• O John Ma, Gary Gaddis, Mark T Steele, David Cowan, and Kary Kaltenbronn.
• Department of Emergency Medicine, Truman Medical Center, University of Missouri-Kansas City School of Medicine, 2301 Holmes Street, Kansas City, Missouri 64108, USA. john.ma@tmcmed.org
• Emerg Med Australas. 2005 Feb 1;17(1):24-30.

ObjectiveThe objective of this study was to examine the effect of ultrasound experience level on emergency physicians' Focused Assessment with Sonography for Trauma (FAST) exam accuracy and emergency physicians' confidence in using FAST findings to assist in managing patients with blunt trauma.MethodsThis prospective, consecutive enrolment study evaluated adult trauma team activation blunt trauma patients. Based on the number of post-training FAST exams carried out, 11 attending emergency physicians were grouped into A (<25 exams, n = 4), B (26-50 exams, n = 4) or C (>50 exams, n = 3). The FAST exam was carried out prior to other diagnostic studies. The emergency physicians were asked to prospectively judge their perception of the need for surgery, abdominal CT or no further tests. All study patients ultimately underwent CT, diagnostic peritoneal lavage or laparotomy. Among each physician group, the number of subsequent CT scans deemed necessary by the emergency physician after a 'normal' FAST was calculated and compared.ResultsAccuracy was greatest in group C. Sixty-nine of 80 patients in group A had a normal FAST exam; emergency physicians deemed CT necessary in 68/69 cases (99%; confidence interval [CI] 92-100%). Eighty-two of 98 patients in group C had a normal FAST exam; emergency physicians deemed CT necessary in 19/82 cases (23%; CI 15-34%). Physicians in groups B and C were less likely to order CT after a normal FAST than group A (P < 0.001).ConclusionsFAST accuracy was greatest among more experienced emergency physicians. A normal FAST exam assisted more experienced emergency physicians with the perceived need to order significantly fewer CT scans than less experienced emergency physicians.

26 keywords...

hide…