Background: The gold standard for voice rehabilitation post-total laryngectomy is tracheoesophageal puncture and placement of an indwelling voice prosthesis (VP). The incidence of an accidental bronchial aspiration of the VP is up to 23.5% and there exists no standardised diagnostic algorithm, nor specific data on the radiologic features of these prostheses. This imaging study aims to characterize the distinct imaging characteristics of common VPs, and to make recommendations on optimal radiographic protocols to avoid the potentially fatal consequence of a missed aspirated prosthesis. Methods: Six commonly used VPs were placed into an imaging phantom and plain radiography and computed tomography (CT) imaging were performed under two dosage conditions, two reconstruction algorithms and application of iterative reconstruction (IR). Primary outcomes were visibility of VP, description of VP appearance, measurement of CT attenuation in Hounsfield Units (HU) and assessment of the optimal protocol for diagnostic purposes. Results: All prostheses were visible by plain radiography but were best seen with standard dose CT lung protocol with IR. On CT, all prostheses have HU measurements of at least >1,500 and usually >2,000. Conclusions: This study supports the use of CXR as first line in the investigation of a missing VP. If no VP is identified on CXR then CT is the second line investigation. A CT neck and chest covering the entire airway with IR reconstruction optimised for lung windows should be performed for completeness. If a VP is identified and localised, the patient should undergo urgent bronchoscopy for retrieval.