Objectives: Soft tissue air may raise suspicion for several life-threatening illnesses. Physical examination has limited sensitivity in detecting air, and computed tomography and magnetic resonance imaging are time-consuming and expensive. Sonography can show soft tissue air, but the sensitivity and specificity in this setting are currently unknown. Therefore, the purpose of this study was to assess the performance characteristics of sonography in depicting the presence, amount, and affected tissue plane in a cadaver model of soft tissue air.
Methods: We conducted a single-blinded observational cadaver study. Cutaneous sites were selected and marked (≈4 or 5 sites on each of 6 cadavers). An examiner injected air at some sites, recording both the depth (plane) and volume. A second examiner, who was blinded to the injection procedure, examined each site with sonography and recorded the presence or absence of air and the plane (subcutaneous or intramuscular). The results were compared, and statistical analysis was performed.
Results: Twenty-eight sites on 6 cadavers were used. Sites were not used if they were damaged or were within 10 cm of previous dissection. Air was injected in 20 of 28 sites; the remaining 8 were sham injections. Sonography showed air in all of the 20 sites with air injected. No air was detected in 7 of the 8 remaining sham sites, yielding 1 false-positive case. The plane was accurately assessed in all cases. Sensitivity for detection was 100%, and specificity was 87.5%.
Conclusions: Sonography accurately showed the presence, amount, and anatomic plane of soft tissue air in cadavers with sensitivity of 100%. It warrants investigation as a screening test in patients at high risk for subcutaneous air from conditions such as necrotizing fasciitis and gas gangrene.