CAEP/ACMU 2001 Scientific Abstracts - Oral Presentations: 30-44

CAEP Abstracts

CJEM 2001;3(2):121-134

| 001-014 | 015-029 | 030-044 | 045-059 | 060-074 | 075-093 |

030 Survey of computer use in emergency medicine staff and residents.
Cummings G, Sher A, Hayward R, Rowe BH. University of Alberta, Edmonton, Alberta.

OBJECTIVES: Easily accessible computer resources made available in emergency medicine (EM) may improve the use of valid health information in both the clinical and non-clinical setting. This study surveys the EM staff after access to a computer-based resource of evidence-based health information. METHODS: CHE/CLINT is a single sign-on, password-protected, desktop application for medical information resource management. An EM-specific desktop was produced with journals, online resources, and important links and access was provided to 70 staff and resident EM users at 2 major teaching hospitals affiliated with the University of Alberta. Following instruction and 3-6 months of use, a survey was distributed via email to all users. Up to 3 email send outs were performed. The 22-question survey examined user computer background, use of CHE/CLINT, and resource needs. Staff were divided into users (>2 log-ins) and non-users (<2 log-ins). RESULTS: Overall, 58 (80%) of staff returned completed questionnaires; 56 (76%) eligible users had logged into this program. Physicians in the non-user vs. user groups perceive the system to be an equally valuable resource (4.8 vs. 4.2 out of 5; p > 0.1). Non-users were more often Macintosh users (37% vs. 7%, p < 0.01) and less comfortable using the CHE/CLINT application (2.5 vs. 4.4 out of 5; p < 0.01). Nonusers did not differ from users in comfort with computers, internet access, or other demographic factors (all p > 0.1). However, users more frequently select electronic sources first when searching for solutions to information needs (43% vs. 0%, p < 0.01). CONCLUSIONS: While this knowledge-based single sign-on system has been explored by many users, it has not yet gained widespread acceptance by EM physicians. However, even non-users remain active in other methods of information retrieval. Further research is required to determine how to encourage EM staff to more effectively acquire information relevant to their practices.
Key words: medical informatics

031 Ambulance diversion increases transport delays for patients with chest pain.
Schull MJ, Morrison LJ, Vermeulen M, Redelmeier DA. Department of Emergency Services, and Clinical Epidemiology Unit, Sunnybrook and Women's College Health Sciences Centre, Toronto, Ontario.

OBJECTIVES: Emergency department overcrowding and ambulance diversion is increasing across North America. We sought to determine the impact on ambulance transport of cardiac patients. METHODS: Data on consecutive chest pain patients transported to hospital by ambulance in Toronto, Canada, were obtained for 2 time periods: February to May 1997 (n = 1563) vs. February to May 1999 (n = 1782). These 2 periods were chosen because they reflected times of low and high emergency department (ED) overcrowding respectively. Quantile regression of 90th percentile intervals was utilized for all analyses. Multivariate analyses modeled transport intervals according to period, weekday, time of day, geographic location of patient, dispatch priority, case severity, return priority, and the number of other chest pain patients transported within two hours of the index transport. RESULTS: There were no significant differences in patient characteristics according to time period (i.e., weekday, time of day, dispatch priority, case severity, return priority, or geographic location), despite a 14% increase in the number of patients transported in 1999 (p < 0.0001). The 90th percentile System Response Interval increased 10% from 1997 to 1999 (9.7 min [95% confidence interval (CI) 9.5-10.0] vs. 10.7 [95% CI 10.4-10.9] respectively), while the On-Scene Interval decreased 8% (28.0 min [95% CI 27.3-28.6] vs. 25.8 [95% CI 25.2-26.3] respectively). The largest delay was in Transport Interval, which increased by 28% from 1997 to 1999 (13.4 min [95% CI 12.8-13.9] vs. 17.2 [95% CI 15.9-18.2] respectively). The Total Prehospital Interval did not change significantly. In multivariate analyses, time period remained a significant predictor of delayed Transport Interval (p < 0.0001). CONCLUSIONS: There has been a 28% delay in the ambulance Transport Interval for chest pain patients associated with increased ambulance diversions. On an ecologic level this may result in some patients with acute coronary syndromes dying due to delays to definitive treatment.
Key words: emergency overcrowding, emergency medical services

032 Use of a structured care map in the emergency department to improve asthma care.
Chahal A, Rowe BH, Spooner CH, Wilson D, Senthilselvan A. University of Alberta, Edmonton, Alberta.

