 1 to Dec. 31, 2006). Independent, double data extraction was performed for all imaging reports and results were categorized as normal, suggestive, stone seen or nonrenal disease. Charts of patients who had suggestive US were reviewed to determine if a stone was documented on further imaging, and if urologic intervention was required within 90 days. Results: Of 857 renal US ordered during the study period, 182 (21.2%) were classified as suggestive. Fifty-two (28.6%) of these patients underwent additional imaging where 21 (11.5%) stones were identified. All but 1 of these stones was less than 7 mm (mean 4.8 mm, SD 1.7 mm). However, 13 (7.1%) patients required urologic intervention. Conclusion: There is a significant rate of urologic intervention within 90 days for patients with renal US suggestive of urolithiasis. Close clinical follow-up or use of CT may be warranted in this population. Future prospective research is needed to better define the role of US in the ED management of renal colic. Keywords: renal ultrasound, renal colic, urologic intervention

118 PREDICTIVE value of xanthochormia by spectophotometry in suspected subarachnoid hemorrhage (SAH) in the emergency department (ED)

Emond M, Perry JJ, Rachid A, Baril P, Nadeau L. Laval University, Québec City, QC

Introduction: The usual investigation tools of acute headache in the ED include CT scan and lumbar puncture. The absence of xanthochromia is used to exclude SAH when the initial CT scan is normal. Few studies have evaluated the diagnostic ability of xanthochromia by spectophotometry in the ED. Our objective was to establish the predictive value and change in angiography rate of different method of xanthochromia detection. Methods: A retrospective cohort study of patient with suspected SAH was realized from 2003 to 2006. Charts were reviewed if patients were more than 14 years of age, had a GCS of 15 and had a chief complaint of nontraumatic acute headache and an initial normal CT scan. Charts were identified using laboratory databases. Extraction of patient data permitted the evaluation of different xanthochromia detection. The definition of positive SAH was a positive angiography investigation and a positive visual xanthochromia or more than 5 × 106 red blood cells/L in the last LP tube. Sensitivity, specificity, predictive values and change in angiography rate were calculated. Results: There were 363 patients included, 5 (1.3%) had a positive SAH. There were no differences in baseline characteristics. Predictive values of different xanthochromia detection were as they appear in Table 1.Conclusion: Predictive values of xanthochromia by spectophotometry revealed low sensitivity with high specificity. A higher sensitivity goal will increase projected angiography rate and use of resources.Keywords: xanthochromia, spectrophotometry, subarachnoid hemorrhage

Table 1, Abstract 118.

	Sensitivity (95% CI)	Specificity (95% CI)	Projected angiography rate	% change in angiography rate
Visual	40 (5–85)	98.3 (96.4–99.4)	2.2	–71
Hendrik	80 (28–99)	87.4 (83.4–90.6)	13.6	+181
Chalmers	40 (5–85)	97.5 (95.3–98.8)	3	–60
Chalmers revised	40 (5–85)	98 (96–99)	2.5	–67
UK NEQAS	40 (5–85)	94.9 (92.1–96.9)	5.5	–26

119 VALIDITY of the Canadian Emergency Department Triage an Acuity Scale at a tertiary care center in Saudi Arabia

Aldarrab A, Bakhedher G, Almulhim M, Khawaji Y. King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia

Introduction: The Canadian Emergency Department Triage and Acuity Scale (CTAS) is a 5-level triage tool, with I being the most acute. It has been shown to be a reliable and valid triage tool in Canada. In 2002, CTAS was first instituted at our tertiary care center in Saudi Arabia. However, it has never been evaluated in our population. We sought to evaluate the validity of CTAS in predicting hospital admission at our tertiary care center in Saudi Arabia. Methods:Our emergency department (ED) sees 45 000 visits per year with an overall admission rate of 9%. A retrospective study of all patients presenting to our ED from Sept. 1 to Sept. 30, 2007, was performed. The CTAS category and ultimate disposition for each patient was recorded. For this study, ED death and hospital transfer were considered as admissions. Results: During the study period, 4108 patients visited the ED and 3636 were included in the final data analysis. Four hundred seventy two patients were excluded due to missing or incomplete data. The total number in each triage category was 18 (CTAS I), 150 (CTAS II), 896 (CTAS III), 2098 (CTAS IV), 474 (CTAS V). Hospital admission rates were 67% (CTAS I), 47% (CTAS II), 18% (CTAS III), 3% (CTAS IV), 1% (CTAS V).Conclusion: At our tertiary care center in Saudi Arabia, CTAS appears to be a valid triage tool that predicts patient’s disposition from the ED. Keywords: CTAS, triage, Saudi Arabia

