Radiographic changes after anterior cervical discectomy and their clinical impact.

Dear EditorWewouldliketopresentourexperiencewithanteriorcervicaldiscectomywithoutfusion(ACD).Thistechniquewasusedinthe department of the senior author during the years 1991–1999. A total number of 145 patients were operated on.Despite many studies concerning ACD, we felt that a reallylong term follow-up is lacking. This retrospective study eval-uates clinical and radiological results, with the longest pub-lished follow-up of 12.8 years. Our aim was to determineradiological biomarkers of poor outcome.Materials and methodsOnly patients with complete documentation who enabledus to contact them were eligible for enrollment, whichended up at 51 patients in the cohort. Four patientsunderwent three-level surgery, 27 two-level, and 20 one-level surgery. The number of surgeries in each level was2 at level C3/4, 13 at level C4/5, 42 at level C5/6, 28 atlevel C6/7, and 2 surgeries at level C7/Th1.AP lateral and dynamic X-ray was performed. The sagittalalignment of the cervical spine was measured using Cobb’sangle method between the inferior margin of C2 and theinferior margin of C7. The sagittal alignment was also evalu-ated at the level of surgery by measuring the angle betweenlines drawn at the superior margin of the superior and theinferior margin of the inferior vertebral body. Bone fusion atthe level of the surgery was studied in lateral and dynamicprojections using the criteria of Lee at al [4]. Odom’scriteriawere used for clinical evaluation [7].ResultsFour patients showed kyphosis of the cervical spine(7.84 %). Lordosis or straight axis was present in 47patients (92.16 %). Kyphosis in the level of surgery,which was considered as radiological failure, was pres-ent in 19 cases (37.25 %). 41 patients had definitefusion, and 5 patients had probable fusion; 3 patientshad possible pseudoarthrosis (5.88 %), and 2 had defi-nite pseudoarthrosis (3.92 %). Clinical examination di-vided the group (according to Odom’s criteria) intoclinical success with excellent (39.22 %), good(27.45 %), or fair (19.61 %) results and into failurewith poor results (7 patients, 13.72 %). Within thecohort we did not find any patient who required imme-diate reoperation.There was no correlation between the number of operatedsegments and poor clinical outcome.Radiographic and clinical correlationsThe cervical spine kyphosis was considered as unnaturalalignment and, therefore, should be associated with pooroutcome. However, no patient from the poor outcome grouppresented C2-7 kyphosis. Local kyphosis at the level of sur-gery, which was present in 5 out of 7 patients in the poor