2007 CAEP/ACMU Scientific Abstracts - Poster Presentations: 161-189

2007 Scientific Abstracts

CJEM 2007;9(3):183-226

Abstracts: 1-30, 31-44, 46-48, 100-130, 131-159, 161-189, 191-216

Poster Presentations

161 OVERVIEW of triage diagnostic test properties for medical priority dispatch and the Canadian Triage Acuity Scale

Panylyk AH. City of Edmonton, Emergency Medical Services, Edmonton, AB

Introduction: Medical Priority Dispatch (MPD) is a measure of pre-clinical triage used in Emergency Medical Services as a diagnostic test without benefit of scientific scrutiny. In contrast to clinical triage, the Canadian Triage and Acuity Scale (CTAS) is a converging work-in-progress. Methods: To find and assess studies, those that evaluate the validity and or reliability of MPD and CTAS for adult patients in an urban Canadian emergency setting at the individual patient level. A search for studies was performed using the SEARCH Canada Vividex Resource Library. Studies detecting a diagnostic-related group or level of acuity as the outcome measure, either individually or in combination, were selected. One reviewer screened the title and abstract of citations. Full text articles for potentially eligible studies were obtained and cross referenced to identify additional publications. Two reviewers independently assessed the full text of the article for relevance and study rigour. Differences were resolved by discussion. Results: Citations for 68 articles were retrieved. The screening process culled 34 articles. Review of full text articles excluded an additional 28 articles. Six primary diagnostic studies were selected. None of the studies assessed MPD. All of the studies estimate reliability test properties for CTAS. None of the studies focus on test validity. Conclusions: Pre-clinical triage is a topic of research that has been ignored. Research in clinical triage is directed towards the reliability of CTAS. Reliability is an important first step in an evidence-based approach to a new triage system. Key Words: Emergency Medical Services, Triage, MPD, CTAS

162 EVALUATION of a regional EMS acute stroke protocol in Southeastern Ontario

Chu J, Jones G, Reed A, Halladay M, Bolton C. Department of Emergency Medicine, Queen's University, Kingston, ON

Introduction: A Regional Acute Stroke Protocol (RASP) was implemented in 1999 in Southeastern Ontario to provide regional access to acute stroke care. Paramedics were trained to identify patients eligible for treatment with tPA and to bypass the closest hospital in favour of direct transport to the Southeastern Ontario Health Science Centre in Kingston. The stroke team was "activated" by the paramedics prior to patient arrival in the ED. This study examined whether paramedics could accurately identify patients with acute stroke and transport patients from a wide geographic area in time to receive thrombolytic therapy. Methods: Data from Ambulance Call Reports (ACRs) was collected prospectively and entered into a database by the Regional Base Hospital for Southeastern Ontario. Inhospital data was collected by the Ontario Stroke Registry at Kingston General Hospital. This study examined the ACRs and hospital records for all patients with stroke or TIA encountered by paramedics in SEO for 2004/2005. Results: In 2004/2005, paramedics encountered 1501 patients with a working diagnosis of TIA or acute stroke. 337 (22%) of these patients met RASP criteria (symptoms compatible with acute stroke, known time of onset and able to reach the stroke center within two hours) and were transported directly to the stroke center with stroke team activation. The majority of patients were excluded from the RASP because of time or symptoms resolving prior to departure from scene. 46% of RASP patients bypassed a closer hospital. Only one patient was identified as not meeting RASP criteria and being incorrectly transported to the stroke center. Four patients (0.3%) patients were determined on chart review to have met RASP criteria but were not transported as such. Twenty-two (22)% of RASP patients received tPA.Conclusions: Paramedics are able to accurately apply a protocol that identifies patients with acute stroke and transport patients from a wide geographic area to a regional stroke center in time to receive thrombolytic therapy. Key Words: Stroke, Paramedics, Protocol

163 PARAMEDIC offload delays associated with prolonged response times

Cleve PD, Welsford M. Department of Emergency Medicine, McMaster University, Hamilton, ON

Introduction: Early access is the first link in the Chain of Survival for sudden cardiac death. With CPR, defibrillation within 8 minutes increases survival rate to 20%. The success rate can be doubled if defibrillation is provided within 4 minutes and paramedic arrival within 8 minutes. With ED overcrowding, offload delays are commonplace. The Hospital Emergency Department and Ambulance Effectiveness Working Group has published it's plan to reduce the offload times for paramedic crews to a target of 30 minutes. We conducted a retrospective review of offload times in the Hamilton area, correlating the response times with offload delays seen during the same 24 hour periods. Methods: All consecutive ambulance dispatches with an initial priority code of 3 or 4 were retrieved from the Hamilton Health Sciences Base Hospital Program's database for 2005. The average response time for each day was compared based on the percentage of calls within the same 24-hour period with a prolonged offload delay. An offload delay was defined a priori as being greater than 60 minutes from the time of arrival to hospital until the time reported as being clear of the hospital. Results: During the 2005 period studied, 161199 ambulance calls recorded with dispatch priority 3 or 4 were included in the study. 6.7% of calls had offload delays greater than 60 minutes. The average response time to the scene was 8m:16s. For every 10% of calls being delayed by more than 60 minutes, there is an associated 21 second delay arriving to the scene (p<0.01). Conclusions: Paramedic crews under supervision of the Hamilton Base Hospital Program are commonly delayed more than the Working Group's 30-minute target. A small, yet statistically significant, linear relationship exists between increasing offload delays and concomitant delays in the early access to paramedic care in these highest priority calls. Further work is needed to determine if clinical outcomes are affected by offload delays beyond the 30-minute target. Key Words: Paramedics, Offload times, Dispatch