120 PREDICTIVE factors of delayed complications of acute minor thoracic injuries (MTI).

Shields J-F, Emond M, Guimont C, Pigeon D. Laval University, Québec City, QC

Introduction: MTI with or without ribs fractures are associated to delayed significant complication (DC) in nearly 10% of patients. It has been suggested that age more than 65 years and the number of fractures could predict those DC. According to a previous study there is an important lack in standardized recommendation for follow- up of these injuries. Our objective was to derive predictive factors of DC. Methods: A multicentre retrospective cohort study was performed in three university-affiliated Canadian EDs from 2004 to 2006. Patients older than 16 years old with an MTI, including suspected or proven rib fracture with a normal chest radiograph, were included. Charts of patient with any intra- or extrathoracic complication at initial visit in ED were excluded. Main outcome was the presence of hemothoraces/pneumothoraces at any follow-up visit. Any other significant complications (e.g., lung contusion, pneumonia) composed the secondary outcomes. Univariate and multiple regression analyses were used to compare profiles of patients with or without delayed complications. Results:There were 588 charts analyzed; the most frequent DC was hemothorax (3.57%).

Consultation delay of less than 24 hours from trauma was significantly associated with delayed hemothoraces (OR 4.5, 95% CI 1.3–15.3). Age greater than 65 and the number of rib fractures were not statistically associated with delayed hemothorax. However, those older than 65 (OR 2.3, 95% CI 1.1–4.8) or taking antiplatelets medications (OR 2.2, 95% CI 1.1–4.4) were associated with the presence of at least 1 DC. Conclusion: The incidence of delayed hemothoraces is lower than expected. Age and the number of fracture failed to predict delayed hemothoraces. Surprisingly, delay in consultation and use of antiplatelets agents were predictive factors of any DC. A larger prospective study with systematic follow-up is needed to confirm those results. Keywords:minor thoracic injury, hemothorax, chart review

121 A review of gastrointestinal decontamination for overdose patients at the Ottawa Hospital

Natarajan S, Kuo A, Gee A, Patel R, Corman C, Kanji S. The Ottawa Hospital (General Campus), Ottawa, ON

Introduction: The risks and benefits of gastrointestinal decontamination are considered in the emergency department for all poisoned patients. Recently we experienced several potentially avoidable serious adverse events related to gastrointestinal decontamination that necessitated the creation and implementation of institutional guidelines at our hospital based on published recommendations for the use of activated charcoal. The purpose of this study was to evaluate our adherence to our institutional guidelines for the use of activated charcoal in poisoned patients. Methods:We performed a retrospective chart review of consecutive poisoned patients presenting to our emergency department between Jan. 1 and Dec. 31, 2006. Patients transferred from other institutions and those with noningested poisonings were excluded. Data describing patient demographics, overdose characteristics, gastrointestinal decontamination use and eligibility, and outcomes were collected. Eligibility and use of gastrointestinal decontamination was independently reviewed and adjudicated by 3 separate reviewers using predefined criteria from institutional and published guidelines. Results: Data was collected from 192 consecutive cases of poisonings during the study period. Activated charcoal was administered in eight of 33 eligible cases (24%) and in 8 other cases where patients did not meet institutional eligibility criteria. Three of the 16 patients who received activated charcoal had an adverse event. Conclusion: The majority of patients deemed to be eligible for gastric decontamination with activated charcoal did not receive therapy for their ingestion. Half of the patients who received charcoal did not meet institutional eligibility criteria. Further educational efforts to raise awareness of institutional guidelines and earlier recognition of eligible patients are warranted to ensure appropriate use of activated charcoal and minimize avoidable adverse events. Keywords: gastrointestinal decontamination, overdose, chart review

