2004 CAEP/ACMU Scientific Abstracts - Poster Presentations (#53 to #75)

CAEP Abstracts

CJEM 2004;6(3):173-212

abstracts:  001-016 | 017-032 | 033-052 | 053-075 | 076-092 | 093-111 | 112-132

053 Learning curve for paramedics administering a reconstituted study drug infusion during cardiac arrest resuscitation in a randomized controlled trial.

Abu-Laban RB, van Beek CA, Christenson JM, Ip J, McKnight RD, MacPhail IA, Dailly I, Woods R. Division of Emergency Medicine, University of British Columbia, Vancouver, BC

INTRODUCTION: The "tPA in PEA Study" involved advanced life support paramedics assessing enrollment criteria, providing standard therapy, and administering a reconstituted study drug infusion during CPR (Abu-Laban et al, NEJM 2002; 346:1522-8). As there is little research on paramedic training for clinical trials, we sought to test the hypothesis that paramedic experience would result in a "learning curve" reduction in time from eligible rhythm to commencement of study infusion over the course of the trial. Secondarily, we evaluated protocol violations over the trial. METHODS: Training for the tPA in PEA Study involved direct education, a 20-minute training video, a written orientation package, and a pocket card. Ongoing feedback was provided on enrolled patients and inadvertent exclusions. The study database contains prospective Utstein-style data on 233 patients randomized to receive a study drug infusion. Using an explicit approach, this database was reviewed for information on the rapidity of study drug commencement and protocol violations over the course of the trial. Appropriate statistics for trend were applied with a p <= 0.05 significance threshold. RESULTS: A linear regression showed an insignificant increase in the time from eligible rhythm to study drug infusion over the course of the trial (0.005 minutes/day, p = 0.13). Protocol violations occurred in 26 cases (11.2%), and the Cochrane-Armitage test showed a downward trend in the number of protocol violations for quartiles of the study population (p = 0.03). A smoothed plot demonstrated that protocol violations declined steadily until day 300 of the 596-day trial. CONCLUSIONS: In a cardiac arrest study with a structured training and feedback program, we found no evidence of a reduction in time from eligible rhythm to commencement of study drug infusion. Protocol violations, however, steadily declined over the first half of the study. These findings have implications for planning and training in future prehospital studies.

Key words: cardiac arrest, resuscitation, emergency medical services

054 Predictors of survival for out-of-hospital chest pain patients in the OPALS study.

Stiell IG, Ong MEH, Nesbitt L, Jaffey J, Wells GA, Beaudoin T, for the OPALS Study Group. Department of Emergency Medicine, University of Ottawa, Ottawa, ON

INTRODUCTION: We previously demonstrated the addition of an ALS EMS program led to a reduction in mortality and improvement for chest pain patients. In this study, we sought to determine which specific factors were associated with better survival. METHODS: This multicenter before-after controlled clinical trial was part of the Ontario Prehospital Advanced Life Support (OPALS) Study, looking specifically at chest pain patients. During the before phase, care was provided at the BLS-D level. During the after phase, ALS providers performed endotracheal intubation and administered IV drugs. We performed a stepwise logistic regression analysis to identify independent predictors of survival, regardless of phase. RESULTS: 10,089 patients were enrolled during two 9-month BLS and ALS phases. The Table (see next page) shows the predictors with % of survivors (N = 9,742) and non-survivors (N = 345), odds ratios for survival with 95% CIs. CONCLUSION: This largest controlled trial of out-of-hospital chest pain patients shows that blood pressure, EMS code "life threatening", and myocardial infarction impact survival. Regarding interventions and survival, only nitroglycerin was positively associated while both furosemide and intubation were negatively associated. Randomized trials are needed to further assess the value of specific EMS treatments for chest pain.

Key words: cardiac arrest, resuscitation, emergency medical services

055 Vaccine delivery by paramedics for an urban influenza immunization program: A public health - EMS partnership.

Schwartz B, Henry B, Shaddock D, Varia M, Popov D. Department of Emergency Medicine, Sunnybrook & Women's Health Sciences Centre, University of Toronto, Toronto, ON

INTRODUCTION: Immunization in hard-to-reach populations (e.g. homeless) is a challenge for Public Health. Barriers include nurses' reluctance to attend shelters, tent cities and other venues due to safety concerns, lack of acceptance of public health initiatives by residents, and lack of access to traditional public health clinics. Emergency Medical Services (EMS) paramedics regularly interact with these populations, thus providing an opportunity to improve their immunization status. We describe an innovative partnership between EMS, Base Hospital and Public Health designed to administer influenza vaccine to individuals in hard-to-reach urban populations. METHODS: Our Immunization Program offers influenza vaccine free of charge using a combination of delivery strategies. Paramedics were trained in the administration of influenza vaccine, under indirect medical oversight. Efforts were made to reach populations that were heretofore not accessed by Public Health. Standardized data forms were used to determine the number of people immunized, the profile of vaccine recipients and the public response towards influenza vaccination and the various delivery strategies to inform the planning of future clinics. RESULTS: Influenza vaccine was provided to 898 individuals from 2000-2003, at drop-in centers, shelters, detoxification units and community housing projects. Average age was 43.3 years (range 16-90) and 83.6% of patients were male. 286 (32.5%) persons gave their residence as no fixed address and 300/794 (37.7%) of respondents had never before received influenza vaccine. Immunization was not performed on 34 (3.8%) patients due to contraindications. 734/735 respondents stated they would use the service again. CONCLUSIONS: Provision of the influenza vaccine through a partnership between Public Health and EMS, can be effective in increasing influenza immunization rates in the community and particularly high-risk populations.