164 QUALITATIVE analysis of a paramedic research question bank: a determinant of paramedic research priorities

Jensen J, Travers A, Patrick G. Emergency Health Services, Nova Scotia

Introduction: The Emergency Health Services (EHS) Prehospital Research Initiative's Paramedic Research Question Bank (QB) houses research questions developed by paramedics, administrators and educators in standardized evidence based medicine (EBM) format. The purpose of the QB is to encourage EBM practice within the paramedic profession and can establish paramedic research priorities. Methods: Research questions were submitted by paramedics during facilitator guided sessions and online submission in a provincial EHS system. The QB was qualitatively analyzed by a focus group that clustered the questions within three domains (clinical, systems and education) into themes and then established the research priority within each domain. This was achieved using a Delphi approach among a group of EHS experts. Results: The qualitative analysis was on questions entered into the QB during the time period of September 2004 to January 2006, in which there were 79 (51%) clinical questions, 17 (11%) education questions and 58 (38%) systems questions, a total of 154 questions. Through thematic analysis the clinical research priorities were found to be: pharmacology research, patient management decisions research, airway research and expanded clinical interventions research. Paramedic education delivery is the educational research priority and human performance and systems performance, and system status planning are research priorities in the systems domain. Conclusions: A paramedic QB is an effective means of determining the research priorities of paramedics within a provincial prehospital system. Key Words: Prehospital, Qualitative, Research priorities

165 INAPPROPRIATE claims of equivalence/non-inferiority in failed randomized trials in emergency medicine

Hall AK, Tang P, Fan J, Upadhye S. Division of Emergency Medicine, McMaster University, Hamilton, ON

Introduction: Failed superiority-designed randomized controlled trials (SupRCTs) often make inappropriate claims of equivalence(EQ) or non-inferiority(NI). We sought to examine how often this occurs in RCTs published in Emergency Medicine(EM) journals. Methods: Duplicated electronic and hand-searches of 6 leading EM journals were conducted to find all published RCTs in the 2005 calendar year. Articles were included if they were considered to be failed SupRCTs. Included articles were subsequently analyzed in detail using a data extraction tool based on previously published NI/EQ design and reporting standards. Agreement analyses of search strategies, inclusion decisions and individual article analyses were conducted. Results: The search yielded 64 reported RCTs, of which 17 were not truly comparison RCTs. Of the 47 remaining RCTs, 24 were excluded for explicit NI/EQ design (2) or successful differences reported (23). Of the remaining 22 failed SupRCTs, 13 made an appropriate claim of failure to demonstrate superiority, whereas 9 made false claims of NI/EQ. Conclusions: The majority of superiority-designed RCTs published in 2005 either successfully reported differences or made appropriate claims of failure. However, a significant proportion of failed superiority-designed RCTs (9/22) made inappropriate claims of equivalence or non-inferiority. Key Words: Methodology, Non-inferiority, Superiority, RCT

166 VALIDITY of patient-reported prescription filling as an outcome in emergency medicine research

Hohl CM, Abu-Laban RB, Zed PJ, Brubacher JR, Sobolev B, Tsai G, Kretz P, Nemethy K, Bjilsma JJ, Purssell RA. Department of Emergency Medicine, Vancouver General Hospital; Faculty of Medicine, University of British Columbia, Vancouver, BC

Introduction: Research on adherence with Emergency Department (ED) discharge prescriptions frequently utilizes unvalidated patient- reported outcome measures. In British Columbia, all prescriptions filled in community pharmacies are entered into a provincial prescription database (PharmaNet) through mandatory functions prior to dispensing. The objective of this study was to compare the validity of patient-reported prescription filling with a PharmaNet criterion standard. Methods: This was an a priori designed secondary analysis of a study examining adherence with ED discharge prescriptions. The study was carried out in a tertiary care center with an annual ED census of 65,000. Research assistants (RAs) collected information on consenting ED patients discharged with a prescription. Two weeks after the index visit RAs contacted patients by phone, and then used PharmaNet to determine whether patients had filled their prescription(s). Data on prescriptions for medications available over the counter was excluded. Identical a priori defined predictor variables for prescription filling were fit in two logistic regression models using each outcome measure for comparative purposes. Results: Data was obtained on 258 patients. During the phone interview 225 patients (87.6%) stated that they had filled their prescriptions compared with 207 patients (80.2.2%) on PharmaNet (difference 7.4%, 95% CI 1.0% to 13.6%, p=0.02). Both regression models identified the same predictor variable (illicit drug use) as statistically significant. The effect size estimates for this variable were similar and confidence intervals overlapped widely (PharmaNet outcome: OR 5.5, 95% CI 1.2-25.0; patient-reported outcome: OR 8.2, 95% CI 1.8-37.3). Conclusions: Despite yielding higher estimates of prescription filling using a patient-reported outcome measure, this measure yielded similar results in multivariate regression modeling compared to a criterion standard and has validity. Key Words:Prescription-filling, Validation, Discharge prescriptions