122 THE potential of web-based graphical user interface design to contribute to medical error in clinical decision support systems

Graham TAD, Bullard MJ, Holroyd BR, Kushniruk AW, Meurer DP, Rowe BH. Department of Emergency Medicine, Faculty of Medicine and Dentistry, University of Alberta Hospital, Edmonton, AB

Introduction: Emergency physicians (EPs) are expected to adhere to complicated treatment algorithms and decision support rules. The University of Alberta hosts eCPG, an intranet-based Clinical Decision Support System (CDSS) that contains a variety of electronic clinical tools that include patient handouts, order sets, assessment tools, clinical practice guidelines with varying degrees of interactivity and other clinical updates. Two CDSS pertaining to community acquired pneumonia (CAP) and neutropenic fever (NF) had secondgeneration prototypes developed as part of a project to add extra functionality to the eCPG site. Inherent in the development process was a detailed analysis of the usability of the existing and prototype CAP and NF applications. Methods: Seven volunteer EPs were observed performing tasks based on standardized mock patient encounters. Screen graphical user interface (GUI) interaction was recorded with automated screen capture software along with audio recording. Audio recording of EPs using the tools were transcribed verbatim, analyzed and correlated with events in the automated screen capture videos to identify relevant incidents. Results: A total of 422 events were recorded in the 56 sessions. Seven main categories of events emerged: 1) negative comments; 2) positive comments; 3) neutral comments; 4) events (e.g., subject ignores drug dosing recommendation); 5) problems; 6) slips (corrected mistakes); and 7) mistakes. In total, there were 169 negative comments, 55 positive comments, 130 neutral comments, 21 events, 34 problems, 6 slips and 5 mistakes identified. Of the 5 mistakes, 2 could have potentially led to adverse events. Conclusion: Previous studies have shown that technology-induced error can lead to adverse medical outcomes. Direct observation of clinical users showed that GUI issues can lead directly to mistakes that could potentially lead to adverse events. Performing usability testing early in design phases of CDSS could identify and mitigate such problems.Keywords: clinical decision support system, graphical interface, medical error

123 ULTRASOUND versus CT scans for estimation of stone size in renal colic

Edmonds ML, Yan JW, McLeod SL, Theakston KD, Sedran RJ. Division of Emergency Medicine, Faculty of Medicine, University of Western Ontario, London, ON

Introduction: Although computed tomography (CT) is a frequently employed imaging modality for suspected renal colic because of its diagnostic accuracy, ultrasound (US) is an alternative that does not expose the patient to radiation. It has been suggested that US overestimates the size of stones. As part of a review of the utility of US in the management of emergency department (ED) patients with possible renal colic, we compared the reported stone sizes for patients who had stones seen on both US and CT.Methods: A retrospective chart review was completed for all adult patients who had an ED-ordered US for suspected renal colic. Data was gathered from 2 tertiary care EDs with a combined annual census of 95 000 during a 1-year period (Jan. 1 to Dec. 31, 2006). Reported stone sizes were compared in patients who underwent renal CT within 90 days of ED-ordered US and had a stone seen on both tests. Results: There were 857 ED-ordered renal US during the study period. Of the 241 renal US where a stone was seen and measured, 44 had a CT done in follow-up and there were 30 stones seen on CT which could be compared. There was significant correlation between the stone size reported on US versus CT (r = 0.53, p < 0.01). US overestimated the size of the stone in 19 (63.3%) cases, with the mean overestimate being 1.7 mm (range 0.3–9.0 mm, p = 0.046). However, in 17 (56.7%) cases, the size reported on CT and US differed by 2 mm or less. Conclusion: Although there was significant correlation between the sizes of stones seen on US versus CT, on average US overestimated the stone size. Stone size is often used to predict the need for urologic intervention; therefore, this may affect the utility of US in determining the need for urologic follow-up and treatment. Keywords: ultrasound, computerized tomography, stone size

