The human cochlea exhibits extensive anatomic variations. These variations will influence the location of cochlear implant arrays and affect the potential of hearing preservation surgery. Our results may explain the surgeon's difficulties sometimes to insert electrode arrays even in so-called "normal" cochleae.

This is a review of the anatomical characteristics of human cochlea and the importance of variations in this anatomy to the process of cochlear implantation (CI). Studies of the human cochlea are essential to better comprehend the physiology and pathology of man's hearing. The human cochlea is difficult to explore due to its vulnerability and bordering capsule. Inner ear tissue undergoes quick autolytic changes making investigations of autopsy material difficult, even though excellent results have been presented over time. Important issues today are novel inner ear therapies including CI and new approaches for inner ear pharmacological treatments. Inner ear surgery is now a reality, and technical advancements in the design of electrode arrays and surgical approaches allow preservation of remaining structure/function in most cases. Surgeons should aim to conserve cochlear structures for future potential stem cell and gene therapies. Renewal interest of round window approaches necessitates further acquaintance of this complex anatomy and its variations. Rough cochleostomy drilling at the intricate ''hook'' region can generate intracochlear bone-dust-inducing fibrosis and new bone formation, which could negatively influence auditory nerve responses at a later time point. Here, we present macro-and microanatomic investigations of the human cochlea viewing the extensive anatomic variations that influence electrode insertion. In addition, electron microscopic (TEM and SEM) and immunohistochemical results, based on specimens removed at surgeries for lifethreatening petroclival meningioma and some well-preserved postmortal tissues, are displayed. These give us new information about structure as well as protein and molecular expression in man. Our aim was not to formulate a complete description of the complex human anatomy but to focus on aspects clinically relevant for electric stimulation, predominantly, the sensory targets, and how surgical atraumaticity best could be reached.

Mean diameter A was 9.3 mm. The mean basal turn length was 22.8 mm, the two first turns were 35.1 mm and the total length was 41.2 mm. Linear regression analyses indicated a coefficient of determination (R(2)) of 0.74 for diameter A and the basal turn length, R(2) = 0.70 for the two-turn length and R(2) = 0.39 for the total length.

There was no incidence of total loss of residual hearing in any of the patients. A slight deterioration of low frequency thresholds occurred in some patients. Mean hearing loss at 125-500 Hz was 14.4 dB at 1 month following surgery and 15.6 dB after 1 year. Insertion angle (300-540°) and depth (17.5-28.5 mm) were not statistically correlated to hearing loss.