167 AN international survey of emergency physicians knowledge, use, and attitudes towards the Canadian CT head rule

Eagles D, Stiell IG, Clement CM, Brehaut JC, Taljaard M, Kelly A-M, Mason S, Kellermann A, Perry J. Department of Emergency Medicine, University of Ottawa, Ottawa, ON

Introduction: The derivation and validation of the Canadian CT Head Rule (CCHR) has been published in emergency medicine and general medical journals. Little, however, is known of its international diffusion and use. The purpose of this study was to determine the knowledge, attitudes and behaviour of emergency physicians (EPs) in Australasia, Canada, the UK and US regarding the CCHR. Methods: A prospectively conducted self-administered email and postal survey was sent to members of 4 national EP associations using a modified Dillman technique. Random samples of members from ACEM (Australasia), CAEP (Canada), BAEM (UK) and ACEP (US) were sent a prenotification letter followed by at least 4 mailouts. Awareness, use and attitudes regarding the CCHR were analyzed using descriptive and univariate statistics with 95% CIs. Results: Overall, 1150/2103 (54.7%) responses were received. The respondents were male (74%), mean age 42.5 years and had mean 12 years experience. Conclusions: Knowledge of the CCHR varied widely by country. The CCHR was viewed favourably across multiple measured dimensions. Of those aware, there is a moderately high level of use. Of those not currently using the CCHR, most would consider future use. A better understanding of the factors related to increased use of decision rules will facilitate strategies to enhance derivation, dissemination and implementation of future rules. Key Words: CT head, CCHR, Dissemination, Implementation

Table 1, Abstract 167
	Australia	Canada	UK	US
Response rate	54%	69%	44%	49%
Aware of CCHR	83%	87%	63%	35%
Use of CCHR	55%	83%	44%	29%
Consider future use	67%	84%	33%	77%
CCHR is useful in my practice	68%	87%	78%	74%
Patients benefit from use	69%	88%	70%	72%
Improves use of resources	66%	83%	53%	82%
Using another rule/strategy	38%	12%	60%	21%

168 SYSTEMATIC qualitative review of clinical decision rules for syncope in ED for predicting adverse outcomes

Thiruganasambandamoorthy V. Department of Emergency Medicine, University of Ottawa, Ottawa, ON

Introduction: Disposition of syncope patients is a challenging task for emergency physicians. The aim of this study was to systematically review the literature on decision rules for predicting adverse outcomes in adult syncope ED patients. Methods: Single reviewer systematically searched MEDLINE, EMBASE, CENTRAL, CINAHL and HTA databases until April 2006. English studies with decision rules for adverse outcomes were included. Data was extracted to calculate sensitivity, specificity and likelihood ratios (LR). Results: Of 8669 articles identified, 48 were reviewed fully. Only 6 met our eligibility criteria and they used 4 prediction instruments. Quinn's San Francisco Syncope Rule (SFSR) was the only prospectively derived and validated rule that included all serious adverse outcomes. Colivicchi's risk score was prospectively validated and predicts mortality only at one year. Elbesber and Crane retrospectively validated the ACEP guideline, drafted by consensus only and not derived by research. Crane's study predicted mortality at 1 year and had high loss to follow-up. Of the remainder Sarasin's was limited to arrhythmia prediction and Martin's to arrhythmia and mortality at 1 year. Both had poor Receiver Operator Characteristics curve range of 77-88 and 73-83% respectively. Conclusions: Studies were heterogeneous so meta analysis was not done. ACEP guideline was not rigorously derived. Only SFSR adhered to accepted guidelines for developing decision rule but was done in one site only. SFSR still needs further validation to ensure generalizability. Key Words: Systematic review, Syncope, Decision rule, SFSR

Table 1, Abstract 168
Study	Outcome	SENS (CI)	SPEC (CI)	+LR	-LR
Quinn04	Composite	96 (92–100)	62 (58–66)	2.53	0.06
Quinn06	Composite	98 (89–100)	56 (52–60)	2.23	0.03
Colivicchi03	Mortality	97 (81–100)	72 (67–78)	3.55	0.04
Elbesber05	Cardiac cause	100 (86–100)	33 (26–40)	1.49	0

170 IMPAIRED driving interventions: do not forget the passengers

Brubacher JR, Purssell RA, Brown D, Abu-Laban RB, Fang M, Edwin M, Schulzer M. Division of Emergency Medicine, University of British Columbia, Vancouver, BC