OBJECTIVES: This study examines the implementation of a structured asthma care map (CM) to standardize the emergency department (ED) treatment of acute asthma. METHODS: A retrospective before-after review of hospital charts was conducted on 2 15-month periods, before (PRE) and after (POST) the implementation of an ED CM. All patients between 18 and 65 presenting to the ED with a primary diagnosis of acute asthma were eligible. A standardized form was used to abstract data from a random sample of charts from each period. A priori, criteria were established to determine the level of completion and success of the CM. RESULTS: Three hundred patients' charts (150 PRE, 150 POST) were selected. Patient characteristics were similar at the outset; however, patients in the POST period had lower mean peak expiratory flow (PEF). The gender ratio was 1:1; the mean age was 32 years. In the POST period, the care map was used in 100 (67%) cases, documentation of markers of severity increased, use of systemic corticosteroids occurred earlier (p = 0.02) and more often (68% vs. 57%, p = 0.04), and prescription of inhaled corticosteroids increased (57% vs. 41%, p = 0.006); but PEF recordings were uncharged (87%, p = 0.9). Mean length of stay in the ED was longer in the POST period (345 vs. 266 minutes, p = 0.01). Admission rates during both periods were low (PRE: 13% vs. POST: 20%; p = 0.28) and documented use of oral steroids on discharge remained the same (57% vs. 66%, p = 0.12). CONCLUSIONS: A standardized ED asthma CM was widely accepted, improved chart documentation, improved some aspects of in-ED care, and increased the prescription of preventive medications on discharge. However, care is often still delayed and research is required to evaluate other strategies to improve acute asthma care.
Key words: asthma, quality

033 Outpatient treatment of deep vein thrombosis from the emergency department: clinical outcomes, patient satisfaction and economic analysis.
Zed PJ, Cook TR, Filiatrault L, Busser JR. CSU Pharmaceutical Sciences and Department of Emergency Medicine, Vancouver Hospital and Health Sciences Centre, Vancouver, British Columbia.

OBJECTIVES: Low-molecular-weight heparins (LMWH) have been shown to be safe, effective and economically attractive for the outpatient treatment of deep vein thrombosis (DVT). The Vancouver Hospital and Health Sciences Centre (VHHSC) outpatient DVT program was implemented in June 1999 to take advantage of this safe, effective and economically attractive therapy. To determine if the VHHSC outpatient DVT program is safe and effective and to assess the economic impact compared to inpatient management. Also, to evaluate patient satisfaction with the outpatient program. METHODS: Patients were enrolled in the outpatient program after a radiographically diagnosed DVT by an emergency physician. Eligible patients returned daily to hospital to receive tinzaparin 175 u/kg subcutaneously daily for a minimum of 5 days and warfarin until their international normalized ratio (INR) >2 on 2 consecutive days. Incidence of recurrent venous thromboembolic events (RVTE) at 3 and 6 months, bleeding complications and thrombocytopenia were evaluated. Economic assessment was determined by comparing total costs of therapy for each outpatient to data collected retrospectively from inpatients treated with unfractionated heparin and warfarin. Patient satisfaction in the outpatient DVT program was established by a patient satisfaction survey. RESULTS: Fifty-three patients were treated in the program for a mean length of treatment of 5.7 days. No patients developed a RVTE at 3 months (n = 39), while only one patient of the 30 evaluable at 6 months experienced a recurrence (3.3%) post discharge from the program. No patient experienced a major bleeding complication or developed thrombocytopenia while 2 patients (3.8%) experienced a minor bleed, both had INRs within normal therapeutic range (INR = 2-3). Average cost/treatment/inpatient was $2,214.99 compared to $556.27 for outpatient therapy, representing a cost savings of $1,658.72 for every DVT treated using the DVT outpatient program. These 53 patients represent 302 admission days avoided and a $87,912.16 cost avoidance as a result of outpatient management. Overall, 93% of patients were very satisfied or satisfied with the outpatient program. CONCLUSIONS: The ED-based outpatient DVT program is safe, effective and economically attractive in addition to providing high patient satisfaction.
Key words: deep vein thrombosis, low molecular weight heparins

034 An analysis of stroke patient eligibility for tissue plasminogen activator in a Canadian city.
Barber PA, Demchuk A, Hill M, Buchan A. Departmenrt of Clinical Neuroscience, University of Calgary, Calgary, Alberta.

