CAEP/ACMU 2000 Scientific Abstracts: 11-14

CAEP Abstracts

CJEM 2001;2(3):169-187

011 Is the Lateral Chest Radiograph Required in the Diagnosis and Management of Children with Suspected Pneumonia? A Randomized Clinical Trial.
Lynch T, Gouin S, Larson C, Patenaude Y. McGill and Montreal Universities, Montreal, Que.

OBJECTIVES: To determine if the addition of the lateral chest radiograph (CXR) to the frontal CXR influences the diagnosis and management of children with suspected pneumonia in a pediatric emergency department (ED).
DESIGN: Randomized clinical trial.
SETTING: Montreal Children's Hospital ED.
POPULATION: Patients 1­16 years of age in whom a CXR was ordered for the clinical suspicion of pneumonia, were eligible for enrollment from May 1998 to December 1999. Exclusion criteria were: chronic cardiac or respiratory disease, concurrent asthma exacerbation, pneumonia confirmed by CXR in the previous 8 weeks or antibiotic use in the previous 2 weeks, gastroesophageal reflux disease and spastic quadriplegia. Critically unstable patients were also excluded. Patients were recruited while one of 11 Board-Eligible/Certified Pediatric Emergency Physicians (PEP) was attending in the ED. Frontal and lateral views of the chest were obtained for each patient. Patients were randomized into one of 2 groups. In Group A, only the frontal CXR was available for PEP review while in Group B, both views were returned for PEP review. The PEP CXR interpretations for each group were compared to a gold standard taken as consensus agreement of at least 2 out the 3 radiologists who interpreted both views obtained. Post ED-visit, patient management changes (i.e., missed pneumonias by PEP = false negatives) were based upon the initial review of the 2 views by a group of 7 radiologists.
RESULTS: There were 534 eligible children during the study period; 33 families declined to participate. Of the remaining 501 patients, 240 were randomized to Group A (1 view) and 261 to Group B (2 views). Both groups were similar in: sex, age, clinical symptoms and signs, treatment provided, and final disposition. The PEP CXR interpretation was equal in accuracy (true positive and true negative diagnoses) between group A and B (70% vs. 72%, RR ± 95% CI = 0.95, 0.78, 1.15). The proportion of false negatives was lower in group A compared to group B (3.4% vs. 4.6%, RR ± 95% CI = 0.85, 0.50, 1.43).
CONCLUSIONS: The addition of the lateral CXR to the frontal CXR does not improve the diagnostic accuracy of the PEPs in the diagnosis of pneumonia in children 1­16 years of age. Furthermore, the addition of the lateral CXR may increase the proportion of false negative results.
Key words: adiography, pneumonia, diagnosis, pediatric

012 Comparison of the Predictive Accuracy of Physician Judgement vs. the Canadian C-Spine Rule.
Stiell I, Wells G, De Maio V, Vandemheen K, Clement C, Greenberg G, Lesiuk H, Brison R, Cass D, Dreyer J, Eisenhauer M, MacPhail I, McKnight D, Morrison L, Reardon M, Schull M, Worthington J. Clinical Epidemiology Unit, University of Ottawa, Ottawa.

OBJECTIVES: To compare the predictive accuracy of emergency physicians' clinical judgement to the Canadian C-Spine Rule, a recently developed and highly sensitive clinical decision rule for the use of cervical spine radiography.
METHODS: This prospective cohort study was conducted as a component of the Canadian C-Spine/CT Head (CCC) Study in 10 Canadian EDs and involved alert and stable adult trauma patients. Physicians prospectively evaluated patients prior to radiography and performed interobserver assessments when feasible. Physicians also estimated, based upon clinical judgement alone, the probability of unstable cervical spine injury from 0% to 100%. Patients underwent radiography, CT, and follow-up to determine clinically important cervical spine injury. Analyses included comparison of areas under the receiver operating characteristic (ROC) curve with 95% CIs and the kappa coefficient.
RESULTS: Among 6,277 patients enrolled over 18 months, the mean age was 36.5 (range 16­97); 50.6% were male; and 64 (1.0%) had a clinically important injury. The physicians' kappa value for a 0% predicted proba bility of injury was 0.46 (95% CI 0.28­0.65). The respective areas under the ROC curve for predicting C-spine injury were physician judgement 0.85 (95% CI 0.80­0.90) and Canadian C-Spine Rule 0.87 (95% CI 0.85­0.89), (p = 0.22). Using a threshold of 0% predicted probability of injury, the respective indices of accuracy for physicians and Canadian C-Spine Rule were sensitivity 92.2% vs. 100% (p <0.001) and specificity 53.9% vs. 46.6% (p <0.001).
CONCLUSIONS: While physician judgement shows good overall discrimination between injury and non-injury cases, inter-observer agreement is only fair, and sensitivity is unacceptably low. The accuracy, reliability, and acceptability of the Canadian C-Spine Rule should be explicitly and prospectively evaluated prior to widespread clinical use.
Key words: cervical spine, radiography, clinical prediction rule, utilization

