Purpose: To investigate subsidence of the Charite total disc arthroplasty (TDA) and to identify if a discrepancy between vertebral endplate and the Charite footprint predispose to subsidence.

Method: Between July 2001 and May 2008, 69 patients underwent a Charite TDA (DePuy Spine, Raynham, MA). They were prospectively followed at 3, 6, 12 months, and once a year thereafter. The following measurements were performed on the replaced motion segment using a lateral radiograph:

The anterior-posterior (AP) dimension of the end plates.

The ratio between the actual and radiographic AP length of the metal endplate was calculated and utilized as the correction factor for the error of magnification on all other radiographic measurements.

Results: At L5-S1 the mean subsidence was 1.87 mm and occurred exclusively at the posterior part of the inferior end plate of L5. The mean posterior uncoverage was 3.5 mm (L5) and 0.27mm (S1). At L4-L5 the mean subsidence was 1.48 mm (L4) and 0.56 mm (L5). Posterior uncoverage of L4 and L5 vertebrae were 4.81 and 2.22 mm, respectively. Subsidence of more than 1 mm was present in all cases where the posterior uncoverage of the end plate with the TDA was more than 2 mm (odds ratio: 5.7). Subsidence was non – progressive in all cases. An anatomic mismatch exists between L5 and S1 endplates in the AP dimension; in more than half the patients S1 is shorter than L5.

Conclusion: The radiographic measurements suggest an increased likelihood of subsidence with more than 2 mm of posterior uncoverage of the end plate by the TDA. The endplate AP length of S1 is frequently less than that of L5. Implant selection based on the smaller S1 endplate may produce worrisome uncoverage of the L5 inferior endplate leading to an increased risk of subsidence and possible catastrophic failure. TDA design should afford modularity to compensate.