124 BEST better early stroke treatment: implementation of nursing questionnaire aids in triage of acute ischemic stroke patients

Hoff AM, Yassa AS, Bellolio MF, Vaidyanathan L, Kashyap R, Enduri S, Suravaram S, Brown RD, Decker WW, Stead LG. Mayo Clinic College of Medicine, Rochester, MN

Introduction: As part of the recommendations of the current guidelines for the management of acute stroke, hospitals must create efficient pathways to rapidly identify and evaluate stroke patients. To minimize delays in triage, we developed an organized protocol to speed the clinical assessment, the performance of diagnostic studies, and decisions for early management. Methods: A random cohort of 43 patients was collected after implementation of the Better Early Stroke Treatment (BEST) Nursing Questionnaire in the setting of a tertiary care ED. BEST Nursing Questionnaire includes:

• Patients name
• Location of referring ED
• Time of symptom onset
• Consideration of t-PA by outside physician
• t-PA administration
• Estimated time of arrival

Data collected included symptom onset to ED presentation, consideration of t-PA, t-PA administration, and ED length of stay (LOS). Descriptive statistics and Wilcoxon/Kruskal–Wallis tests were performed in JMP software (SAS institute, Version 6.0). Results: The BEST Nursing Questionnaire was completed in all 43 patients prior to arrival at the tertiary care ED. Only 16 of the 43 patients were considered to be candidates for t-PA by the outside facilities. Of those 16, a total of 4 received t-PA prior to their arrival at the tertiary care ED. All patients were triaged to the specialized area of the ED upon arrival and managed by a standardized acute stroke protocol. The ED LOS was available in 35 of 43 patients. The median ED LOS was 2.83 hours (interquartile range 2.08–5.25 hr). This time is significantly lower than the recorded times during the previous 12 months, with a median of 3.55 hours (interquartile range 2.63–4.70 hr), p = 0.046. Conclusion: Implementation of the BEST Nursing Questionnaire aids in management of AIS patients by triaging patients to a specialized area of the ED and decreasing ED LOS. The questionnaire enhanced ED throughput by shortening the median ED LOS by approximately 1 hour which is of paramount importance given the current overcapacity trends. Keywords: acute ischemic stroke, triage, length of stay

125 IMAGING modalities for the assessment of acute appendicitis in the emergency department

Grushka DH, McLeod SL, Edmonds ML. Division of Emergency Medicine, London Health Sciences Centre, Schulich School of Medicine and Dentistry, University of Western Ontario, London, ON

Introduction: Appendicitis is routinely on the differential for patients that present to the emergency department (ED) with an acute abdomen. Clinical diagnosis can be inaccurate with a reported negative appendectomy rate (NAR) of 16%–40%. Several studies have indicated that despite the increased use of diagnostic imaging, NARs have remained relatively unchanged. The objective of this study was to review the utilization of imaging modalities, factors associated with choice of imaging modality and the NAR at our institution.Methods: A retrospective chart review was completed for all adult patients with an ED diagnosis of appendicitis. Data was gathered from 2 tertiary care EDs (combined annual volume 95 000) during a 1-year period (Apr. 1, 2006, to Mar. 31, 2007). Results: Of 250 patients with a diagnosis of appendicitis, 30 were managed without imaging. One hundred and twelve patients had an ultrasound (US), 128 had a computed tomography (CT) scan and 15 had both. Younger patients (< 45 yr) were less likely to have a CT scan (26.6% v. 71.7%, p < 0.001) compared to older patients (> 44 yr), with no difference found between males and females (44.4% v. 42.1%). Younger patients were more likely to have an US compared to older patients (49.4% v. 15.2%, p < 0.001), with young females receiving more US than males (45.2% v. 28.2%, p = 0.005). The NAR was similar in those patients managed with US (6.5%) and CT (4.7%); however, the NAR was statistically higher for those patients that did not have imaging (20.0%, p < 0.05). Conclusion: There was a high rate of imaging overall with 51.2% of patients receiving a CT scan prior to appendectomy. Although the rate of misdiagnosis was lower in those who received imaging, the most appropriate use of these diagnostic tests warrants further study owing to costs, time in the ED and concerns regarding the lifetime cumulative radiation exposure attributed to CT. Keywords: appendicitis, imaging, radiation exposure