OBJECTIVES: Alteplase (tissue plasminogen activator [tPA]) is an effective treatment of acute ischemic stroke if used within 3 hours. However, thrombolytic therapy in hyperacute stroke will not have a major impact on death and dependency unless it is accessible to more patients. We identified the reasons why most patients with ischemic stroke did not receive intravenous alteplase and assessed the community impact in a large North American city. METHODS: Consecutive patients with acute ischemic stroke were identified in a prospective registry of stroke patients at a university teaching hospital between October 1996 and December 1999. Additional patients with ischemic stroke were identified that were admitted to one of three other Calgary hospitals during the study period. The Oxford Community Stroke Programme Classification was used to record type and side of stroke. The primary aim of this study was to determine the reason patients with ischemic stroke did not receive intravenous alteplase therapy. RESULTS: Of 2,165 stroke patients presenting to the university hospital 1,168 (53.9%) were diagnosed with ischemic stroke, 31.8% with intracranial hemorrhage (intracerebral, subarachnoid or subdural) and 13.9 % with transient ischemic attack. 84/314 (26.7%) patients admitted within 3 hours of stroke symptom onset received intravenous alteplase. Among patients arriving within 3 hours of onset, the major reasons for exclusion were mild stroke (13.1%), clinical improvement (18.2%) and perceived protocol exclusions (12%). Of those patients whose strokes were considered too mild or were documented to have had significant improvement, 32% either remained dependent at hospital discharge or died during hospital admission. Delay in presentation to emergency department beyond 3 hours excluded 854/1168 (73.1%). A total 1,806 ischemic stroke patients were admitted to Calgary hospitals during the study period, of which 4.7% received intravenous alteplase. CONCLUSIONS: The effectiveness of thrombolysis is dependent on understanding why patients are excluded from its use. Efforts need to be made to improve both the eligibility of patients for alteplase therapy and the appropriate interpretation of national guidelines through health education programs, and by developing organized acute stroke care infrastructure within communities.
Key words: stroke, tissue plasminogen activator, thrombolysis

035 DC cardioversion in atrial fibrillation: a survey of practice by Canadian emergency physicians.
Borgundvaag B, Ovens H. Mount Sinai Hospital, Toronto, Ontario.

OBJECTIVES: Atrial fibrillation (AF) is the cardiac dysrhythmia most commonly encountered by emergency physicians (EPs). There are currently no guidelines for elective direct current cardioversion (DCCV) of patients in AF. We theorize this procedure could be performed safely by EPs. As a first step in establishing practice guidelines for DCCV, we conducted a survey of Canadian EPs regarding their practice. METHODS: All members of the Canadian Association of Emergency Physicians (CAEP) with a Canadian address (n = 1,288) were mailed a 15-point questionnaire regarding demographics and practice patterns. Chi-square analysis was performed to analyze practice patterns overall and also in relation to demographic variables. RESULTS: We received 659 (51%) responses. Five percent, 22% and 66% of EPs reported referring their patients to internists/cardiologists for rate control, chemical cardioversion and DCCV respectively. Reported rates for EP performance of DCCV were significantly higher for physicians who had completed a residency program in emergency medicine. A similar relationship was not seen for chemical cardioversion. There were significant differences in reported referral rates for both chemical and electrical cardioversion related to type of residency training. The magnitude of this effect appeared much greater for DCCV. Additionally, EPs who reported seeing more patients/month with AF indicated a significantly higher probability of chemically cardioverting these patients themselves. Such a relationship was not reported for DCCV. CONCLUSIONS: Although the majority EPs responding to this survey reported referring patients to internists/cardiologists for DCCV, those who reported completing a residency program in emergency medicine were significantly more likely to have ever performed DCCV, as well as more likely to report incorporating DCCV as a management strategy for management of uncomplicated AF. Potential exists to reduce both cost and length of stay in emergency departments through education in the use of DCCV for uncomplicated AF.
Key words: Atrial fibrillation, cardioversion, therapy

036 Canadian Activase for Stroke Effectiveness Study (CASES).
Hill MD, Buchan AM, for the CASES Investigators. University of Calgary, Calgary, Alberta.