Key words: emergency medical services, influenza, public health, vaccination

056 Heat-related illness and emergency medical services utilization.

Shaddock D, Varia M, Henry B, Schwartz B. Department of Emergency Medicine, Sunnybrook & Women's College Health Sciences Centre, University of Toronto, Toronto, ON

INTRODUCTION: Heat-related illness affects susceptible individuals, causing increased emergency volumes, hospital admissions and mortality. The impact of heat-related illness on Emergency Medical Services (EMS) Systems has not been described. Our objective was to determine whether heat alert (HA) days were associated with increased EMS utilization. METHODS: Emergency transports and use of selected interventions on days defined as heat alerts in July and August 2002 were compared with those on days that were not so designated. A priori conditions deemed to be associated with heat-related illness (e.g. asthma) were compared to conditions that were unlikely to be heat-related (e.g. minor trauma). Public health determination of HA days was based on climate conditions and historical mortality data. RESULTS: There were 16 HA days from July 1 to August 31, 2002. The number of EMS transports on HA days (mean 459.3; 95% CI, 440.7 to 478.0) was slightly higher than on non-HA days (439.7; 95% CI, 430.0 to 449.4; p = 0.056). There were a significantly higher number of transports associated with heat-related illness on HA days (mean 73.0; 95% CI, 68.9 to 77.1) compared to non-HA days (mean 64.1; 95% CI, 61.1 to 67.2; p = 0.003). There was no difference in the mean number of non-heat related transports on HA days compared to non-HA days. CONCLUSIONS: Heat alert days are associated with a higher number of EMS transports, with a particular increase in calls associated with heat illness. Training of EMS staff should prepare them for increased call volumes on heat alert days and should include recognition and management of persons who have disorders which are exacerbated by hot weather.

Key words: emergency medical services, heat illness

057 Half a million strong: The Emergency Medical Services (EMS) response at the Toronto Rolling Stones Concert.

Feldman MJ, Lukins JL, Schwartz B, Burgess RJ, MacDonald RD. Department of Emergency Medicine, Sunnybrook & Women's College Health Sciences Centre, University of Toronto, Toronto, ON

INTRODUCTION: Although the medical response to mass gatherings has been described in the literature, there are few published reports of very large single-day gatherings involving hundreds of thousands of spectators. We describe the experience and outcomes of the EMS response to the largest single-day ticketed concert held in North America. METHODS: Medical care at the outdoor summer rock concert was provided between 0800 and 0200 hours by volunteer first aiders, paramedics, physicians and nurses. Sites of care included ambulances, medically-equipped golf carts, bicycle paramedic units, first aid tents, and a 234-bed medical facility that included a field hospital, a rehydration unit, and a paramedic screening desk at the entrance. Records from the tents, ambulances, screening desk, and rehydration unit were obtained. Data abstracted included patient demographics, chief complaint, time of incident, treatment, and disposition. RESULTS: Over 450,000 people attended the concert and 1870 sought medical care (42 per 10000). No records were kept for those seeking only water, sunscreen, or bandages. Records were obtained for 1205 patients, whose average age was 28±11 years, and 61% were female. 795 patients (66%) were cared for at one of the first aid tents, as the crowd sometimes restricted access to the medical facility. The common chief complaints included headache (321 patients; 27%), dehydration or heat-related illness (148; 12%), nausea or vomiting (91; 7.6%), musculoskeletal complaints (83; 6.9%), and shortness of breath or asthma (79; 6.6%). Peak activity occurred between 1400 and 1900 hours, when an average of 100 patients per hour sought medical attention. Twenty-eight patients (0.6/10000) were transferred to off-site hospitals. CONCLUSIONS: Outdoor rock concerts are considered high-risk events in the mass gathering medicine literature. The EMS response represents a significant public safety, medical, and logistical undertaking. Our experience may assist EMS planners at future large-scale mass gatherings.

Key words: mass gatherings, emergency medical services

058 Implementation and outcomes of an EMS-based rehydration unit at a mass gathering.

Lukins JL, Feldman MJ, Verbeek PR, Summers J. Department of Emergency Medicine, Sunnybrook & Women's College Health Sciences Centre, University of Toronto, Toronto, ON

INTRODUCTION: Heat-related illness is a significant cause of illness at outdoor mass gatherings during summer. We describe the experience and outcomes of a medical rehydration unit staffed by paramedics using medical directives developed for heat-related illness. METHODS: The 48-bed rehydration unit was adjacent to a field hospital at the largest single-day ticketed rock concert held in North America, with an estimated attendance of 450,000. Paramedics initiated rehydration therapy for patients with mild to moderate heat-related illness. Patients were also reviewed by the supervising base hospital physician. Data abstracted from patient charts were analysed for patient demographics, chief complaint, admission times, length of stay, treatment, and disposition. RESULTS: 143 patients (3/10,000 attendees) were managed in the rehydration unit. The average age was 24 ± 9 years, 72% were female. The common chief complaints were syncope or presyncope (21%), nausea or vomiting (16%), and headache (13%). 40% of patients received parenteral fluids. The average length of stay was 94 ± 67 minutes. 107 patients (75%) were discharged to self-care, 17 (12%) were transferred to the field hospital, 4 patients left against medical advice, 2 required transfer to a hospital off-site and in 12 cases, records of patient disposition were incomplete. CONCLUSIONS: A paramedic-staffed rehydration unit designed to manage heat-related illness at a mass gathering has not been previously reported. This strategy can divert large numbers of patients from requiring treatment at field hospitals and could be used by health care planners at future mass gatherings where heat-related illness is expected to have a major impact.