126 ASTHMA presentations by children to emergency departments in a Canadian province: a need for better follow-up

Rosychuk RJ, Voaklander DC, Klassen TP, Senthilselvan A, Marrie TJ, Rowe BH. Department of Pediatrics and Emergency Medicine, University of Alberta, Edmonton, AB

Introduction: Asthma has a high prevalence among North American children and presentations for exacerbations to the emergency department (ED) are common. In this study, the epidemiology of acute asthma presentations by children in the province of Alberta is described. Methods: ED visits were identified in the Ambulatory Care Classification System databases and linked to other provincial administrative databases to obtain all ED and follow-up encounters for asthma made by children less than 18 years of age during 6 fiscal years (April 1999 to March 2005). Information extracted included demographics, ED visit timing and subsequent visits to physicians; all data were coded by trained medical records nosologists. Data analysis included descriptive summaries and directly standardized visit rates (DSR). Results: During the study period, 94 187 ED visits for asthma were made by 45 385 distinct children, averaging of 2.1 visits per child. The rates were 21/1000 in 99/00 and 19.8/1000 in April 2005. Annual visits for Aboriginal children were higher (124/1000) than any other group. Until age 14, more male than female children presented; a noticeable spike reaching 47.6/1000 visits for boys aged 1 to 4 years old compared to 26.9/1000 for girls; these rates decline and reverse in adolescent ages. Daily, weekly, and monthly trends were also seen. While 89.2% of these ED visits resulted in discharge, these encounters were followed by infrequent physician follow-up (29%) within 1 week of discharge. Repeat ED visits within 1 week (5.4%) and 180 days (23.8%) were identified.Conclusion: Asthma is a relatively common presenting problem in Alberta EDs. These results suggest that the rates have remained relatively constant; however, disparities exist based on age, sex and socio- economic/cultural status. Follow-up reassessments with physicians within a week are lower than expected and may explain the repeat ED presentations. Interventions to reduce asthma-related ED visits by children are needed. Keywords: asthma, pediatrics, continuity of care

127 LOW-IMPACT pelvic fractures presenting to the emergency department

Dodge G, Brison R. Queen’s University, Kingston, ON

Introduction: Elderly patients with low-impact pelvic fractures present a challenge of appropriate disposition for emergency department (ED) physicians. Few studies describe patterns of occurrence, ED-based clinical care and outcomes for these patients. Methods:Patients treated for a pelvic fracture in a tertiary care ED in Kingston, Ontario, in 2005/06 were identified in emergency discharge databases using ICD-10 codes. A retrospective chart review of ED and hospital records collected information on the injury event, clinical care provided and outcomes at 1 year on those with low-impact pelvic fractures. Results: Of 132 pelvic fractures captured, 77 were identified as low-impact pelvic fractures. Patients were predominantly female (82%), with an average age of 81 years and all but 3 reporting comorbidity. Common presentation was by ambulance (74%) following a fall (94%) with 82% the result of simple falls from standing. Fractures involved the pubic rami (86%), sacrum (25%), acetabulum (21%), ilium (4%) and ischium (1%) with multiple sites in 35%. The median length of stay in the ED was 9.4 hours. Consults included orthopedics (43%), medicine (8%), social work/homecare (33%). Twenty-five (32%) were admitted (orthopedics 20, medicine 5). Median length of stay in hospital was 13 days. Ten patients had surgical stabilization, most of the acetabulum. Five patients died in hospital, 4 precipitated by pneumonia and 1 by myocardial infarction. Eight additional patients died within 1 year of discharge from ED or hospital. Postinjury, 18% lived independently and 16% walked without aids versus 42% and 38%, respectively, before the injury. Conclusion: Low-impact pelvic fractures affect an elderly patient population with substantial pre-existing comorbidity. They are important injuries that affect independence and seem associated with an increased risk of death. Additional research is needed to develop appropriate and safe disposition plans for rehabilitating these persons postinjury. Keywords: low-impact pelvic fracture, geriatrics, prognosis

