**4. Clinical results**

As a result of the US FDA IDE studies, there have been a number of thorough clinical evaluations of cervical disc arthroplasty. These trials have been designed as non-inferiority studies, comparing cervical disc arthroplasty to ACDF. The Prestige ST results at two years follow-up were reported by Mummaneni et al. (Mummaneni et al, 2007) The study, similar to all of the FDA IDE studies, was a prospective, randomized, multicenter trial comparing Prestige ST cervical disc arthroplasty to ACDF for one level pathology. Five hundred fortyone patients were enrolled with 1:1 randomization; over 75% of patients were available for two-year follow-up. There were no statistically significant differences in the revision

unconstrained arthroplasty would allow unrestricted motion while a fully constrained arthroplasty would allow only flexion/extension without any anterior-posterior translation. However, compared to large joints, the differences in constraint in cervical disc

**Implant Bryan® Prestige ST® Prodisc-C® PCM-V®**

Danek

Metal-on-metal; Stainless steel

Anterior flanges with screws

Variable; superior to disc space

A number of arthroplasties have been evaluated biomechanically versus simulated fusion in human cadaveric models. These routinely show increased adjacent segment motion and increased adjacent segment disc pressures in the fusion simulations compared to cervical

As a result of the US FDA IDE studies, there have been a number of thorough clinical evaluations of cervical disc arthroplasty. These trials have been designed as non-inferiority studies, comparing cervical disc arthroplasty to ACDF. The Prestige ST results at two years follow-up were reported by Mummaneni et al. (Mummaneni et al, 2007) The study, similar to all of the FDA IDE studies, was a prospective, randomized, multicenter trial comparing Prestige ST cervical disc arthroplasty to ACDF for one level pathology. Five hundred fortyone patients were enrolled with 1:1 randomization; over 75% of patients were available for two-year follow-up. There were no statistically significant differences in the revision

**Degree of constraint** Unconstrained Semi-constrained Semi-constrained Semi-constrained

**Relative constraint** Least Less Most Less

Synthes Cervitech

Keels LP: Press-fit, ridges

Metal-on-poly; chrome cobalt, UHMWPE

(shallow sphere)

Variable; inferior to disc space

Metal-on-poly; Chrome cobalt, UHMWPE

Ball and trough design Ball and socket design Ball and socket design

Fixed; inferior to disc space

arthroplasties are limited. (Darden & Raposo, pending) (Table 1)

**Manufacturer** Medtronic Sofamor Danek, Medtronic Sofamor

Metal-on-poly; Titanium end plates, polycarbonate urethane nucleus

Milling technique creates concave endplate surfaces to fit convex endplates of the device.

Variable; lies at center of the mobile nucleus

Table 1. Summary of implant features and design characteristics.

disc arthroplasty. (DiAngelo et al, 2003; Dmitriev et al, 2005)

Biconvex nucleus articulating with upper and lower endplates

**Bearing surface materials**

**Bearing surface geometry**

**fixation**

**Primary (immediate)** 

**Implant center or rotation**

**4. Clinical results** 

surgeries at the index level (3.4% ACDF, 1.9% Prestige ST). However, the rate of surgery at adjacent levels was statistically higher for ACDF (3.4% versus 1.1%, p=0.0492). Neurological success, defined as maintenance or improvement in the neurological exam, was better with Prestige ST (92.8%) than ACDF (84.3%). Clinically, the patients were evaluated by Shortform 36 (SF-36), Visual Analogue Scale (VAS) and the Neck Disability Index (NDI). While both groups improved significantly from the preoperative state, there was no statistically significant difference between the groups at final follow-up. Overall success was defined as an NDI improvement ≥ 15 points, maintenance of the neurological status and the absence of implant-related adverse events. The arthroplasty group showed overall success in 79.3% of the patients compared to 67.8% in the ACDF group. As a sidebar, the Prestige ST patients were able to return to work on average at 45 days postoperatively, compared to 61 days for the ACDF patients.

