Comparison of computerized tomography and ultrasound for diagnosing soft tissue abscesses

Patients presenting with a suspected skin abscess that underwent both US and CT imaging were eligible for inclusion. Two physicians blinded to patient characteristics and other imaging results prospectively reviewed the CT and US images for pre-defined image elements, and in circumstances where there was disagreement between these interpretations, a third physician adjudicated the findings. The presence or absence of an abscess cavity was noted on imaging. Imaging detail was summarized using a pre-specified 4-point scale based on the degree of visible detail with higher numbers corresponding to greater detail. The clinical presence of an abscess was defined by surgical evacuation of purulence. Sensitivity and specificity for both CT and US were calculated using Chi square analysis. Comparison between imaging detail was performed using a Student's T-test. Data are presented with (95% confidence intervals) unless otherwise noted.Over an 18 month period 612 patients received a soft tissue bedside ultrasound with 65 of those patients receiving a CT for the same complaint. 30 of these 65 patients had an abscess located in the head and neck (37%), buttock (17%), lower extremity (17%), upper extremity (13%), torso (13%), or hand (3%). US demonstrated a sensitivity and specificity for the diagnosis of abscess of 96.7% (87.0% to 99.4%) and 85.7% (77.4% to 88.0%) respectively. The overall sensitivity and specificity of CT for the diagnosis of an abscess was 76.7% (65.5% to 82.8%) and 91.4% (81.8% to 96.7%) respectively Overall image detail ratings were superior for US compared to CT (3.5 vs 2.3, p = 0.0001).US is more sensitive then CT, but CT is more specific for superficial soft tissue abscesses. US demonstrated more visible detail within the abscess cavity compared to CT.The incidence of skin and soft tissue infections in the ambulatory setting has more than doubled over a ten-year period to 3.4 million emergency department visits in 2005 [1]. The epidemiology of skin and