OBJECTIVES: Therapy for acute stroke using rtPA was approved in Canada in February 1999. The Canadian Activase for Stroke Effectiveness Study Group was formed to study the use of recombinant tissue-type plasminogen activator (rtPA) in Canada in a 2-year post-marketing study. To both prospectively assess the safety of rtPA in the Canadian context and to examine whether the efficacy of rtPA for acute stroke, demonstrated in randomized trials, can be translated into effectiveness in routine clinical practice across Canada. METHODS: The CASES group is a multi-stakeholder collaboration among the Canadian Stroke Consortium (CSC), the Canadian Stroke Society, the Heart and Stroke Foundation of Canada, Hoffmann-La Roche Canada, the Canadian Stroke Network and physicians across the country. Patient information is being collected prospectively and evaluated in a blinded fashion. Each centre has been asked to have study protocol approved by the local research ethics board. Demographics, stroke risk factors, blood pressure, biochemistry, hematology, and computed tomography (CT) scans are being collected. NIHSS and mRS scores are being collected. Outcomes will be monitored at discharge and at 3 months. RESULTS: 754 patients have been entered in the study to date. 45% of patients are independent (mRS 0-2) at 3 months and 29% have minimal or no neurological findings on examination (NIHSS 0-1). The rate of protocol violation is low at 12.5%. The rate of symptomatic intracerebral hemorrhage is 4.6% (95% CI 3.3-6.4). CONCLUSIONS: CASES is an ongoing prospective evaluation of the effectiveness of rtPA in acute stroke. The symptomatic hemorrhage rate is 4.6%. 90-day outcomes are commensurate with those observed in randomized trials.
Key words: stroke, tissue plasminogen activator, thrombolysis

037 Measures of the effectiveness of resident research curricula: a systematic review of the literature.
Frank JR, Bandiera G. Centre for Research in Education, University of Toronto, Toronto, Ontario.

BACKGROUND: There is increasing interest in developing the scholarly skills of residents. In Canada, the RCPSC has deemed "Scholar" one of 7 essential physician roles for Canadian specialty residency programs to incorporate. However no consensus exists on appropriate outcomes for measuring the effectiveness of resident research curricula. OBJECTIVES: To systematically assemble and characterize the outcome measures used to describe the effectiveness of resident research curricula. METHODS: We performed a search of the Medline, Healthstar, PsycLit, and ERIC databases from 1966 - July 2000 using the keywords: "resident", "internship and residency", and "research". Additional resources were sought in bibliographies. The search was limited to English journal literature. Publications were included if they were descriptions or studies of resident research curricula that employed outcome measures of the program intervention. Citations were selected independently by each of us, and disagreements were resolved by consensus. Outcome measures were extracted and compared. RESULTS: The Medline search produced 71 citations. 6 additional citations were found using ERIC and bibliographies. Relevant studies and outcome measures used were selected and kappas calculated. The majority of studies involved family medicine residents in the U.S., and a variety of educational interventions. Studies employed traditional academic productivity measures in 63% and satisfaction scores in 57%. Ninety-three percent of papers were positive. CONCLUSIONS: There is heterogeneity in methods and outcomes used to measure the effectiveness of resident research curricula. No meta-analysis was possible. A consensus is needed to facilitate comparison of interventions and sharing of best practices.
Key words: systematic review, research, postgraduate training

038 Utility of developmental screening in the pediatric emergency department.
Millar KR, Graboski C, Fox M, Joubert GI. Department of Pediatrics, Section of Emergency Medicine, University of Western Ontario, London, Ontario.