A small prospective study compared results for Prestige LP and ACDF at a minimum of two years follow-up. Single and multilevel procedures were evaluated by VAS, NDI, SF-36 and Japanese Orthopedic Association scores. Clinically, while both groups improved significantly, there was no statistical difference between them. Motion was preserved in the Prestige LP group at a mean of 13.9° on flexion/extension lateral radiographs of two years. (Peng et al, 2011)

For the Bryan Cervical Disc Replacement, Goffin et al reported on the European experience, a multicenter, prospective, nonrandomized study, including both single-level and multi-level implants. Ninety-eight patients were evaluated at the 4 to 6 year followup point, 89 single-level patients and 9 two-level patients. The patients maintained improvement clinically at all evaluation periods. Approximately 90% of the patients had good or excellent results by Odum's criteria. The success rate for the arthroplasties, estimated by Kaplan-Meier analysis was 94% at 7 years postoperatively. One patient had removal of the arthroplasty for progressive spinal cord compression due to posterior osteophytes. (Goffin et al, 2010)

The Bryan FDA IDE study results at two years were published by Heller et al (Heller et al, 2009). Four hundred sixty-three patients enrolled, with 242 having a single-level Bryan Cervical Disc Replacement and 221 having single-level ACDF. The Bryan patients had statistically significantly improved NDI and VAS scores compared to the ACDF group at two years follow-up. Other clinical parameters improved equally between the two groups. Overall success at final follow-up was better in the Bryan patients (82.6%) versus the ACDF controls (72.7%), (p=0.010). As with the Prestige IDE patients, the Bryan patients returned to work sooner than did the ACDF patients.

Riew et al (Riew et al, 2008) evaluated a subset of patients enrolled in the Prestige ST or Bryan IDE studies that were determined to have a cervical myelopathy, defined as being hyperreflexic, having clonus or having a Nurick grade ≥ 1. In most of the patients, the cause of the myelopathy was a disc herniation. Because of enrollment criteria, multilevel cervical disease or patients with ossification of the posterior longitudinal ligament (OPLL) were excluded. A total of 107 patients in both studies were deemed myelopathic and underwent cervical disc arthroplasty. Compared to the ACDF patients, arthroplasty patients with myelopathy showed similar clinical improvement. There were no arthroplasty patients who deteriorated neurologically, suggesting that myelopathy confined to a single disc level without OPLL or retrovertebral osteophytes can be treated successfully with cervical disc arthroplasty.

Cervical Disc Arthroplasty 509

Fig. 7. Lateral flexion radiography- implanted ProDisc-C with heterotopic ossification,

preserved motion.

Murrey et al (Murrey et al, 2009) published the two year ProDisc-C IDE study. The study was structured similarly to the Prestige ST and Bryan studies, with 209 patients enrolled, 103 who underwent ProDisc-C arthroplasty and 106 who had an ACDF. Both groups showed improvement by all clinical after parameters; NDI, SF-36, VAS arm and neck pain scores and neurological success. A significant difference was in the rate of re-operation: 8.5% in the ACDF group versus 1.8% in the ProDisc-C group (p= 0.033).

Delamarter et al further evaluated the ProDisc-C IDE patients as well as 136 patients who received ProDisc-C in the continued access phase of the study, with a minimum of four years follow-up. (Delamarter et al, 2010) Demographic data remained similar between the arthroplasty and ACDF groups. All clinical parameters improved equally in both groups at all follow-up periods (p<.0001). A significant difference in the study was the rate of secondary surgical procedures. At the four year follow-up point, 12 (11.3%) of the ACDF patients had additional surgery, while only three (2.9%) of the ProDisc-C patients required further procedures (p=.0292). The ACDF patients primarily required additional surgery for pseudoarthrosis at the index level; however, six (5.6%) ACDF patients had surgery at an adjacent level. Three ProDisc-C patients were converted to fusion for axial pain. No ProDisc-C patients had to have surgery at adjacent levels. In the continued access arm of the study, one ProDisc-C patient required additional surgery to reposition the implant and two were converted to fusion for axial neck pain.

Pimenta et al (Pimenta et al, 2007) prospectively evaluated patients undergoing the PCM cervical disc replacement. Seventy-one single-level and 69 multi-level arthroplasties were performed. While both groups improved, the multi-level patients showed improved scores compared to the single-level patients. The mean NDI improvement was better in the multilevel PCM group (p=0.021). While the overall IDE results have yet to be reported, Philips et al (Phillips et al, 2009) showed in a small set of the IDE patients that arthroplasty was viable at levels adjacent to a prior fusion.