Key words: mass gatherings, emergency medical services

059 Impact of a bimodal SARS outbreak on paramedic resources of a large urban EMS system.

Verbeek PR, Silverman AC, McClelland IW, Burgess RJ. Department of Emergency Medicine, Sunnybrook & Women's College Health Sciences Centre, University of Toronto, Toronto, ON

INTRODUCTION: To describe the impact of a bimodal SARS outbreak on paramedic resources of a large urban EMS system. METHODS: During a bimodal SARS outbreak (SARS-1 and SARS-2) in a large North American city, a paramedic home quarantine (HQ) program was implemented. During the second phase of the outbreak, a supplementary "work quarantine" (WQ) program was also implemented to permit asymptomatic paramedics with a low-risk exposure paramedics to continue working under close medical surveillance. The number of paramedics on quarantine each day and the type of quarantine in place during each phase of the outbreak was determined. The mitigating effect of WQ in preserving the paramedic resource during the second phase of the SARS outbreak was determined. RESULTS: SARS-1 lasted 30 days (between March 17 and April 15, 2003). During SARS-1, there were 234 paramedics placed on HQ. The peak number of paramedics on quarantine during SARS-1 was 146 on day 12. The total number of HQ days was 1615. During the peak five days of SARS-1, the total number of HQ days was 664. SARS-2 lasted 18 days (between May 22 and June 8, 2003). During SARS-2, there were 292 paramedics placed on either HQ or WQ. The peak number of paramedics on quarantine was 236 on day 7 (78 paramedics on HQ and 158 paramedics on WQ). The combined number of quarantine days was 1637. During the peak five days of SARS-2, the combined number of quarantine days was 910. Of these, paramedics were available for work on 708 (78%) days due to the WQ program. CONCLUSIONS: A SARS outbreak can be responsible for large numbers of paramedics requiring quarantine within a short time frame. The use of WQ may be a useful way of mitigating the impact of a SARS outbreak on paramedic resources.

Key words: SARS, quarantine, emergency medical services

060 Prehospital transcutaneous cardiac pacing for symptomatic bradycardia or bradyasystolic cardiac arrest: a systematic review.

Sherbino J, Verbeek PR, MacDonald RD, Sawadsky BV, McDonald AC, Morrison LJ. Division of Emergency Medicine, Department of Medicine, University of Toronto, Toronto, ON

INTRODUCTION: Our objective was to determine the efficacy of prehospital transcutaneous cardiac pacing (TCP) in adult patients with symptomatic bradycardia (SB) or bradyasystolic cardiac arrest (BACA). METHODS: Medline, EM Base and Science Citation Index were searched from 1966 to 2003 using the terms: prehospital/emergency medical services, external/transcutaneous, and pacing. CIHR and NIH were contacted for unpublished trials. Two reviewer teams blinded to source and author conducted hierarchical selection and quality assessment (Jadad 5 point scale). Kappa agreement at each level was measured. All data points were abstracted by consensus. RESULTS: Thirty-one unique articles were identified and 7 selected (Kappa agreement; title:0.85, abstract:0.78, full article:0.82). Article quality was <3 in all trials. There were 3 case series (BACA, n = 215), 3 unblinded randomized control trials (1 BACA, 2 BACA+SB), and one subgroup (SB) analysis. Criteria for inclusion in BACA and SB were comparable. All BACA studies included asystole (6/6), post defibrillation asystole (4/6) and pulseless bradycardia (6/6). SB was defined as HR <60, BP <90 (1/3); and slow HR, GCS < = 12, decreased cap refill (2/3). All patients received standard ACLS care. In the case series of paced patients 0/215 survived to hospital discharge and 2/103 survived to admission. In the BACA controlled trials 56/509 (paced) vs 58/497 (control) survived to admission; and 16/509 vs 14/497 survived to discharge. In the SB subgroup of one trial, 6/6 (paced) vs 2/7 (control) survived to admission (p 0.01); and 5/6 vs 1/7 survived to discharge (p 0.01). The final study combined the SB subgroup of a second control trial with a case series (same TCP protocol), where 7/27(paced) vs 3/24 (control) survived to admission (p 0.20); and 4/27 vs 0/24 survived to discharge (p 0.07). CONCLUSIONS: In the prehospital setting, there is no evidence to support TCP use in BACA. A lack of evidence suggests further research to determine the efficacy of TCP in symptomatic bradycardia.

Key words: cardiac arrest, dysrhythmia, emergency medical services

061 Biomechanical evaluation of casualty evacuation by stretcher.

Fortier CJ, Ferreira JJ, Costigan PA, Deakin JM. Department of Family Medicine, McMaster University, Hamilton, ON

INTRODUCTION: Emergency rescue and military personnel are responsible for transporting injured people from sites of industrial and natural disasters and terrorist attacks to areas where further medical expertise is available. In Canada, Search and Rescue Technicians and other members of the Canadian Armed Forces are expected to perform this physically demanding task. The standardized simulation of stretcher carriage evaluated in this study is used to assess their occupational fitness. METHODS: We conducted a randomized controlled crossover study of the vertical and anterior-posterior (AP) forces during a two person stretcher carry and a simulation (front of the stretcher was lifted by a person, back of the stretcher supported on wheels). The forces were measured by strain gauges on aluminum stretcher handles and compared using paired t-tests. These tasks were performed by 11 participants on level ground and 7 participants on inclined terrain. The stretcher had two sets of 32-kilogram (kg) weights to mimic the load of a 70-kg anthropometric manikin on the front handles. RESULTS: On flat terrain, the mean vertical forces were greater in the simulation (236.6 ± 3.6 versus 221 ± 2.4 newton (N), p < 0.001) while the mean AP forces were less (5.9 ± 4.5 versus 15.8 ± 5.4 N, p < 0.003). During the ascent, the mean vertical forces in the simulation were decreased (252.7 ± 12.9 N versus 272.8 ± 10.3, p < 0.048) with increased mean AP forces (212.9 ± 8.0 versus 144.0 ± 8.2 N, p < 0.045). Since the rear stretcher bearer provided 32 % of the AP force on the uphill, the same 70 kg manikin could be simulated on both the uphill and level terrain with a 35 kg weight placed 0.27 metres from the front of the stretcher. CONCLUSIONS: This study quantified the forces in casualty evacuation by stretcher. Furthermore, we were able to demonstrate that the safety of this simulation could be improved by decreasing the overall weight on the stretcher while keeping the demands of the simulation similar to those of a two-person stretcher carry.