OBJECTIVES: Early identification of developmental delay (DD) has proven benefits to a child. Less than 30% of children with serious DD are identified prior to school entry. Children often utilize walk in clinic (WC) and emergency department (ED) services for primary care and may not be screened for DD. The Parents' Evaluation of Developmental Status (PEDS) is a parental self administered validated screening tool that has not been studied in the ED. Our objectives were to determine: the ED incidence of parental concern of DD, the proportion of unaddressed concerns, and to identify high risk ED subgroups. METHODS: The PEDS screening tool was administered in the ED to parents of children aged 2-8 yrs. We recorded the child's age, sex, diagnosis, number of visits to WC and ED over the past year, primary care doctor, PEDS score, whether DD concerns discussed with a doctor, and if not, why not, and had developmental assessment been done. Exclusion criteria: parents unable to read English or critically ill child. RESULTS: n was 413. 98.8% had a primary care doctor. 46.7% had concerns about their child's development -- only 50.8% had discussed their concerns. DD risk frequency was high (HR) in 14.8%, moderate (MR) in 17.7%, and low in 67.6%. Concerns were not discussed in 28% of HR and 53% of MR patients. Developmental assessment had not been done in 39% of HR and 88% of MR patients. Risk of DD highly correlated with increase number of ED visits (p = 0.0001). Children with more ED visits were likely to screen at high risk (0-1 visits = 11%; >5 visits = 50% [p < 0.001]). DD risk did not correlate with age, diagnosis, or WC usage (p > 0.057). CONCLUSIONS: A high number of children at risk of DD were identified for the first time in the ED (13% of 413). Those at highest risk visited the ED the most. PEDS screening in the ED is easily administered and could identify unrecognized children with DD.
Key words: developmental delay, public health, pediatric

039 Do residents interpret pediatric chest radiographs accurately?
Alghamdi A, Lynch T, Gouin S, Larson C, Patenaude Y. University of Western Ontario, London, Ontario.

OBJECTIVES: The objective of this study was to evaluate the accuracy of pediatric chest radiograph interpretation by residents in emergency medicine and pediatrics compared to pediatric radiologists. METHODS: A consecutive sample of 18 chest radiographs performed for suspected pneumonia on children (1-8 years) who presented to an emergency department over a 3-day period was independently interpreted by 10 emergency medicine residents and 10 pediatric residents. Patients with chronic cardiorespiratory disease were excluded. Each resident was asked to indicate the presence or absence of pneumonia as defined by the presence of pulmonary opacities. Each resident was able to review the information on the original radiographic requisition but was blinded to the emergency department management and the radiologist's interpretation. The 18 radiographs were independently interpreted by 3 pediatric radiologists. Consensus agreement of at least 2 out of 3 radiologists was considered to be the gold standard. Kappa statistics assessed the level of agreement between the residents in each program and the radiologists. RESULTS: The level of agreement between the emergency residents and the radiologists was moderate (average kappa ± 1 SD = 0.47 ± 0.19). The average concordance rate between the emergency residents and the radiologists was good (average concordance rate ± 1 SD = 74 % ± 10 %). Similarly for the pediatric residents, the level of agreement was moderate (average kappa ± 1 SD = 0.50 ± 0.15) and the average concordance rate was good (average concordance rate ± 1 SD = 75 % ± 8 %). CONCLUSIONS: The level of agreement in the interpretation of pediatric chest radiographs by both residents in emergency medicine and pediatrics with pediatric radiologists was moderate. Despite their lack of clinical evaluation, the accuracy of residents in both programs compared to pediatric radiologists was good.
Key words: radiography, postgraduate training

040 Emergency department revisits as a computer-automated quality assurance target.
Bullard M, Chui T, Lin C, Hu P, Liaw S. University of Alberta, Edmonton, Alberta.

OBJECTIVES: Quality assurance (QA) is often mandated by institutional accreditation leading to pressures on emergency departments (EDs) from administration to conduct QA. Unfortunately few if any resources are allocated to support these endeavours. Early ED revisits are potential markers of medical error or inadequate discharge and are targeted as an ED QA focus area. To efficiently use staff resources and establish a valuable QA process, an ED intranet was programmed to automatically alert the MD to repeat visits within 72 hours. Revisits were classified by the treating MD and later by blinded reviewers to identify complaints or diagnoses that would warrant nurse or physician education or departmental policy reviews. METHODS: During the 2-month study period, for any revisit within 72 hours, the computer generated mandatory screens required the physician to indicate the revisit reason (patient, medical, disease, system or unknown factor). These charts were later reviewed and classified by 2 independent reviewers with a 3rd MD adjudicator when needed. A random telephone sampling of all non-admitted patients was performed to determine ED revisits and a patient perceptions of reasons. RESULTS: There were 238 revisits (1.6%) for 14,690 patients over 2 months; 32% of revisits resulted in hospitalizations. The treating physician classified 81% of the revisits as patient and 10% as medical factors. The reviewers assessed revisits as resulting from patient (28%), 44% disease (44%), medical (23%), and system (2%) factors (k = 0.76). Among those classified as medical factors, 66% were admitted at the second visit. The most common chief complaints were abdominal pain (22%) and SOB (10%); while the most common diagnoses were gastroenteritis (13%) and cancer (9%). CONCLUSIONS: With 50% of revisits potentially preventable through improved medical care or education, computerized medical records have the potential to easily identify and track revisit events and contribute to QA projects.
Key words: quality, medical informatics