128 FACTORS associated with troponin I elevation in middle-aged adults presenting to the emergency department

Sommers JM, Klett MJ, Mancusso M, Allard J, Baer C, McAvinue T. Dalhousie University, Halifax, NS

Introduction: Chest pain (CP) is a frequent and potentially ominous complaint in the emergency department (ED). Despite low incidence of acute myocardial infarction (AMI), many troponins (cTnI) are drawn in middle-aged adults representing significant resource utilization. Strategies aimed at risk stratification to guide ED assessment could aid in optimizing cTnI assessment in this cohort. Methods:Subjects included all men aged 40–55, and women aged 40–65 who had cTnIs drawn in the Moncton Hospital ED, Moncton, NB, in 2005. Over 100 variables were assessed. The main outcome measure was cTnI elevation (> 99% upper limit normal). Results: There were 1593 charts reviewed (627 men, mean age 51.2 yr). Of these, 91 (5.7%) patients had a positive cTnI. There were 29 factors significantly associated with cTnI elevation identified. Male sex (p < 0.001), smoking (p = 0.007), dyslipidemia (p = 0.005), and obesity (p = 0.01) were associated with positive cTnI. Diabetes, hypertension, and family history were not. Pain radiation to both arms (p < 0.001), diaphoresis (p < 0.001) and retrosternal location (p = 0.002) were also associated with cTnI elevation. Any ECG change, including nonspecific ST changes (p = 0.005), was associated with positive cTnI, as was leukocytosis (p < 0.001) and random glucose (p = 0.01). Factors negatively associated with cTnI elevation were prior number of ED visits (p < 0.001), pain reproduced by palpation (p = 0.02), sharp CP (p = 0.023), brief duration (p = 0.04). Conclusion: Although evaluation of CP in middle-aged adults represents significant ED resource utilization, no studies have identified factors associated with cTnI elevation in this population. We identified aspects of patient history, physical and ECG findings that might be used to better identify those patients at greater risk of AMI. Future studies should target strategies for risk stratification of this patient population and methods to optimize cTnI use. Keywords: troponin, chest pain, chart review

129 EMERGENCY department troponin I assessment in middleaged adults: patterns of use

Sommers JM, Klett MJ, Mancusso M, Allard J, Baer C, McAvinue T. Dalhousie University, Halifax, NS

Introduction: Chest pain (CP) is a frequent and potentially ominous complaint in the emergency department (ED). Despite low incidence of acute myocardial infarction, many troponins (cTnI) are drawn in middle-aged adults representing significant resource utilization. Understanding patterns of utilization can aide in planning/resource allocation for the ED. Methods: Subjects included all men aged 40–55 and women aged 40–65 who had cTnIs drawn in the ED at the Moncton Hospital, Moncton, NB, in 2005. The aim was to identify factors associated with cTnI elevation; a subanalysis was performed to describe the patient population and utilization patterns of cTnI in middle-aged adults. Results: There were 1593 charts rreviewed (627 men, mean age 51.2 years). There were 2529 cTnIs performed in study subjects. Of patients, 1157 (72.6%) had chest pain, 165 (10.4%) had pre-existing coronary artery disease and 243 (15.3%) had diabetes. Ninety-one (5.7%) patients had a positive cTnI. Smoking (p = 0.007), dyslipidemia (p = 0.005), and obesity (p = 0.01) were associated with positive cTnI. Diabetes, hypertension and family history were not. Most patients had 1 (n = 860) or 3 (n = 599) cTnIs drawn. Of pateints, 684 (42.9%) were discharged after 1 cTnI. Mean previous number of ED visits for CP was 1.12 (SD 2.53); 490 (30.8%) patients had an ED diagnosis of CP not yet diagnosed (NYD). There were 618 (38.8%) patients admitted, 172 (10.8%) were assessed in the rapid chest pain assessment clinic. Conclusion:Despite a relatively low incidence of AMI, a large number of middleaged adults have cTnI evaluation and are admitted to hospital. A significant proportion of patients were diagnosed with CP NYD. CP in middle-aged adults poses a significant diagnostic challenge and resource utilization in the ED. Future research should target strategies of risk stratification for optimal cTnI assessment in this cohort.Keywords: troponin, chest pain, utility