Key words: emergency medical services

062 Succinylcholine assisted intubation by general duty paramedics in the prehospital setting: comparison of success in a prospective cohort and an historical control.

Hall C, Fick G, Iwanow R, Doig C. Department of Emergency Medicine, Calgary Health Region, University of Calgary, Calgary, AB

INTRODUCTION: There is understandable reluctance to use a potentially dangerous pharmacological protocol for intubation in the prehospital setting where intubation skill can be variable and conditions unfavourable. In Canada, Rapid Sequence Intubation (RSI) to facilitate airway capture in the field has been limited to aeromedical flight crews. This study describes the success rate of prehospital endotracheal intubation (ETT) in a general duty urban advanced life support (ALS) Emergency Medical Services (EMS) system utilizing RSI with succinylcholine (sux) compared to intubation in the same system prior to the RSI protocol. METHODS: A prospective observational cohort (4 years of RSI use) was compared with a retrospective historical control (2 years prior to RSI) in the City of Calgary, population ~905,000. Proportion of successful prehospital intubation was compared in adult patients (pts) undergoing ETT attempts. Data was collected for all intubation attempts regardless of the use of RSI. RESULTS: 853 pts intubated during the RSI protocol, 520 (61%) were in cardiac arrest and did not require RSI. During RSI years, in non-arrested pts, 226/333 (39% total intubations) did not receive sux for ETT attempts at paramedics' discretion. 107 (12.5% total intubations) non-arrested pts were intubated using RSI with sux. Intubation was successful in 83.2% (89/107) non-arrested pts intubated with sux vs. 52% (52/100) non-arrested pts in the historical control, p <0.0005. Odds ratios (OR) for intubation success with sux was 2.41 (95% CI 1.05,5.51); OR for ETT success with midrange GCS and sux was 0.73 (95% CI .31,1.73); OR for ETT success with trauma mechanism and sux was 3.63 (95% CI 1.71,7.72). Cric rate for failed ETT attempts changed by 4.3% (-10.7, 2.2) from 8% (8/100) in control to 3.7% (34/107) in RSI years, p 0.24. CONCLUSIONS: Cinylcholine administration via RSI protocol in selected pts is a significant predictor of intubation success. Influence of RSI on cric rate requires further study.

Key words: Rapid Sequence Intubation, airway management, emergency medical services

063 Use of special events medical directives for paramedics at a mass gathering.

Feldman MJ, Lukins JL, Schwartz B, Burgess RJ. Department of Emergency Medicine, Sunnybrook & Women's College Health Sciences Centre, University of Toronto, Toronto, ON

INTRODUCTION: Medical care at mass gatherings presents unique challenges. Many patients do not require transport to medical facilities. Paramedics provide the majority of care at many mass gatherings, but do not typically possess the authority to treat and release patients. We describe medical directives to allow paramedics to treat and release patients with minor injuries and illnesses at a mass gathering. METHODS: Our base hospital provides medical oversight to paramedics in a large urban EMS system. Medical directives allowed paramedics to provide four medications (acetaminophen, dimenhydrinate, diphenhydramine, and Polysporin®) for common complaints, and oral fluids for uncomplicated heat-related illness. Paramedics were instructed to assess vital signs, allergies, contraindications to treatment, and signs or symptoms of serious underlying illness. After treatment, patients could be released or transferred to a site medical facility. These directives were implemented at the Toronto Rolling Stones concert in July 2003. Patient records were collected and demographics, chief complaint, time of incident, treatment, and disposition were obtained. RESULTS: Over 450,000 people attended the concert, with 1870 presenting for medical care. A significant proportion required water, sunscreen, or bandages and records were not taken of these encounters. Records were obtained for 1205 patients of whom 407 received medications under the directives. The disposition was not recorded in 13 cases. 299 patients were treated with acetaminophen, of whom 269 (90.0%) were released and 23 (7.7%) required additional care. 62 patients received dimenhydrinate, of which 44 (71%) were released and 14 (23%) required transport. 36 patients received diphenhydramine, of whom 34 (94%) were released. Ten patients received Polysporin® for minor wounds. CONCLUSIONS: A treat and release medical directive for paramedics providing care at mass gatherings allows the release of selected patients and may divert patients from requiring care at a medical facility.

Key words: mass gathering, emergency medical services

064 How widely used is the Canadian c-spine rule by emergency physicians?

Brehaut J, Stiell IG, Visentin L, Graham I, Grimshaw G. Department of Emergency Medicine, University of Ottawa, Ottawa, ON

INTRODUCTION: The Canadian C-Spine Rule (CCR) was recently derived (2001) and validated (2003) and shown to be highly sensitive for c-spine injury. Little, however, is known about the uptake of the CCR by clinicians. We sought to determine the actual use of the CCR by emergency physicians (EPs) as well as factors associated with its use. METHODS: We conducted a national mail survey of a random sample of 399 practising EPs, using a modified version of Dillman's Total Design Method whereby 1 pre-notification letter and 3 sequential survey mailings and reminders were sent. The survey included 24 questions related to use of the CCR, factors of use, clinical decision rules in general, and professional status. We conducted univariate chi-square and t-test analyses. RESULTS: Of the 261 EPs who responded (69.4% of those contacted) characteristics were: mean age 41.0, male 75.9%, EM certified 80.4%, fulltime 72.8%, teaching hospital 66.5%. 82.8% said they knew of the CCR and, of those, 75.5% said they currently use it (62.5% of all respondents). 91.4% of physicians who use the CCR do so "always" or "most of the time". The most common mode of learning about the rule was through journal articles 37.0% or CME 31.9%. Respondents' attitudes towards the CCR are indicated by strong or moderate agreement on a 6-point scale: Useful in my practice 87.4%, Efficient use of time 82.5%, Easy to use 76.1%, Easy to learn 74.6%, Easy to remember 60.4%, Too much trouble 5.8%, Too unsafe 2.8%, Increase chance of lawsuits 2.4%. In general, EPs rate the 4 most important factors in their adoption of any new decision rule to be (mean score on 9-point scale): Easy to use 1.9, Easy to remember 2.5, Easy to incorporate into practice 2.9, Saves time in the ED 3.0. CONCLUSIONS: A surprising number of emergency physicians are aware of the CCR, are currently using it, and use it most of the time. EPs value ease of use and ease of incorporation for decision rules in general and have very positive attitudes towards the CCR.