041 A comparison of emergency physician judgement versus a preliminary clinical model to predict severe urinary tract obstruction in renal colic.
Papa L, Stiell IG, Wells GA, Mahoney J. University of Ottawa, Ottawa, Ontario.

OBJECTIVES: To compare the predictive accuracy of emergency physicians' judgement versus a recently developed clinical model for predicting severe urinary tract obstruction in patients with suspected renal colic. This is the first prospectively derived clinical model addressing the need for urgent imaging for severe urinary tract obstruction. METHODS: This prospective cohort study of suspected renal colic patients was conducted in 2 tertiary care hospital EDs staffed with certified ED physicians. Physicians evaluated patients prior to imaging and were asked to estimate patients' pretest probability of severe obstruction based on their clinical judgement on a scale from 0% to 100%. All IVP's were reviewed by a radiologist and a urologist to identify those cases with severe ureteral obstruction. The clinical model included 4 clinical variables: vomiting, rebound tenderness, persistent pain in the ED despite IV narcotics and residual pain at discharge. Analysis of data included a comparison of the areas under the receiver operating characteristic (ROC) curves with 95%CIs. RESULTS: Of 115 patients evaluated over a 6-month period, 17 (15%) had severe obstruction. Physicians' median pretest probability of severe obstruction was 17% (95% confidence interval [CI] 14%-20%). The respective areas under the receiver operating characteristics (ROC) curve for predicting severe obstruction were physician judgement 0.64 (95% CI 0.51-0.77) and clinical model 0.89 (95% CI 0.74-1.03), (p < 0.001). Using a threshold of 10% predicted probability of severe obstruction, the respective indices of accuracy for physicians and the preliminary clinical model were sensitivity 53% vs. 90% and specificity 58% vs. 85%. CONCLUSIONS: Compared to physician judgement our preliminary clinical model for detecting severe obstruction is much more accurate. This clinical model should be prospectively validated and may prove quite useful for emergency physicians to use to determine the need for urgent imaging in cases of suspected renal colic.
Key words: renal colic, diagnosis

042 Non-contrast helical computed tomography versus intravenous pyelography for the investigation of suspected acute urolithiasis: a systematic review.
Worster A, Preya I, Hanies T. McMaster University, Hamilton, Ontario.

OBJECTIVES: In 1994, non-contrast helical computerized tomography (NHCT) of the abdomen was introduced as a method of investigating suspected acute urolithiasis. A number of studies report to varying degrees that NHCT is more accurate than other investigations in identifying patients with acute urolithiasis. Although the results of these studies are consistently in favour of NHCT, no formal review of the literature to confirm that NHCT is more accurate than IVP in this setting has been reported. METHODS: Computerized MEDLINE and hand searches for studies comparing NHCT and IVP in patients with suspected acute urolithiasis were conducted by 2 investigators in an independent and blind fashion. The results of each of the studies meeting the inclusion criteria were independently tabulated by hand by each investigator. The random effects pooled estimates of the positive and negative likelihood ratios with 95% confidence intervals were calculated along with the probabilities from calculated Z-tests and calculated Q-values for heterogeneity. RESULTS: No disagreement was found between the 2 investigators in terms of articles that met the inclusion criteria or between the results of the studies. Four studies involving a total of 296 patients met all of the a priori criteria. The positive likelihood ratios (LR+) for NHCT and IVP are 17.794 (6.901, 45.880) and 8.526 (4.157, 17.487) respectively. The negative likelihood ratios (LR-) for NHCT and IVP are 0.052 (0.028, 0.099) and 0.337 (0.205, 0.555) respectively. Heterogeneity between the four studies was significant in only one category, IVP LR- (p < 0.001). CONCLUSIONS: NHCT appears to have both a higher LR+ and lower LR- than IVP in the diagnosis of acute urolithiasis. The magnitude of these values has the potential to generate significant changes from pretest to posttest probability in the diagnosis of this disorder.
Key words: computerized tomography, renal colic, pyelography

043 How do physicians rate patients' pain after knowing patients' actual pain scores?
Papa L, Stiell IG, Wells GA, Mahoney J. University of Ottawa, Ottawa, Ontario.

