EBM --- Journal Reading Presenter ：李政鴻 Date ： 2005/10/26

Users’ Guides to the Medical Literature III. How to Use an Article About a Diagnostic Test A. Are the Results of the Study Valid ？ Roman Jaeschke, MD. MSc; Gordon Guyatt, MD. MSc; David L. Sackett, MD. MSc; for the Evidence-Based Medicine Working Group JAMA; Jan 2, 1994; 271, 5

78 year old woman –Increasingly SOB over the last 24 hours –Chest discomfort (worse by taking a deep breath) –10 days after abdominal surgery –PE : residual tenderness in the abdomen and scattered crackles at both lung bases. –CXR: small right pleural effusion(ten days ago) –ABG: PO2 of 70, with a saturation of 92% –EKG:only non-specific changes Clinical situation

Suspect pulmonary embolus. –Ventilation-perfusion scan: "intermediate probability" for PE. –Order full anticoagulation. –Understanding is based on local practice rather than the properties of V/Q scanning from the original literature. Clinical situation

Search 1.Medical subject (MeSH) headings "pulmonary embolism". 2.“Radionuclide imaging" as a subheading 3.“Comparative study" 4.Need a study comparing V/Q scanning to some reference standard. The PIOED investigation. Value of ventilation perfusion scan in acute pulmonary embolism: results of the prospective investigation of pulmonary embolism diagnosis JAMA :

Reading Articles about Interpreting Diagnostic Test Results Are the results of the study valid? –Primary guide –Secondary guide What are the results of the study? –Likelihood ratios for the test results Will the results help me in caring for my patients? –Reproducibility –Applicable –Wound change my management? –Get better?

Primary Guides Independent, blind comparison with reference standard? Compare with truth Appropriate patient sample spectrum for clinical practice? CEA in colorectal cancer: Not a good diagnostic tool. –Elevated in advanced cancer –Only for follow-up and screen In the PIOPED study –high, medium, and low clinical suspicion of PE –Appropriate patient sample was chosen.

Goal of primary guide Convinced that the article –Appropriate spectrum of patients –Underwent the independent, blind comparison of a diagnostic test and a reference standard  Unbiased estimate of real accuracy

Secondary Guides Did the results of the test being evaluated influence the decision to perform the reference standard? Verification bias (or work-up bias): –the properties of a diagnostic test will be distorted if its results influence whether patients undergo confirmation by the reference standard Second reference standard –For those whose test results are negative E.g. exercise stress testing, PIOPED study

For sensitivities and specificity of V/Q scans for acute pulmonary embolus. Design: –Multicenter, Prospective, blinded study –933 random samplings of V/Q scans from 1493 patients. –Recruitment was done prior to V/Q scan completion. F/U: Telephone interviews at 1, 3, 6, and 12 months. Intervention: – 755 of 933 had pulmonary angiogram after the V/Q scan. –The others had close follow-up and were considered negative for pulmonary embolus if they were discharged without anticoagulation or had no events within one year of follow-up. PIOPED

Conclusion: large magnitude of bias 69% 92% Angiogram perform rate

PIOPED-no angio part I Add 150 patient (low/ near normal,normal) 136  no, 14  uninterpretable Follow up 1 year  no clinical PE, no anticoagulation

PIOPED-no angio part II Another 50 patient (high, intermediate), no angio 1.small proportion 2.the same clinical characteristics

Were the methods for performing the test described in sufficient detail to permit replication? Must tell you how to use it. All issues that are important –Preparation of the patient (diet, drugs to be avoided, precautions after the test) –Performance of the test (technique, possibility of pain) –Analysis and interpretation of its results. Secondary Guides

Reading Articles about Interpreting Diagnostic Test Results Are the results of the study valid? –Primary guide –Secondary guide What are the results of the study? Will the results help me in caring for my patients?