Introduction: Motor vehicle crashes (MVCs) kill over 2,700 Canadians annually. Drivers with an illegal blood alcohol content (BAC) cause 34% of all crashes involving fatalities, however minimal research has been done on the future driving behavior of passengers. The primary objective of this observational study was to determine the proportion of injured passengers treated in hospital, categorized by BAC, who subsequently engage in impaired driving activity (IDA).Methods: We retrospectively identified all passengers injured in an MVC who were treated in our tertiary care, urban Emergency Department (1999-2003) or registered in our provincial trauma registry (1992-2005) and had a BAC measured. Injured passengers were categorized into three groups according to their BAC: Group 1: BAC=0, Group 2: 0<BAC<=17.3 mmol/L BAC 3: Group and limit), (legal>17.3 mmol/L. IDA was determined from police records (1989-2005), and defined as any of the following: a conviction for impaired driving; a 24hr or 90 day license suspension for impaired driving; involvement in a MVC where police listed alcohol as a factor; or presentation to a hospital with a BAC above the legal limit following an MVC. Results: 1336 passengers met inclusion criteria: 713 in Group 1; 174 in Group 2; and 449 in Group 3. IDA following the index hospital visit was identified in 78 passengers in Group 1 (10.9%, 95% CI 8.8%-13.4%); 48 passengers in Group 2 (27.6%, 95% CI 21.3%-34.6%, p<0.001 vs Group 1); and 158 passengers in Group 3 (35.2%, 95% CI 30.9%-39.7%, p<0.001 vs Group 1). Many passengers had engaged in IDA prior to their index visit: 12.8% in Group 1 (95% CI 10.5%-15.4%); 24.1% in Group 2 (95% CI 18.2%-30.9%); and 40.1% in Group 3 (95% CI 35.6%-44.7%). Conclusions: Passengers injured in a MVC frequently engage in subsequent impaired driving, particularly those who have a BAC above the legal limit. These findings suggest that rehabilitation programs and legal efforts targeting impaired or high risk drivers should also target high risk passengers. Key Words: Impaired driving, Alcohol, Motor vehicle collision

171 ANTERIOR shoulder dislocation reduction: current practice

Usher DJ, Woolfrey K, Cacic I, Cleve P. Division of Emergency Medicine, McMaster University, Hamilton, ON

Introduction: The management of anterior shoulder dislocations in the emergency department presents several important issues. Despite being a relatively straightforward injury to deal with, there is no gold standard approach and a simple reduction can prove costly and time consuming. The purpose of this study was to evaluate time spent in the ED, resources required, success rates and complications. Methods: The study included both an online survey and a medical record review. The survey was submitted to 56 ER physicians in our region. Our chart review included 38 anterior shoulder dislocations analyzed by two individual data abstractors using a computerized abstraction form. Results: The response rate for our survey was 66%. Overall, perceptions were that shoulder reduction has a high success rate with few complications and typically patients wait roughly 30-60 minutes for their procedure and an additional 30-60 minutes from the procedure to discharge. The Chart review demonstrates that the wait times typically are greater than 2hrs for shoulder reduction and an additional 2 hrs to be discharged. Conclusions: Current management of anterior shoulder dislocation reductions prove to be very time consuming and require many ER resources. Several studies have demonstrated that intraarticular lidocaine is an alternative to procedural sedation but further studies are needed. If effective, this technique may significantly reduce ER time, cost and resources. Key Words: Shoulder dislocation, Wait times, Shoulder reduction

172 ASSOCIATION of injury mechanism with the risk of cervical spine fractures

Thompson WL, Stiell IG, Clement C, Wells G, Brison RJ, for the Canadian C-Spine Rule Study Group. Departments of Community Health and Epidemiology and Emergency Medicine, Queen's University, Kingston, ON; Departments of Epidemiology and Community Medicine and Emergency Medicine, University of Ottawa, Ottawa, ON

Introduction: A full understanding of an injury event and the mechanical forces involved should be important in predicting specific anatomic patterns of injury. Yet, information on the mechanism of injury is often overlooked as a predictor for specific anatomic injury in clinical settings. This study measures the relationship between mechanism of injury and risk for cervical spine fracture. Methods: This case-control study is a secondary analysis of data collected from the Canadian C-Spine Rule (CCR) Study. Data were collected from 1996 to 2002 and included patients presenting to the emergency department of nine tertiary care centres after sustaining acute blunt trauma to the head or neck. Cases are patients categorized in the CCR study as having a clinically important cervical spine fracture. Controls had no radiological evidence of cervical spine injury. Bivariate and multivariate unconditional logistic regression models were used. Results are presented as odds ratios with 95% confidence intervals. Results: Among the 17208 patients in the CCR study, 320 (1.9%) were diagnosed with a cervical spine fracture. Axial loads, falls, diving and non-traffic motorized vehicle collisions were injury mechanisms significantly related to an increased risk of fracture. For motor vehicle collisions, the risk of cervical spine injury increased with posted speed, being involved in a head-on collision, a rollover or not wearing a seatbelt (p<0.05). Cervical spine fracture occurrence was negligible in simple rear end collisions (1 in 3694 cases; OR=0.015, 95% CI=0.002-0.104). Conclusions: This study demonstrates quantitatively the relationship between specific mechanisms of injury and the risk of a cervical spine fracture. Knowledge of injury mechanism will assist emergency health providers in their care of trauma patients. Key Words: CCR, Mechanism, Injury, Fracture risk

175 IMPACT of a difficult airway course on participants perceived ability

Dowling SK, Donnon T, Rigby I, Lord J. Departments of Emergency Medicine, Community Health Science, and Critical Care Medicine, University of Calgary, Calgary, AB