OBJECTIVES: Little is known about the relative pain severity perception of patients and physicians in cases of renal colic. This study compared the pain scale ratings of patients and treating emergency department (ED) physicians in cases of suspected renal colic. METHODS: We conducted a prospective cohort study of suspected renal colic patients at 2 tertiary care hospital EDs. A 10-cm visual analogue scale (VAS) was used to assess patients' pain on arrival to the ED. After initial patient assessment, the emergency physician was shown the patient's VAS score and was asked to record their impression of the patient's pain on another VAS. At discharge, patient's pain was again recorded on a VAS scale to assess treatment. Analysis included comparison of VAS scores with the Wilcoxon Signed Ranks Test. RESULTS: 169 patients with suspected renal colic were evaluated. Mean patient VAS score on arrival was 7.1 cm (95% CI 6.6-7.6) versus mean physician VAS score 6.2 cm (95% CI 5.7-6.7) (p < 0.001). Physicians agreed with patients in 22 (13%) cases, gave patients a higher score in 40 (24%) cases and gave patients a lower score in 107 (63%) cases. Only 5% of patients were felt by physicians to be exaggerating their pain. At discharge, patients' mean VAS score was 1.0 cm (95% CI 0.7-1.3) with over 50% of patients leaving the department in a pain free state and 77% leaving with pain less than 1 cm. CONCLUSIONS: Physicians often rated the admission pain intensity lower than did the suspected renal colic patients. Despite this, physicians rarely thought patients were exaggerating their pain and most patients left the ED in a pain-free state after adequate analgesia therapy. The role of rating pain intensity in cases of suspected renal colic should be further evaluated in future studies.
Key words: renal colic, analgesia, pain

044 Delayed ultrasound in patients with abdominal pain and/or vaginal bleeding in the first trimester of pregnancy.
Hendry JN, Naidoo Y. Waikato Hospital, Hamilton, New Zealand.

OBJECTIVES: In patients with abdominal pain and/or vaginal bleeding in the first trimester of pregnancy ruling out ectopic pregnancy is essential. Clinical exam alone cannot rule out ectopic. The literature suggests that emergent ultrasound is required for appropriate diagnosis and disposition. In reality this is often unavailable. We theorized that a simple clinical decision rule could predict the safety of delayed outpatient ultrasound in these patients. METHODS: Ectopic pregnancies over a two-year period were retrospectively reviewed for clinical status at first presentation. Before data extraction "Clinically Stable" low risk patients were arbitrarily defined by the following characteristics; 1) systolic blood pressure greater than 100, 2) heart rate less than 100, 3) no risk factors for ectopic pregnancy, 4) no findings of peritonism, cervical excitation, adnexal mass or tenderness. During the study period emergent ultrasound was generally only available for patients with high clinical suspicion of ectopic pregnancy. In those patients meeting the definition of clinical stability time delay to ultrasound and whether any adverse events occurred during this delay was determined. RESULTS; One hundred seventeen cases of ectopic pregnancy were reviewed. Thirty-seven met the definition of clinical stability. These patients waited a median of 14 hours for ultrasound with 62% waiting more than 12 hours. No adverse events occurred. The risk of adverse events using this strategy is less than 3% however confidence intervals (CIs) are wide (95% CI 0%-14%). CONCLUSIONS: Preliminary results suggest that pregnant patients with abdominal pain and/or vaginal bleeding in the first trimester who meet low risk clinical criteria may have ultrasound delayed 12-8 hours without risk of adverse events. Further prospective studies are warranted to confirm the safety of this strategy.
Key words: ectopic pregnancy, diagnosis

| 001-014 | 015-029 | 030-044 | 045-059 | 060-074 | 075-093 |