Diagnostic yield of cone beam computed tomography for small foreign body detection in the hand in comparison with radiography, MSCT and MRI: an ex vivo study.

INTRODUCTION

Detection of symptomatic foreign bodies (FB) after penetrating hand injuries can be challenging. Multiplanar radiography is most frequently used for FB detection and may be complemented by multislice computed tomography (MSCT) if suspected FBs cannot be identified and clinical symptoms are persisting. Cone beam computed tomography (CBCT) is a promising imaging modality for traumatology aside from fracture detection. The aim of this study was to evaluate the diagnostic yield of CBCT for different small FBs in the hand in comparison with radiography, MSCT and magnetic resonance imaging (MRI).

METHODS

In ten cadaveric hands of voluntary body donors, 20 different FBs (metal, glass, stone, wood, thorn) in predefined sizes (0.5, 1 and 2mm) were randomly placed in the central hand and the basal phalanges. All hands were imaged using radiography, 256-slice CT, CBCT, and 3T MRI. A total of 200 subcutaneous and intramuscular particles were analyzed for their visibility by two observers at two time points. The Cohens Kappa coefficient was calculated as a measure of interobserver agreement and intraobserver reliability. The particle detection rate between different imaging modalities was compared using McNemar Chi-tests.

RESULTS

CBCT and MSCT provided a higher detection rate (94.6% and 86.3%) for detecting metal, glass and stone particles compared to standard radiography (70.0%; each p<0.001). MRI did not provide a diagnostic benefit. Wood particles and thorns were not reliably recognizable by any imaging technique. The interobserver agreement (K=0.768; p<0.001) and the intraobserver reliability for both observers (K=0.914 and K=0.907; p<0.001) were good. The dose length product (DLP) was 2-fold lower in CBCT than in MSCT (39.2 ± 2.1 vs. 81.4 ± 2.9 mGy*cm; p<0.001).

CONCLUSIONS

In this ex vivo study, CBCT provided a high detection rate for small metal, glass, and stone particles while the radiation exposure was significantly lower compared to MSCT. These results suggest that CBCT instead of MSCT seems a reasonable option in supplementary diagnostics to exclude of FBs. The primary use of CBCT instead of radiography may be considered for symptomatic patients with expected small radiopaque particles <1mm. Organic FBs can be visualized indirectly in MRI and CBCT/MSCT by entrapped surrounding air.

LEVEL OF EVIDENCE

Level I, diagnostic study.