23 The first-generation ProDisc (Aesculap AG & Co., Tuttlingen, Germany) was developed in 1989 by Thierry Marnay, M.D., at the Clinique du Parc, Montpellier, France. Between March 1990 and September 1993, Marnay and Louis Villett, M.D., (Dunkerque, France) implanted 93 ProDisc I devices into 64 patients at either one or two levels. No further devices were implanted and they waited until the results were evaluated at a mean follow-up of 8.7 years (range, 7–11 years). After adequate time had elapsed, investigators made an exhaustive effort to locate all patients. From the initial 64 patient cohort, three patients had died and three patients were not found. The remaining 58 patients (95% of all living subjects) were studied. By Marnay’s evaluation, all implants were intact and functioning. There was no evidence of subsidence or migration. There was a significant reduction in subjective back and leg pain. In terms of satisfaction, 92.7% of patients were either “satisfied” or “entirely satisfied.” There were no differences between the single-level and two-level procedures. Importantly, “there were no device related safety issues, untoward effects, complications or adverse effects.”1 The second-generation design, ProDisc II (Aesculap AG & Co.), was released to European markets in December 1999. The new device featured several changes in implant design. In 1999, the device had been acquired by Spine Solutions, Inc., which was founded by the Viscogliosi Brothers, LLC. In October 2001, the first ProDiscs were implanted in the United States under a Food and Drug Administration (FDA) Investigational Device Exemption (IDE). In early 2003, ProDisc was acquired by Synthes-Stratec (Oberdorf, Switzerland). The end plates of the first-generation ProDisc were fashioned entirely of titanium alloy. The ProDisc II end plates are a cobalt-chrome-molybdenum (CoCrMo) alloy coated with a plasmapore titanium alloy that allows for bone growth into the device. The initial ProDisc design had dual keels on each end plate. The second-generation model has a single central keel and two spikes on each end plate that provide immediate rotational stability and allow for bony ingrowth. The cranial end plate has a highly polished concave bearing surface that articulates with the convex polyethylene core. The central core of the device is an ultra high molecular weight polyethylene (UHMWPE) liner that snaps into the caudal end plate, creating a semiconstrained device. The monoconvex design of the liner allows it to be placed without requiring overdistraction of the disk space. The third modification to the first-generation device, the addition of a modular polyethylene, allows the surgeon more options in the reconstruction of the disk space. The resulting motion has been compared with a ball and socket articulation.2,3 The device allows 13 degrees of flexion, 7 degrees of extension, 10 degrees of lateral bending, and ±3 degrees of axial rotation4 (Fig. 23–1). The implant was designed to approximate the natural physiological center of rotation that normally exists just inferior to the superior end plate of the caudal vertebra.5 Cadaveric studies comparing instantaneous axes of rotation of the ProDisc II to radiographically normal L5–S1 segments has shown that there is not a significant difference between the paths of motion.6 During flexion and extension, there was similar vertical motion, whereas with lateral bending, there was comparable horizontal motion (Fig. 23–2). In addition, the model showed that there was an increase in the coupled motions of rotation and lateral bending. Of specific interest was the resultant unloading of the facet joints. There was a decrease in facet shear forces by 37%. Although a cadaveric model cannot clearly answer clinical questions, it implies that pristine facet joints may not be a requirement with a semiconstrained device. Current studies are attempting to clarify this clinical question. Huang et al retrospectively reviewed the radiographs of Marnay’s initial cohort.7 They have shown that 66% of ProDisc I implants had > 2 degree range of motion at a mean follow-up of 8.7 years. Of the two thirds of disks that continued to exhibit measurable motion, the mean range of motion was 5.5 degrees (range, 4.1–7.5 degrees), which was actually less than measured normal ranges of motion in a group of asymptomatic subjects.8 Perhaps the most important finding is the correlation between range of motion and adjacent level degeneration. The authors defined degeneration as greater than 2 mm of disk collapse, anterior osteophyte formation, or greater than 3.5 mm of dynamic instability. Patients who maintained at least 5 degrees of motion at their most cephalad ProDisc had a 0% chance of adjacent level degeneration. On the contrary, patients with less than 5 degrees of motion had a 34% prevalence of adjacent segment degeneration. This may be the most compelling evidence that the ProDisc can protect from adjacent segment degeneration with the caveat that a threshold 5 degrees of motion is maintained.9 Interestingly, the same authors showed that ranges of motion were greater in the ProDisc II. In the later model, the average range of motion at the L4–L5 level was 10 degrees (range, 8–18 degrees) and at L5–S1 it averaged 8 degrees (range, 2–12 degrees).3 The second study had a mean follow-up of 1.4 years, so it is unclear whether the improved range of motion reflects improvements based on the second-generation design, better surgical technique, or better patient selection. On the other hand, it is possible that this motion could change with a longer follow-up period. No clear correlation existed between disk mobility and clinical outcomes. Additionally, no significant correlations were found between lack of motion and age, weight, lumbar level replaced, number of levels, or history of prior surgery. Huang and associates7 found that female gender was correlated with a higher likelihood of not having measurable motion, even though there was no difference in mean range of motion between male or female patients. Instead, there were more outliers in the female group. None of the radiographs showed signs of instability above or below the level of the disk replacement. However, 24% of patients developed either loss of height or annular traction osteophyte formation adjacent to the replaced level. One investigator showed that there was an increase in lumbar lordosis after ProDisc arthroplasty.10 However, a second study showed no difference in sagittal alignment after implant with the ProDisc II prosthesis.3 This study also showed that there was a significant correlation between lack of motion, poor sagittal plane alignment, and the development of adjacent level breakdown. This may lend credence to the theory that the improved motion allows the spine to “seek” a more desirable lordotic position and may be essential in protecting against adjacent level disease. Le Huec investigated the role of UHMWPE in dampening vibration and shock transmissibility in the ProDisc’s metal-on-poly bearing surface compared with a metal-on-metal design.11 The shock absorption ratio was measured in a bio-mechanical testing apparatus. Despite the appreciably lower modulus of elasticity of the UHMWPE (0.8 GPa) versus the chrome-cobalt (235 GPa) there was no difference in load dissipation. Unfortunately, the clinical implications of this finding are not easily decipherable given the lack of information about dynamic loading (shock or vibration) of the human disk in vivo. The ProDisc is typically implanted through a left, mini-anterior retroperitoneal approach (although a transperitoneal approach may be considered under special circumstances). The incision may be transverse for single-level or longitudinal paraumbilical for multilevel reconstructions. The anterior rectus sheath is opened and the rectus abdominis is retracted laterally. The retroperitoneal space is entered inferior to the arcuate line. The peritoneum is retracted medially.12 At our institution, a Balfour self-retaining retractor is utilized longitudinally, and Wiley handheld retractors are used medially and laterally rather than self-retaining retractors. Our access surgeons believe that the great vessels can be more safely handled with the proprioceptive input from a handheld retractor, but table-based self-retaining systems are certainly acceptable. At our institution, the access surgeons remain scrubbed throughout the entire procedure to safely retract and protect the vessels.
ProDisc Lumbar Artificial Disk
Prosthesis Design
Biomechanical Studies
Implantation Technique