Key words: Canadian C-Spine Rule, clinical prediction rule

065 Attitudes of emergency physicians toward the Ottawa ankle rules.

Brehaut J, Stiell IG, Visentin L, Graham I, Grimshaw J. Department of Emergency Medicine, University of Ottawa, Ottawa, ON

INTRODUCTION: The Ottawa Ankle Rules (OAR) have been shown to be highly accurate in excluding ankle fractures in a meta-analysis of 32 studies from 12 countries and involving 15,581 patients (Bachmann BMJ 2003). Little, however, is known about how the OAR are used by clinicians. We conducted a survey of emergency physicians (EPs) to address this issue. METHODS: This national mail survey included a random sample of 399 EPs and used a modified version of Dillman's Total Design Method (1 pre-notification letter and 3 sequential survey mailings). The 24 survey questions related to use of the OAR, decision-making factors, attitudes, and professional status. We conducted univariate chi-square and t-test analyses. RESULTS: 261 EPs responded (69.4% of those contacted) and had these characteristics: male 75.9%, mean age 41.0, years since graduation 14.6, EM certified 80.4%, ED census >50,000 46.6%. 89.6% used the OAR "always" or "most of the time", had been doing so for 5.3 years, and learnt about the OAR from: journal articles 34.1%, medical school 31.0%, CME 20.6%. EPs based their ankle radiography decisions on: primarily OAR 42.0%, OAR plus small number of factors 42.0%, OAR and many factors 15.2%. These additional social and clinical factors included: gross deformity 96.1%, communication problem 88.5%, distracting pain 76.5%, fall from height 71.2%, hostile patient 66.5%, patient request 37.7%. Respondents' attitudes towards the OAR are indicated by strong or moderate agreement on a 6-point scale: Easy to learn 95.7%, Easy to use 94.6%, Useful in my practice 92.6%, Efficient use of time 90.3%, Easy to remember 88.7%, Increase chance of lawsuits 2.0%, Too much trouble 1.2%, Too unsafe 0.4%. CONCLUSIONS: A large number of practicing EPs use the OAR and 30% learned about them in medical school. EPs often incorporate other social and clinical factors into their ankle radiography decision-making, which may reduce rule specificity, but, overall, have very positive attitudes towards the OAR.

Key words: clinical prediction rule, Ottawa ankle rule

066 How important is the age 65 criterion in the Canadian C-Spine Rule?

Stiell IG, Brison R, McKnight RD, Clement C, Rowe B, Wells GA, Lee J, Dreyer J, MacPhail I, Holroyd B, Schull M, Eisenhauer M, for the CCC Study Group. Department of Emergency Medicine, University of Ottawa, Ottawa, ON

INTRODUCTION: The Canadian C-Spine Rule (CCR) consists of 3 high-risk, 5 low-risk, and range of motion criteria, is highly sensitive for c-spine injury, and requires all patients aged 65 or older to undergo radiography. In this study, we sought to evaluate the performance of the CCR and its components in these older patients. METHODS: This prospective cohort study was conducted in 9 tertiary care EDs and involved alert (GCS 15) and stable adult trauma patients at risk for neck injury. Physicians performed standardized assessments and completed data forms for patients who then underwent radiography to determine the outcome, clinically important c-spine injury. We conducted recursive-partitioning analyses and calculated unadjusted odds ratios. RESULTS: Among the 8,283 patients, 622 (7.5%) were aged 65 or older. The older patients had 41 (24.3%) of the 169 injuries, yielding an odds ratio of 4.4 for the criterion "Age 65". Comparing the older to the 7,661 younger patients, the odds ratios for injury for the CCR component criteria are illustrated in the Table (below). Applying the complete CCR to all patients, the sensitivity was 99.4% (95% CI 96-100%), whereas for the older patients it was 92.1% (78-97%) and would have missed 3 injuries. CONCLUSIONS: In older patients, several CCR criteria, particularly "Dangerous Mechanism", perform less well and the overall sensitivity of the CCR is insufficient. The "Age 65" criterion remains an important component of the CCR and all potential neck injury patients aged 65 and older should undergo imaging.

Key words: Canadian C-Spine Rule, clinical prediction rule

067 Predicting admission in the absence of pneumonia severity index score.

Blitz S, Rowe BH, Marrie TJ. Division of Emergency Medicine, University of Alberta, Edmonton, AB