130 ADHERENCE of emergency physicians to Canadian Cardiovascular Society Guidelines in the treatment of patients with acute decompensated heart failure

Steinhart B, Semeniuk L, Isaacs D. St. Michael’s Hospital, Toronto, ON

Introduction: The recently published 2007 update of Canadian Cardiovascular Society (CCS) Guidelines recommend the use of intravenous diuretics and vasodilators for treatment of acute decompensated heart failure (adHF). Previously published data from a single regional study (Edmonton, AB) found these agents to be used in a relatively small proportion of patients presenting to the emergency department (ED). We extended the analysis to include both eastern and western Canadian centres to determine the ED practice pattern and also compared our findings to US data. Methods: Medications given in the ED were determined by retrospective chart reviews in consecutive patients presenting to 2 Canadian EDs: Foothills Medical Centre (FMC), Calgary, Alberta (Oct. 1, 2004, to Mar. 31, 2005; n = 223) and St. Michaels Hospital (SMH), Toronto, Ontario (Apr. 1, 2005, to Sept. 30, 2005; n = 123). ICD-10 codes (FMC) and Emergency Department Information System-EDIS (SMH) for adHF admission were used to screen patients. This was compared with US ADHERE Registry data. Results: More patients received antihypertensive medication in the western Canadian centre; neither center utilized vasodilator therapy to any significant extent comparable with ADHERE registry data. Conclusion: Based on new CCS guidelines, treatment patterns before guideline inception show suboptimal therapy for the treatment of adHF in 2 major academic centres in Canada. This study advocates the need to apply these practice guidelines and establish treatment algorithms to guide future treatment of adHF. Keywords: acute decompensated heart failure, clinical practice guideline, guideline adherence

131 ASSESSMENT of medical adventure racing as a tool for teaching acute care skills

Gilic F, Blagojevic A, Brown G. Queen’s University, Kingston, ON

Introduction: Simulation through moulage (representations of clinical scenarios using actors, props and make-up) is often used to teach acute care skills, particularly related to trauma. Medical Wilderness Adventure Race (MedWARs), a series of educational events held around North America since 2001, is a novel method of moulage teaching in an outdoors setting. The objective of the study was to assess the effectiveness of MedWAR participation in increasing the participants’ sense of mastery of acute care topics. Methods: The study consisted of 2 questionnaires, given before and after the event to all willing participants of the MedWAR North race, held near Toronto in March 2006. The race included scenarios on mass trauma, near drowning, exposure, burns, etc. The participants were asked to rate themselves on 14 areas of acute care knowledge and 5 areas of acute care processes on a 5-point Likert scale. Pre- and postrace self-ratings on each item were compared using Wilcoxon matched-pairs nonparametric test. Results: Of 42 participants, 25 filled out the questionnaires. The majority (56%) were medical students, the rest being residents, practising MDs and paramedics. Eleven out of 14 categories were found to be significantly (p < 0.05) increased postrace (Scene assessment, Triage, Secondary survey, C spine, Neuro, Chest, Abdo, MSK and Exposure injury, Shock and Evacuation). Primary Survey, CPR and Airways were not improved. Process questions improved in Teamwork, Calmness and Comfort with acute scenarios (p < 0.05). Organization was not improved. Majority of participants (96%) thought MedWAR was a good learning method, transferable to hospital setting and preferable to didactic teaching. Conclusion: MedWAR participation increased selfreported sense of knowledge and comfort with acute care in the majority of subject areas sampled by the survey. Keywords: Medical Wilderness Adventure Race, moulage, simulation

Abstracts: 1-4, 5-24, 25-43, 44-73, 74-103, 104-131