Prevalence of ghost scans in point-of-care ultrasound for trauma patients: A multicenter study

Abstract

Introduction: Ghost scanning refers to when a point-of-care ultrasound (POCUS) is performed, but no images are saved. This is a common, yet understudied, problem with POCUS examinations and has consequences for patient care, billing, and quality assurance. Previous studies were single-site and have shown a high rate of ghost scanning, but there is a lack of direct comparison across institutions. The study aim was to compare the prevalence of ghost scanning across multiple institutions for patients with trauma activations who received an Extended Focused Assessment with Sonography in Trauma (eFAST) examination.

Methods: This was a multi-center, retrospective cohort study that took place between July 1, 2021, and June 30, 2023. Study sites included four urban Level 1 trauma centers with emergency medicine residencies and emergency ultrasound fellowships. Inclusion criteria were emergency department patients >18 years old who arrived as the highest level of trauma activation at each site. Patients were excluded if there was incomplete or missing electronic medical record (EMR) data, if the patient was determined to not be a trauma by EMR review, or if the patient was downgraded from a trauma. Investigators manually reviewed EMR documentation for performance of an eFAST examination, and then searched their respective image repository for corresponding saved images. The primary study outcome was the prevalence of ghost scanning at each site. Descriptive statistics were reported, along with site comparisons using an ANOVA test with a post-hoc Tukey honestly significant difference (HSD). Secondary outcomes included a comparison of ghost scanning rates for each site broken down both by trauma type (blunt or penetrating) and eFAST examination result (positive or negative).

Results: A total of 6078 patients were included in the analysis from the four study sites. Approximately 35 % (2182) of all trauma activations had an eFAST performed, ranging from a low of 24.5 % (444) to a high of 49.0 % (596). The overall ghost scanning rate was 49.8 % for the cohort, with 1086 images saved. The ghost scanning rate across institutions was highly variable, and ranged from 21.1 % to 93.2 % at the included sites. There was a significant difference between all sites when compared as independent pairs (p < 0.001). A statistically significant decrease in ghost scanning rates for positive eFAST examinations was seen at site 2 (p = 0.001), and site 3 had a statistically significant increase in ghost scanning rates for penetrating trauma (p = 0.02).

Conclusion: Ghost scanning is problematic and has implications for accreditation, liability, billing, quality assurance, and patient care. There was variable, but universally high rates, of ghost scanning for patients with a trauma activations at the included sites. These results show that ghost scanning continues to be a concern in the emergency department and further studies are needed to test interventions to decrease its rates.

Keywords: FAST; Ultrasound; ghost scanning; phantom scanning; trauma.