INTRODUCTION: The pneumonia severity index (PSI) class is sometimes used as an indicator of admission for patients presenting to the Emergency Department (ED) with community acquired pneumonia (CAP). Frequently, a PSI class can not be calculated due to insufficient data and this study examines an alternative approach. METHODS: Between 11/2000 and 11/2002, all CAP patients presenting to 6 regional EDs were assessed according to a specified treatment pathway. ED physicians were encouraged to compute a PSI score on those patients who may have required admission. This analysis considers the patients in whom a PSI score was not determined. Logistic regression was used to determine predictors of admission. Sensitivity and specificity was used to determine cutoff values for continuous variables. Odds ratios (OR) and 95% confidence intervals (CI) are reported. RESULTS: 1362 of 3818 entered on the pathway did not have a PSI. 774 patients with complete data on the variables included in the model were included in the final analysis. Oxygen saturation <92 (OR: 11.6; 95% CI; 7.6, 17.7), age>68 (OR: 4.5; 95% CI: 2.9, 7.0), altered mental status (OR: 4.3; 95% CI: 1.7, 11.0), triage score<4 on a scale of 1 to 5 (OR: 3.1; 95% CI: 1.8, 5.5), pulse>94 (OR: 2.0; 95% CI: 1.2, 3.1) and respiratory rate >25 (OR: 1.7; 95% CI: 1.1, 2.6) were significant predictors of admission. Results were transformed to scores and summed to yield an admission risk score, which ranged from 0 to 24. Risk scores were classified into low (<5), moderate (5-8) and high (>8) probability of admission, based on ROC cut-points. Admissions occurred in 2%, 13%, and 60% (p = 0.001), respectively; death occurred in 0%, 1.4%, and 7.5% (p = 0.001), respectively. CONCLUSIONS: An admission risk score based on common variables, which are easily collected in the ED, can successfully predict the probability of admission for CAP patients. The same score was predictive of death; however, not of relapsing back to the ED (data not shown). This model requires prospective validation prior to implementation.

Key words: pneumonia severity index, pneumonia

068 Investigating the impact of lowering the injury severity score cutoff for major trauma in pediatrics.

Kaida A, Petruk J, Sevcik W, Latoszek K, Ohinmaa A, Jacobs P, Fraser-Lee N, Blitz S, Rowe BH. Population Health and Research, Capital Health, Edmonton, AB

INTRODUCTION: Most trauma registries use injury severity scores (ISS) > 11 to designate major trauma. While this may be reasonable for injured adults, many children with ISS < = 11 require hospitalization and consume health care resources. We investigated the patient characteristics and outcomes of children with ISS 9-11 requiring hospitalization and compared them to those in the pediatric trauma registry (ISS > 11). METHODS: All pediatric patients (under 17 years old) admitted for trauma with an ISS of at least 9 at all 6 hospitals within a linked regional health authority from 04/02 to 03/03 were included. Data for ISS> 11 were obtained from the trauma registry; data for ISS 9-11 were extracted from chart review using the same trauma personnel. RESULTS: 483 patients were admitted during the study period: 360 (75%) had ISS 9-11, and 123 (25%) had ISS>11. Most injuries (54%) were the result of falls, followed by transportation injuries (28%). Median age of patients was 9 years and 62% were male. Patients with ISS > 11 were older (median age = 12 vs. 8, p = 0.001). Patients were more likely to have stayed in the ICU (37% vs. 2%) or died (13% vs. 0%) in the ISS > 11 group (both p < 0.001). Patients more frequently stayed > 1 day in hospital (86% vs. 42%; p<0.001) in the ISS > 11 group; total hospitals days were high in both groups (ISS 9-11 = 876; ISS > 11 = 1229). Patients with ISS 9-11 were most likely to be admitted to orthopedics (88%), whereas patients with ISS > 11 were more likely to be admitted to neurosurgery (37%) or intensive care (28%). CONCLUSIONS: Using a cut-point of ISS > 11 to designate major trauma in children severely under-estimates the frequency of trauma requiring hospitalization in this community. While patients with ISS 9-11 often result from falls that require short stays and all survive, they contribute significantly to the economic burden of injury. Thought should be given to changing the definition of major trauma in children to include these patients.

Key words: injury severity score, trauma

069 A population-based snapshot of pediatric neurotrauma.

Grant VJ, Macpherson A, MacKay M, Lipskie T, Mongeon S, Osmond MH. Division of Emergency Medicine, Department of Pediatrics, Children's Hospital of Eastern Ontario, Ottawa, ON

INTRODUCTION: Pediatric neurotrauma accounts for significant morbidity and mortality in childhood. Community injury prevention activities are best guided by local data. Our objective is to describe epidemiological features of a population-based sample of pediatric neurotrauma. METHODS: The Canadian Hospital Injury Reporting and Prevention Program (CHIRPP) is a national surveillance program that collects injury data at 10 pediatric and 5 general EDs. To obtain population-based data, CHIRPP was expanded to prospectively collect injury data on children 0-19 years from all the EDs and Urgent Care Centres in Ottawa from March 2002 to February 2003. We performed descriptive statistics with 95% CIs. RESULTS: 30,151 patients were enrolled over one year. 3112 (10%) of injuries were coded as neurotrauma (injury to the head and/or spine). The most common activity associated with all neurotrauma was sports (24%). Of all neurotrauma, 64% were minor head injuries, 18% were concussions and 2% were intracranial injuries. The Table (below) shows activities associated with all neurotrauma injuries and intracranial injuries specifically. The 15-19 year age group represented a higher proportion of intracranial injuries compared with all neurotrauma (37% vs 22%, p = 0.01). There was no specific association seen between all neurotrauma and season of year or day of week. However, when looking specifically at intracranial injuries, there was an association seen with the summer months (p = 0.01). CONCLUSIONS: Intracranial injuries accounted for only 2% of all neurotrauma, and were more commonly seen in the summer months, associated with vehicular use, and most often seen in youth aged 15-19 years. Road and vehicle operator safety should be a high priority in this community's injury prevention strategy.

Key words: pediatric, injury surveillance, head injury

070 Do cases of pediatric neurotrauma in an urban setting preferentially present to a pediatric tertiary care centre emergency department (PED)?