Introduction: Airway management is an essential skill for both critical care and emergency physicians as well as certain allied health care providers. Despite being a life-saving skill, there can be significant morbidity and mortality associated with failure to appropriately manage a difficult airway. Recently, a group of Calgary Health Region (CHR) physicians developed a Difficult Airway Management Course targeting residents, staff and allied health care providers. We conducted a pre-post course questionnaire to assess the impact of the CHR Difficult Airway Management Course on participant's self-reported ability, knowledge, and comfort with this subject matter. Methods: The pre-post questionnaire assessed the participant's self-reported knowledge (general understanding of procedure), skills (ability to perform procedure), and attitudes (comfort level with procedure) using 12 questions on a 5-point scale from poor to excellent. We used a student's t-test to compare the pre- and post-course results. Results: The response rates for the pre-post course questionnaires were 96% (52 of 54) and 56% (30 of 54) respectively. Of the participants, 46% were from the division of Emergency Medicine, 27% from Respiratory Therapy, and 15% were from Critical Care. A comparison of the pre- and post-course survey results showed that participants reported improved knowledge [t (76) = 3.76, p <.001], skills [t (72) = 2.97, p < .01] and attitudes [t (71) = 3.31, p < .001] with respect to the airway topics covered during course. Conclusions: The CHR Difficult Airway Management course resulted in significant improvements in participants' perceived knowledge, skills and comfort with respect to difficult airway management. Key Words: Difficult airway, Medical education, Self-assesment

176 EFFECT of simulation based training on self-assessment of comfort with management of hemodynamically unstable patients

Langhan TS, Walker I, Rigby I, Donnon T, Lord JA. Department of Emergency Medicine, Foothills Medical Centre, University of Calgary, Calgary, AB

Introduction: Procedural skills training in emergency has traditionally been taught through a process referred to as "see one, do one, teach one". New ethical arguments support achieving competence in procedural skills prior to attempts with actual patients. We sought to assess the impact of procedural skills simulation training on the self-assessed competence of Emergency Medicine residents in performing critical resuscitation skills. Methods: A convenience sample of Emergency Medicine residents were surveyed for self-assessed measures of competence in the procedural skills required to manage hemodynamically unstable patients. Study participants were then exposed to an intensive 8 hour simulation based training program focusing on resuscitation procedures, in particular psychomotor skill acquisition. At the completion of the simulation course, the subjects were asked to repeat the self-assessment survey tool. Repeated measures MANOVA was used to assess interval changes in residents' self-assessments of competence. A follow-up survey 3 months after simulation training to assess confidence retention is pending. Results: Data analysis is ongoing. Preliminary results indicate an improvement in self-assessed competence measures following the simulation intervention (p < 0.05). Conclusions: The importance of simulation based medical education cannot be overstated. Emergency Medicine residents acquired additional confidence with resuscitation procedural skills following an intensive simulation based training program. We anticipate that this study, in concert with further studies proving knowledge transfer from the simulation environment to the clinical environment, will drastically change how procedural skills are taught. Most importantly, patient safety is the driving force behind this study. No longer is it acceptable to "practice' on patients. Achieving procedural competence without compromising patient safety is paramount, and support for simulation based medical education is a step in that direction. Key Words: Simulation, Resuscitation, Procedural skills, Education

177 RESIDENT code blue survey: confidence and preparedness for running cardiac-arrest resuscitations

Education/Teaching. Sartor JR, Howes D. Department of Emergency Medicine, Queen's University, Kingston, ON

**please note: research is currently ongoing and expected completion date is Feb 15, 2007

Introduction: In most institutions, in-hospital cardiac-arrest resuscitations (Code Blues) outside of the emergency department are coordinated and administered by residents. The type and quality of training provided for residents to act as team leader during these cardiac-arrests is highly variable; this prior training, along with past experience, residency program and level of residency may all impact the quality of care given to patients. We will be conducting an on-line anonymous survey of residents who attend residency programs at one of the five Ontario medical universities regarding their preparedness to act as team leaders during cardiac arrest resuscitations. Methods: An anonymous voluntary on-line survey directed to persons enrolled in recognized residency programs at one of five Ontario universities. Results: Results will evaluate resident confidence in their ability to run cardiac-arrest resuscitations and compare this to the school they attend, their residency level, their residency program, and their prior experience and training with cardiac-arrest resuscitations. Conclusions: It is hoped that conclusions established from this survey may contribute to the development or improvement of training programs in cardiac-arrest management. Key Words: Education, Cardiac arrest, Resuscitation, Preparedness, Residents

181 A lesson in history - clinical history predicts acetaminophen toxicity in children

Goldman RD, Malik K, Verjee Z, Cohen E. Pediatric Research in Emergency Therapeutics (PRETx) Program, Division of Pediatric Emergency Medicine, The Hospital for Sick Children, Child Health Evaluative Sciences, The Research Institute, Department of Pediatrics, University of Toronto, Toronto, ON

Introduction: Acetaminophen is frequently reported as an agent of toxic ingestions, both accidental and non- accidental. We aimed to determine the diagnostic test properties of history in the diagnosis of acetaminophen toxicity in children. Methods: This is a retrospective chart review in a tertiary pediatric emergency department (ED) in Canada. We included all children that had acetaminophen levels drawn in the ED. The main outcome measures were sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). Results: We included 1989 acetaminophen levels drawn during 1905 unique patient visits. Of 1886 analyzed tests, 13 children were considered "toxic" and 55 children were considered "possibly toxic". All but 2 had a clear history of ingestion on arrival to the Emergency Department. There was a statistically significant association between a positive history of ingestion and toxicity (p<.0001). This association did not change with re-classification of the "possible-toxicity" group. The overall odds ratio for toxicity when a history of overdose was given was 71.9 (95% CI 17.5 - 294.5). If all children with a "possibly toxic" acetaminophen levels were considered "toxic" the Sensitivity, Specificity, PPV and NPV (95% CI) for history was 0.97 (0.90 - 1.0), 0.69 (0.66-0.71), 0.10 (0.08-0.13), 0.998 (0.994 -1.0) respectively. Conclusions: Unlike in adults, in children a history of ingestion is an excellent predictor for possibly toxic acetaminophen plasma levels. This may result in a significant reduction of number of acetaminophen levels drawn in the ED. Key Words: Acetaminophen, Toxicity, Toxicology, Predictors

