[Ventral interbody spondylodesis in injuries of the cervical spine. Indications, surgical technique and results].

Lower cervical spine injuries with instability of the anterior and/or posterior column can be treated by anterior interbody fusion and plate fixation. Plates available for anterior instrumentation of the lower cervical spine can be divided into locking or non-locking systems with uni- or bicortical screw purchase. Our biomechanical comparative testing of different screw fixation systems demonstrates improved stability with the use of bicortical purchase. Clinical studies, however, have proven high fusion rates without loss of correction and a low implant related morbidity with the use of unicortical as well of bicortical plate systems. Correct reduction and intraoperative positioning of the unstable cervical spine is crucial to avoid implant related complications. Also, limitations of anterior instrumentation for the treatment of specific lesions of the lower cervical spine have to be considered, e.g. in complex lesions with axial instability or in fracture dislocations with ankylosing spondylitis. Changes or alterations of adjacent segments can be reduced by the use of plates with correct lengths, contact of uninjured adjacent discs with implants should be avoided. A comparative analysis of two patient collectives--89 patients (1972-1983) and 102 patients (1987-1994), all of them treated with bicortical plate fixation--revealed different results in terms of implant failure, operative reduction and loss of correction. All but one surgical fusions had healed radiologically. Implant related complications during the first 3 months after the initial operation were lower in the latter group, only 3 out of 102 patients (3%) with implant loosening versus 7 our of 89 patients (8%) with implant breakage or loosening required surgical revision. In all cases technical errors could be detected. Clinical follow-ups with personal examination was performed in 144 patients: 57 of 72 survivors of series I (79%) after an average time of 11 years 9 months and 87 out of 94 survivors of series II (85%). The radiologic examination revealed 2 patients with screw breakage in series I, one patient with an asymptomatic implant loosening in series II. Only one case was observed with a loss of correction after loosened and early removed hardware. In all other patients there was no difference of radiologic angles between postoperative X-ray and follow-up. 16 patients, 12 of series I, 4 of series II, were fused in a kyphotic position after insufficient preoperative reduction. Radiologic alterations of adjacent segments, i.e. spondylophyts or "spontaneous" fusions, were observed in more than 50% of all patients of both series. However, complaints or persistent pain did not correlate with radiologic findings. Also in both series there was a high percentage of patients with mild, residual neck pain in spite of a very good radiologic result. Only in a very few cases the complaints had to be treated by drugs.