Grant VJ, Macpherson A, MacKay M, Lipskie T, Mongeon S, Osmond MH. Division of Emergency Medicine, Department of Pediatrics, Children's Hospital of Eastern Ontario, Ottawa, ON

INTRODUCTION: Pediatric neurotrauma accounts for significant morbidity and mortality in childhood. Early intervention at a pediatric trauma center may improve outcomes. Our aim is to determine whether cases of pediatric neurotrauma in an urban setting preferentially present to a pediatric tertiary care ED (PED). METHODS: The Canadian Hospital Injury Reporting and Prevention Program (CHIRPP) is a national surveillance program that collects injury data at 10 pediatric and 5 general Emergency Departments (ED). Though the data is detailed it is limited by not being population-based. We performed a prospective cohort study where CHIRPP data were collected over a one-year period (March 2002-February 2003) on all children 0-19 years presenting with acute neurotrauma to all of the EDs and Urgent Care Centres (UCCs) in a large Canadian city (750,000 pop.). We performed descriptive statistics with 95% CIs. RESULTS: During the study period, 30,151 patients were enrolled in the citywide CHIRPP expansion project. Of these, 3112 (10%) were coded to have had neurotrauma (injury to the head and/or spine). 66% of patients with neurotrauma were seen at the PED, while 24% were seen in general EDs and 10% were seen at UCCs. The percentage of patients seen for neurotrauma at the PED was significantly higher than for patients seen with all other injury types (66% vs 44%, p < 0.0001). Of the 3112 patients seen for neurotrauma, 46 were deemed to have significant neurotrauma coded by the presence of an intracranial injury. Of these 46 patients, 67% (p = 0.75) presented initially to the PED. CONCLUSIONS: Compared to all injuries, the presence of neurotrauma is associated with an increased likelihood of initial presentation to a PED. However, site of initial presentation was not further influenced by the severity of neurotrauma. Future studies should investigate treatment and outcomes of these two populations.

Key words: pediatric, injury surveillance, head injury

071 Focussed abdominal ultrasound for blunt trauma in an emergency department without advanced imaging or on-site surgical capability.

Shuster M, Abu-Laban RB, Boyd J, Gauthier C, Mergler S, Shepherd L, Turner C. Mineral Springs Hospital, Banff, AB

INTRODUCTION: Focused abdominal ultrasound for trauma (FAST) is well established in urban emergency departments (EDs). The use of FAST in EDs without advanced imaging or on-site surgical capability has not been studied. The objective of this study was to evaluate whether FAST, when performed by rural emergency physicians (EPs), prompts the immediate transfer of patients who would otherwise have been observed or discharged, and who are later proven to have free intraperitoneal fluid. METHODS: Participating EPs undertook a minimum of 30 hours of ultrasound training in a trauma center under the supervision of certified radiologists. All patients presenting to our rural ED with blunt abdominal trauma between 9/1/2002 and 4/30/2003 were eligible. Following a history and physical examination, the EP documented his/her disposition decision. A FAST was then performed and the disposition reconsidered in light of the FAST results. RESULTS: Sixty-seven FAST exams were performed on 65 patients. Three FAST exams were positive as was subsequent definitive imaging (4.5%: 95% CI 0.9% to 12.5%), while 4 FAST examinations were negative and subsequent definitive imaging (delayed a mean/median of 5.8/9.3 hours) was positive (6.0%: 95% CI 1.7% to 14.6%). One of 38 patients with a negative FAST and no definitive imaging on initial encounter returned 24 hours later because of worsening symptoms (2.6%: 95% CI 0.1% to 13.8%). FAST results did not alter the decision to transfer any patient (0%: 95% CI 0.0% to 5.4%), although one positive FAST may have led to an expedited transfer. CONCLUSIONS: This study failed to demonstrate that FAST alters the disposition of blunt abdominal trauma patients in a hospital without advanced imaging or on-site surgical capability. However, since the 95% confidence intervals indicate that up to 5.4% of transfer decisions could be influenced by FAST, our results are best considered inconclusive and further research in this area is warranted.

Key words: ultrasound, trauma

072 Pediatric all-terrain vehicle injury patterns: a five-year review in southwestern Ontario.

Alawi K, Lynch T, Lim R. Department of Pediatrics, University of Western Ontario, London, ON

INTRODUCTION: All-terrain vehicles (ATV) are a known, significant risk factor for major pediatric trauma. Their use continues by children under sixteen years of age despite position statements by the American Academy of Pediatrics outlining their inherent dangers in this age group. The objective of this study is to characterize the nature of the injuries sustained by pediatric patients involved in ATV accidents in Southwestern Ontario. METHODS: This study is a retrospective chart review of pediatric patients (less than 18 years of age) involved in an ATV accident with an Injury Severity Score (ISS) > 12 who presented to Children's Hospital of Western Ontario between September 1, 1998 and September 30, 2003. Patient medical records were reviewed and injuries were recorded. RESULTS: Seventeen patients between the ages of 8 and 17 were reviewed. The average age was 13.7 years while 76.5 % were less than 16 years of age. Eighty-two percent of the patients were males. The average ISS and length of stay were 22.8 and 9.7 days, respectively. There were no mortalities. Overall, 4 patients required an operative procedure. Seven of the patients (41.2%) sustained splenic injuries which was the most common injury and two of these patients required splenectomy. Liver injury was present in 6 (35.3%) patients. Four of the 5 patients with head injuries suffered isolated head injuries. Helmets were worn by 64.7 % of the patients. Three of the 5 head-injured patients were not wearing a helmet. Two of the five patients with orthopedic injuries required an operative procedure. Renal injuries were noted in 17.6 % of patients. One patient sustained a debilitating brachial plexus injury. CONCLUSIONS: A variety of injury patterns have been described with splenic injury being the most common. ATV trauma continues to pose a significant threat to the pediatric population in Southwestern Ontario.