182 PARENTERAL base therapy in pediatric critical illness: systematic review

Parker MJ, Parshuram CS. Divisions of Emergency Medicine and Critical Care Medicine, The Hospital for Sick Children, The University of Toronto, Toronto, ON

Introduction: Current practice guidelines do not provide clear indications for the use of base therapy in acute severe illness. In clinical practice base therapy is prescribed frequently by physicians caring for critically ill children. Methods: We conducted a systematic review of published literature from 1966 to August, 2006. With an academic librarian we developed a broad search strategy for 7 electronic databases (Medline, EMBASE, CINAHL, Cochrane DSR, ACP Journal Club, DARE, CCTR). We included a variety of subject headings relating to base therapy and critical illness. Eligible studies included parenteral administration of base to children 1 month-18 years who had acute severe illness. We excluded case reports with <4 patients, outpatient studies, and non-English language publications. Articles identified in the primary search were reviewed for exclusion criteria based on title and abstract. The remaining articles were reviewed in full. We abstracted the population, intervention and study design. Results: 1183 articles were identified in the primary search. 125 articles were assessed in full. 11 met inclusion criteria. One randomized controlled trial of 60 adults and children with acute organophosphate poisoning found high dose sodium bicarbonate reduced hospital stay. 10 pediatric observational trials were found. All were retrospective. The location of patients was ICU(5), ER(1), and other in-patient(4). Sodium bicarbonate was the intervention of interest in all studies. Diseases included diabetic ketoacidosis (4), severe asthma(2), need for continuous renal replacement therapy(2), gastroenteritis-induced metabolic acidosis(1), and ethylene glycol intoxication(1). Conclusions: In contrast to the frequent use of sodium bicarbonate in acutely ill children, we found few pediatric studies of this or any base therapy. Further high-quality research including randomized trials is needed to facilitate the development of evidence-based guidelines and inform clinical practice. Key Words: Base therapy, Sodium bicarbonate, Critical care, Pediatrics

183 THE effect of concentrating periods of physical activity on the risk of injury in organized sports in a pediatric population

Fecteau D, Gravel J, D'Angelo A, Amre D, Martin E. Department of Pediatrics, University of Montréal, Montréal, QC

Introduction: Biopsychological markers suggest that overtraining syndrome arises from a continuum and is perpetuated by an imbalance between rest and effort. The current trend in pediatric sport organizations is to regroup activities into weekend tournaments. This practice could theoretically increase the risk of sports injuries given that there is a lack of rest periods. Objectives : To evaluate the association between the risk of sport injury and the effect of consolidation of physical activity for periods of 48 hours and 7 days in children. Methods: A case-crossover study was conducted in an emergency department of a tertiary care hospital. To be included, participants had to be aged between 8 to 16 years old and acutely injured in an organized sport-related event. A standardized questionnaire was used to evaluate the amount of physical activity performed during 28 days prior to the injury. The case periods were the 48 hours and 7 days prior to the injury and were compared to control periods of similar length. The primary analysis was the comparison of the number of hours of activity between the case and control periods. It was prospectively determined that a clinically significant difference is 1 hour for the 48 hours study and of 2 hours for the 7 days study. Results: The mean age of the 48 recruited patients was 13.25 years. The injuries occurred mainly during soccer, basketball, football and hockey. On average, participants performed 356 minutes of organized sports per week and 136 minutes in the 48 hours preceding the injury. Participants reported an increase of 13±24 minutes for the case period in the tournament effect study and of 40±47 minutes for the 7 days prior the injury. Conclusions: The present study failed to disclose any association between the consolidation of organized sport activity and the risk of injury in children. A limitation of the study relies in its retrospective aspect and conducting a prospective study of a large number of athletes would allow a more complete collection of data. Key Words: Organized sports, Injury, Physical activity, Pediatrics

184 A needs assessment for obesity-related anticipatory guidance in the pediatric emergency department

Warkentin J, Chan M, Igric A, Seabrook J, Matsui D, Joubert G. Department of Paediatrics, University of Western Ontario, London, ON

