Middle ear mechanics: using temporal bone experiments to improve clinical methods

Background The middle ear transmits and amplifies sound vibrations from the tympanic membrane via three ossicles to the inner ear. Moreover, it contains two muscles, the stapedius muscle (SM) which protects the inner ear from loud noise, and the tensor tympani (TT) whose function is still debated. The majority of hearing loss caused by disruption of the ossicular chain is a result of chronic otitis media and cholesteatoma. Variations in pathology, surgical skill and individual healing conditions make objective evaluation of ossicular replacement prosthesis in vivo difficult. Prosthesis development and the investigation of trauma mechanisms are affected by the same challenges. With few changes postmortem, the temporal bone (TB) is suitable for studies of middle ear mechanics and allows a controlled environment. Equally important, it allows theories to be tested without patient risk. In this thesis we used human TBs to find factors associated with optimal sound transfer in the two types of ossicular replacement prostheses. Furthermore, we investigated the mechanism and forces involved in rare cases of isolated malleus fractures. We also investigated the morphology, fibre phenotype composition and vascularization of the human middle ear muscles in order to better understand their roles. Materials and Methods Laser Doppler vibrometry (LDV) is an established method of measuring sound transfer in human TBs. We have further developed a surgical model that allows testing of a wide range of prostheses and their placements. In Paper I beneficial factors in partial ossicular replacement prostheses (PORPs) were tested. In Paper II we evaluated different types of total ossicular replacement prostheses (TORPs) including an experimental prosthesis inspired by the single ossicle system of birds. In Paper III the negative pressure trauma typically associated with isolated malleus fractures, produced by a finger being withdrawn from a wet ear canal after a shower or bath, was simulated in TBs. Based on measurement from control persons the forces involved were calculated and measured in models developed for this purpose. The force of the TT was estimated by comparing its cross-sectional area and fibre composition with those reported in published references. In Paper IV we used immunohistochemical, enzyme histochemical, biochemical and morphometric techniques on TT, SM and human orofacial and limb muscle control samples. Results Of the prostheses, PORPs and TORPs with lateral