013 The Availability of "Urgent" Ultrasound in Canadian Emergency Departments. A Survey of Emergency Department Directors.
Bajwa GS, Stiell I, Hebert G, Lee J. University of Ottawa, Ottawa.

OBJECTIVES: To determine the availability of "timely" ultrasound (U/S), accessibility and prevalence of U/S in emergency departments (ED), and the interest of directors in the use and implementation of U/S in EDs.
METHODS: A mailout survey was sent to the directors of 300 of 800 of the largest EDs across Canada. The survey consisted of 13 questions on demographics, availability of U/S in their departments and the potential future use by emergency physicians (EPs).
RESULTS: The response rate to the survey was 52% (155/300) of which 65% were non-teaching hospitals. While 64% of EDs claimed to have after-hours coverage for U/S, the facility to obtain an U/S was rated as difficult 50% of the time. And if U/S was available, 48% of the time it would take between 1 and 3 hours to obtain. 60% of directors surveyed felt that U/S should be performed by EPs and 79% were interested in learning. However, the overwhelming majority, 97%, stated that U/S was not being performed by EPs in their department.
CONCLUSIONS: Only 47% said they could obtain an urgent U/S within an hour and 2% within 15 minutes. Only 55% of non-teaching hospitals have 24-hour coverage, and only 50% of those with coverage reported easy access. Across Canada, 5 of 800 EDs have emergency physicians performing focused U/S. There is, however, substantial interest in implementing ED U/S, and even greater interest in EPs learning to perform U/S.
Key words:ultrasonography, emergency department

014 DVT Treatment Protocols in Canadian Emergency Departments.
Spooner C, Kelly KD, Barker S, Holroyd BR, Bullard M, Voaklander D, Rowe BH. University of Alberta, Edmonton, Alta.

OBJECTIVES: To examine the site characteristics, availability of diagnostic tests, and treatments used for deep-vein thrombosis (DVT) in emergency departments (EDs) across Canada.
METHODS: Surveys were mailed to 225 ED directors across Canada. Introductory letters, follow-up reminders, phone calls and thank-you letters were used to increase participation rates. The survey consisted of 20 questions regarding site demographics, availability of testing modalities, and existence of DVT and "rule-out" DVT protocols. Sites were compared on the basis of whether they had an outpatient DVT management protocol in place or not.
RESULTS: 143 sites (64%) returned surveys (11% French; 89% English), and 50 (35%) provided full details of their DVT protocols. The sample consisted of 43 (30%) teaching hospitals, 83 (58%) non-teaching and 39 (27%) regional referral centres. Most hospitals (80%) had <500 beds, but 78% had >25,000 ED visits/year. Compared to those sites without protocols, those with a DVT protocol were more likely to have a practising hematologist (p = 0.003), employ full time ED physicians (p = 0.005), and have more physician hours of weekday coverage (p = 0.04). At sites with a DVT protocol, leg ultrasound (96% vs. 87%; p = 0.05) and V/Q scans (74% vs. 53%; p = 0.008) were more readily available than hospitals without a protocol. In terms of treatment, protocol sites had dedicated DVT teams (41% vs. 10%; p < 0.001), rule-out DVT protocols (39% vs. 8%; p < 0.001), and used low molecular weight heparin (LMWH) for rule-out DVT pending imaging (90% vs. 61%; p < 0.001) and proven DVTs (97% vs. 77%) significantly more often than non-protocol sites. Most sites were interested in participating in DVT research (73%) and receiving DVT CME (93%).
CONCLUSION: DVT protocols exist in less than half of Canadian EDs, and there is significant variability in service delivery, treatment and monitoring. Standardized protocol development may improve this situation.
Key words: deep vein thrombosis, emergency department, diagnosis