Feasibility and Biomechanics of Multilevel Arthroplasty and Combined Cervical Arthrodesis and Arthroplasty

A new experimental protocol was applied utilizing a simplified postural control model. Multiple constructs were tested nondestructively by interconnecting segmental rods to screws. To investigate how posture and distribution of segmental angles under physiological loads are affected by combined cervical arthroplasty and fusion. Previous studies of biomechanics of multilevel arthroplasty have focused on range of motion and intradiscal pressure. No previous study has investigated postural changes and segmental angle distribution. In 7 human cadaveric C3-T1 specimens, C4-C5, C5-C6, and C6-C7 disks were replaced with ProDisc-C (Synthes). Combinations of fusion (f) adjacent to arthroplasty (A) were simulated at C4-C5, C5-C6, and C6-C7, respectively: fAA, AfA, AAf, ffA, fAf, Aff, fff. C3-C4 and C7-T1 remained intact. A compressive belt apparatus simulated normal muscle cocontraction and gravitational preload; C3-C4, C4-C5, C5-C6, C6-C7, and C7-T1 motions were tracked independently. Parameters studied were segmental postural compensation, neutral buckling, and shift in sagittal plane instantaneous axis of rotation (IAR). With one or more levels unfused, the arthroplasty levels preferentially moved toward upright posture before the intact levels. Neutral buckling was greatest for 3-level arthroplasty, less for 2-level arthroplasty, and least for 1-level arthroplasty. Among the three 1-level arthroplasty groups (ffA, fAf, Aff), arthroplasty at the caudalmost level resulted in significantly greater buckling than with arthroplasty rostralmost or at mid-segment (P