Introduction: Paediatric obesity is a major concern in industrialized nations and is recognized by the Canadian Paediatric Society as an important anticipatory guidance (AG) topic. Our objective is to determine the proportion of non-urgent patient-families interested in obesity-related anticipatory guidance while they wait in the Paediatric Emergency Department (PED). Methods: A survey was distributed to patients and families in the waiting area of the PED of the Children's Hospital of Western Ontario (CHWO). Results: 200 people were surveyed; 92% were parents and 4% were patients. 51% of families had someone in the family dieting, exercising or both for weight loss and 17% of respondents believed at least one child in their family considered themselves overweight or obese. Respondents that considered themselves overweight or obese were significantly more likely to state their children also consider themselves overweight or obese (22.8% vs. 11.1%, p=0.03). This group was also significantly less interested in AG (66.7% vs. 90.1%, p=0.009). Conversely, the unemployed were significantly more interested in information about nutrition or exercise compared to those that were employed (49.0% vs. 31.2%, p=0.023). 41% of all respondents were interested in further information on obesity in the PED, 91% of whom were interested in information on both nutrition and exercise. Of the 22.4% of patients that felt their physician was not meeting their needs for nutrition or exercise-related AG, 76.3% were interested in further information (p<0.001). The preferred method of knowledge dissemination was via brochure (31.5%) followed by an internet-based resource (12%). Conclusions: Obesity is a major AG topic for physicians of all specialties. It is important to address this issue at every opportunity and convey our message with the highest impact possible. Our study demonstrates that a segment of the PED population would be receptive to AG in the PED and suggests different health care environments require diverse AG approaches. Key Words: Obesity, Pediatrics, Prevention, Anticipatory guidance

185 COMPARISON of four pain scales in children with acute abdominal pain in a pediatric emergency department

Bailey B, Bergeron S, Gravel J, Daoust R. Division of Emergency Medicine, CHU Sainte-Justine, Université de Montréal, Montréal, QC

Introduction: Pain documentation is becoming more and more frequent in the emergency department. In children, many pain scales may be used. However, their interchangeability, especially in the evaluation of acute pain, is not well known. The objective of our study was to evaluate the agreements between the visual analog scale (VAS), the standardized color analog scale (CAS), the Wong-Baker Faces Pain Rating Scale (WBFPS), and a verbal numeric scale (VNS) in children with acute abdominal pain suggestive of appendicitis in a paediatric emergency department (PED). Methods: Participants were children aged 8 to 18 years old presenting to a PED with abdominal pain suggestive of appendicitis recruited to participate in a randomized controlled trial evaluating the efficacy of morphine. Patients were initially asked to grade their pain on a plasticized CAS, a paper VAS, a paper WBFPS, and then with a VNS. Thirty minutes after morphine or placebo administration, the same assessment was repeated. All scales were then converted to a 0 to 100 mm; agreements between scores were evaluated using the intraclass correlation (Ric) and the Bland-Altman method. Results: A total of 87 children were included in the study, 58 of them with confirmed appendicitis. All pain scales were correlated between each others (Ric between 0.88 and 0.97, p<0.001). However, except for the VAS/CAS comparison, the limits of agreement between each other were all more than our a priori limit of acceptability of ± 20 mm: VAS/CAS -18.6, 14.4; VAS/WBFPS -20.1, 33.7; VAS/VNS -30.2, 20.8; CAS/WBFPS -18.5, 36.3; CAS/VNS -26.9, 22.1; WBFPS/VNS -38.7, 15.7. Also, the measurements differed by more than 20 mm in more than 10% of the patients for all the scales comparison except for VAS/CAS. Conclusions: Our study suggests that only the results from the VAS and the CAS are interchangeable in children with acute abdominal pain suggestive of appendicitis. In particular, the VNS is not interchangeable with the other evaluated scales. Key Words: Pain Scales, Interchangability, Appendicitis, Pediatrics

186 SAFETY of the modification of the triage level for children 6 to 36 months old with fever

Gravel J, Manzano S. Division of Emergency Medicine, Hôpital Sainte-Justine, Montréal, QC

Introduction: The Paediatric Canadian Triage and Acuity Scale (PaedCTAS) was implemented in Canadian paediatric emergency department (ED) in 2001. An item of the PaedCTAS stipulates that patients aged between 3 and 36 months with fever must be triaged as urgent. To maximize resources use, a modification has been applied in our institution since 2001. The protocol stipulates that children 6 to 36 months old with fever and no sign of toxicity could be triaged as semi-urgent. Objective: Evaluate whether children 6 to 36 months old with fever but no sign of toxicity and triaged as semi-urgent behave as all the other patients triaged semi-urgent. Methods: A retrospective cohort study evaluating all patients triaged in an urban tertiary paediatric hospital during a 6 month period was performed. Data were retrieved from the ED database (Staturg® from Statdev®). Hospitalisation rate was compared for 4 groups: 1. children aged 3 to 36 months with fever triaged urgent, 2. children aged 3 to 36 months with fever triaged non-urgent (following the protocol), 3. all the patients triaged urgent, and 4. all the patients triaged semi-urgent. Results: During the study period, there were 36,285 visits at the ED and 3,477 visits for children aged 3 to 36 months with fever. Of these, the nurses triaged 1322 (38%) patients as urgent and 1869 (54%) patients as semi-urgent. The proportion of hospitalisation for children aged 3 to 36 months with fever triaged urgent was similar to the proportion of all the patients triaged urgent (13.1% vs.13.4%) while the proportion of hospitalisation for children aged 3 to 36 months with fever triaged semi-urgent was similar to the one of all the patients triaged semi-urgent (2.3% vs. 2.0%). No patient downgraded as semi-urgent were admitted to the intensive care unit. Conclusions: Patients aged 6 to 36 months with fever and no sign of toxicity can be triaged as semi-urgent safely. This would modify the triage level of approximately 5% of the population visiting a paediatric ED. Key Words: CTAS, Triage level, Fever, Infants

