Feasibility of pneumoperitoneum diagnosis using point-of-care ultrasound: a pilot study using a fresh cadaver model

Aims: To assess the accuracy of point-of-care ultrasound (PoCUS) in the hands of two trained and blinded emergency physicians (EPs) in detecting very small amounts of free intraperitoneal air injected intra-abdominally, using a fresh human cadaver model. Material and methods: Fifteen cadavers were injected on 3 occasions with predefined quantities of free intraperitoneal air ranging from 0-10 mL. Seven cadavers were injected in the mid-epigastrium (ME), while 8 were injected in the left lower quadrant (LLQ). Each cadaver was scanned after each of the 3 injections by 2 trained and blinded EPs, resulting in 45 scans per sonographer. Scans were performed using previously validated and standardized techniques. All scans were recorded, time-stamped and labeled. For each scan the sonographers indicated "yes" or "no" to whether pneumoperitoneum was detected. A chi square analysis was performed to determine the sensitivity and specificity of PoCUS utilized by each sonographer of pneumoperitoneum based on the location and volume of air injected. Results: Free air (0.25-10 mL) injected into the ME was successfully diagnosed in 36/42 instances (86% sensitivity), but only detected in 10/36 instances when injected into the LLQ (28% sensitivity). Both EPs detected all air injections of >= 2 mL into the ME. Conclusion: Detection of free air originating from the midepigastric region may become a future PoCUS indication for adequately trained EPs.