Key words: All-terrain vehicles, trauma, pediatric

073 Hydro-Québec algorithm: why patients are monitored for 24 hours after an electric shock?

Bailey B, Gaudreault P. Division of Emergency Medicine, Department of Pediatrics, Hôpital Ste-Justine, Montreal, PQ

INTRODUCTION: At the beginning of the 1990s, Hydro-Québec distributed an algorithm to all Quebec EDs to determine the need for cardiac monitoring following an electric shock. Since then, most Quebec EDs use the algorithm. The potential risk factors that justify monitoring patients according to this algorithm are not currently known. METHODS: All patients presenting to one of the 21 participating Quebec EDs following an electric shock with either current that passed through the heart (suggested by sensation or burn marks), tetany > 1 second, lost of consciousness, or voltage source >1000 volts were approached to participate in an observational study of the use of the Hydro-Québec algorithm. ED physicians had to obtain consent and fill in a standardized form. RESULTS: Since October 2000, 122 patients were recruited: 24 were children and 98 were adults including 77 work-related accidents. A total of 58 were exposed to 120-240 volts, 41 to 347-600 volts, 12 to 4000, 14400-25000 or 52000 volts, 1 to 750 volts DC while the voltage was unknown in 10 cases. The reasons for monitoring patients according to the algorithm were: 77 because the current passed through the heart (69 suggested by sensation and 8 by burn marks), 15 because of tetany > 1 second, 3 because of lost of consciousness, and 8 because of a voltage source of more than 1000 volts. A total of 109 patient had a normal initial ECG while 13 patients had an abnormal initial ECG. Cardiac monitoring was normal in all but 2 patients: one patient developed a short PR and another an asymptomatic bradycardia. These two patients had transthoracic current with domestic voltage. CONCLUSIONS: Most patients monitored according to the Hydro-Quebec algorithm had transthoracic current as a potential risk factor. If the number of patients monitored were to be reduced significantly, transthoracic current would have to be eliminated as a potential risk factor for arrhythmia following an electric shock. Therefore, more patients will need to be studied in order to do that.

Key words: electric shock

074 Motor vehicle collision characteristics associated with brain injury in mild head injury patients.

Al-Salamah M, Stiell IG, Clement C, Wells GA, for the CCC Study Group. Department of Emergency Medicine, University of Ottawa, Ottawa, ON

INTRODUCTION: To identify high-risk mechanisms of injury due to motor vehicle collisions (MVC) that are associated with clinically important brain injury in mild head injury patients. METHODS: This study consisted of the derivation and validation phases of the Canadian CT Head rule study. This prospective cohort study involved adults with loss of consciousness, amnesia, or confusion and a GCS score of 13-15. The outcome criterion was clinically important brain injury on CT scan. Study nurses reviewed ambulance reports, ED records, and in-hospital records to classify each case according to 8 dangerous MVC factors. A logistic regression model was built by manual stepwise addition of preinjury characteristics, adding interactions and adjusting for clustering by hospital using multilevel modeling. RESULTS: Among the 5858 patients, there were 1477 MVC patients with the following characteristics: 67 (4.5%) had important brain injury, 13 (0.9%) required urgent neurological intervention, 62% male, 9.7% intoxicated, median age 30 years, 57.5% seat-belted, 25% highway speed (60-100km/hr), 20% high speed (>100km/hr), 18% rollover, 9% head-on collision, 8.7% bulls-eye damage to windshield, 6.6% ejected, 2.3% death in same MVC. The odds ratios with 95% CIs for important brain injury for each mechanism were: High Speed 6.4 (2.1, 19.3), Ejected 3.8 (1.6, 8.9), Highway Speed 3.4 (1.2, 9.8), Rollover 2.9 (1.1, 7.6), Death in the same MVC 1.4 (0.3, 7.7), Head-on Collision 1.2 (0.5, 3.1), Bulls-eye damage to windshield 1.2 (0.4, 3.4), Seat-belted 0.5 (0.2, 1.0). Other important pre-injury variables were: Age (10 year increments) 1.9 (1.4, 2.5), Female 0.4 (0.2, 1.0), Suspected Intoxication 1.9 (0.8, 4.7), Winter 4.0 (1.5, 10.8), Spring 1.8 (0.7, 4.3) and Summer 4.4 (1.6, 12.1). CONCLUSIONS: Certain injury mechanisms place patients at increased risk of brain injury and health care providers should carefully ascertain details of the injury when managing mild head injury patients.

Key words: injury surveillance, minor head injury

075 Follow-up after a paediatric emergency department visit "telephone versus electronic-mail".

Goldman RD, Mehrotra S, Pinto TR, Mounstephen W. Division of Paediatric Emergency Medicine, The Hospital for Sick Children, Department of Paediatrics, University of Toronto, Toronto, ON

INTRODUCTION: The Internet has become an unlimited source of health-related information and health information access by parents. Follow-up after an Emergency Department (ED) visit is important for continuity of care, monitoring changes in patient health and for informing patients of test results taken during their visit. We conducted this study to determine if electronic main (e-mail) would be better than phone for follow-up. METHODS: During two months, research assistants recruited parents of children being discharged from a large tertiary ED in Toronto, Ontario, that have a personal e-mail account and a personal telephone line. Parents were randomized to e-mail or phone follow-up, that included two questions. Parents not replying to the e-mail were contacted via telephone after 10 days. The main outcome measure was the response rate. Data was collected on Microsoft Excel and statistical analyses used SPSS for calculation of response rate and comparisons between the phone and e-mail groups. RESULTS: A total of 337 families were approached. Two hundred and sixty five (79%) had Internet access, personal e-mail and telephone line available and 75% of them check e-mails at least once a day. A total of 198 families were randomized for a follow-up. Of 98 families, 85 (87%) were reached by phone within average of 17 hours (range 0-98). Of 100 families contacted via e-mail only 53 (53%) responded in a mean of 46 hours (range 0-242) (p < 0.0005). Ten (10%) of the e-mails "bounce back". Of the parents in the e-mail group that had received the e-mail and did not respond 21 (57%) did not check the e-mail, did not remember reading the e-mail or had problems with e-mail access, 1 (3%) did not have time to respond and 15 (40%) were not reached even by phone. CONCLUSIONS: E-mail failed to be better than phone in reaching families for follow-up after a Paediatric ED visit. It could be used for non-urgent messages by about half of our patient population that can access the Internet.