187 CANADIAN best practice recommendations for stroke cCare (2006) - stroke prevention in the emergency department

Harris DR, Lindsay MP, Witt J, McDonald AM, Phillips SJ, on behalf of the Canadian Stroke Strategy (a joint initiative of the Canadian Stroke Network and the Heart and Stroke Foundation of Canada). Department of Emergency Medicine, St. Paul's Hospital, Vancouver, BC

Introduction: The Canadian Stroke Strategy Best Practices and Standards Working Group (BPS-WG) was developed to aid in the translation of research into practice for stroke care. The goal of the BPS-WG was to recommend best practices in stroke care appropriate to the Canadian context and based on the highest levels of evidence available. This abstract focuses on the recommendations for secondary stroke prevention that are relevant to emergency physicians. Methods: The recommendations were developed by systematically searching for existing practice guidelines, appraising their quality using a validated tool, selecting recommendations for inclusion in the final document, and obtaining expert feedback on the proposed recommendations. Core performance measures were identified for each recommendation. Emergency physicians played an active role on the working group through all stages of development and review. Results: Of a total of 24 recommendations, seven relate to primary and secondary prevention of stroke, and are relevant to emergency physicians: 1. Life style management (exercise, smoking, diet, weight, alcohol and stress); 2. Blood pressure management; 3. Lipid management; 4. Diabetes management; 5. Antiplatelet therapy; 6. Antithrombotic therapy for atrial fibrillation, and; 7. Carotid intervention. In addition, recommendations also stressed the necessity of public awareness, and patient and caregiver education across the continuum of care. Conclusions: The Canadian Best Practice Recommendations for Stroke Care: 2006, provide guidance to physicians who are responsible for initiating or modifying secondary stroke prevention therapy for patients who present with TIA or minor stroke. Key Words: Stroke, Prevention, Best practice, Secondary prevention

188 CANADIAN best practice recommendations for stroke care (2006) - acute stroke management

Introduction: The Canadian Stroke Strategy Best Practices and Standards Working Group (BPS-WG) was developed to aid in the translation of research into practice for stroke care. The goal of the BPS-WG was to recommend best practices in stroke care appropriate to the Canadian context and based on the highest levels of evidence available. This abstract focuses on the recommendations for acute stroke management that are relevant to emergency physicians. Methods: The recommendations were developed by systematically searching for existing practice guidelines, appraising their quality using a validated tool, selecting recommendations for inclusion in the final document, and obtaining expert feedback on the proposed recommendations. Core performance measures were identified for each recommendation. Emergency physicians played an active role on the working group through all stages of development and review. Results: Of a total of 24 recommendations, eight relate to primary and secondary prevention of stroke, and are relevant to emergency physicians: 1. Acute stroke unit care; 2. Brain imaging; 3. Blood glucose; 4. Acute thrombolytic treatment; 5. Carotid artery imaging; 6. Dysphagia assessment; 7. Acute aspirin therapy, and; 8. Management of subarachnoid and intracerebral hemorrhage. In addition, recommendations also stressed the necessity of public awareness of the signs and symptoms of stroke, and patient and caregiver education across the continuum of care. Conclusions: The Canadian Best Practice Recommendations for Stroke Care: 2006, provide guidance to physicians who are responsible for acute stroke management. Key Words: Acute stroke, Management, Best practice

189 OUTCOME of patients with an emergency department discharge diagnosis of transient ischemic attack

Harris DR, Grafstein E, Stenstrom R, Innes G, Hunte G. Department of Emergency Medicine, St. Paul's Hospital, Vancouver, BC

Introduction: Patients who experience a transient ischemic attack (TIA) are at high short-term risk of stroke. The purpose of this study was to determine the outcome of patients with an emergency department (ED) discharge diagnosis of TIA (ICD-9CM 435.9) as coded by the emergency physician within one year after the event. Mehtods: We undertook a retrospective study in a tertiary care, urban academic hospital in Vancouver, British Columbia utilizing the emergency department administrative database linked with provincial vital statistics data and the regional database of emergency encounters. Only patients with the primary diagnosis of TIA were analyzed, from January 2001 to August 2005. We were interested in the incidence of stroke, recurrent TIA, cardiovascular event (angina, myocardial infarction) or death within 90 days and 1 year after the index event. Results: Over the 55 month period, 329 patients had a discharge diagnosis of TIA from the ED. 184 were men and the average age was 67 years. 72 patients (22%) were referred to neurology in the ED and 42 (12.8%) were admitted to hospital on the index event. Of those admitted, median length of stay was 3.25 days (IQR = 1.8, 5.6). Of those patients discharged, 8 (2.4%) were readmitted to an acute care hospital within 90 days with a diagnosis of stroke. Outcomes are outlined further in the Table. Conclusions: In this study, the risk of stroke, recurrent TIA, cardiovascular event or death is high after a TIA, yet lower than other published estimates. Interventions must be directed at patients presenting with TIA to the ED to prevent further cardiovascular events. Key Words: TIA, Stroke, Short-term risk, Prevention

Table 1, Abstract 189
Event	Number (percent)
TIA < 90 days	13 (4.0)
TIA > 90 days & < 1 year	2 (0.6)
CVA < 90 days	8 (2.4)
CVA > 90 days & < 1 year	2 (0.6)
ACS < 1 year	1 (0.3)
Death < 90 days due to TIA/CVA	1 (0.3)
Composite for 90 days	22 (6.7)
Composite for 1 year	27 (8.2)