The Effect of Laminectomy with Instrumented Fusion Carried into the Thoracic Spine on the Sagittal Imbalance in Patients with Multilevel Ossification of the Posterior Longitudinal Ligament

OBJECTIVE: To determine if there is a difference in either the cervical alignment or the clinical outcomes in cervical ossification of the posterior longitudinal ligament (OPLL) patients who underwent laminectomy with instrumented fusion (LIF) ending at C(6), C(7), or proximal thoracic spine for the treatment of multilevel OPLL, and to find out the appropriate distal fusion level. METHODS: This was a single‐center retrospective study. In total, 36 patients with cervical OPLL who underwent three or more level LIF in our institution between January 2015 and January 2017 were enrolled. They were divided into three groups according to their distal ends: C(6) (nine females and 11 males, 60.45 ± 9.68 years old), C(7) (four females and six males, 61.60 ± 10.29 years old), and T‐group (two females and four males, 64.33 ± 8.12 years old). Radiographic (compression level, classification of OPLL, occupying rate, C(2‐7) cobb angle, C(2‐7) sagittal vertical axis, and fusion level) and clinical outcomes (NDI score, operative time, and blood loss) were compared. Predictors of postoperative sagittal imbalance were also identified according to if the postoperative C(2‐7) SVA was greater than 40 mm. The sensitivity and specificity of preoperative parameters predicting postoperative cervical stability were evaluated via the receiver operating characteristic (ROC) curve. RESULTS: All patients were followed up at least 1 year. The blood loss in T group was significantly more than C(6) or C(7) group. The length of fusion level became significantly longer when the caudal level extended to the thoracic spine. The age, preoperative SVA, and NDI score at follow‐up were significantly greater in the imbalance group. At the final follow‐up, the cervical lordosis tended to be straight and the C(2‐7) SVA tended to be greater when the caudal level of fusion was extended to upper thoracic segment. Further ROC curve analysis suggested that patients’ age had a sensitivity of 75.00%, specificity of 79.17% for cervical stability, and the AUC was 0.844 (P < 0.01), with the cutoff value for age being 66.5 years old. For preoperative SVA, the sensitivity was 58.30%, and specificity was 91.70%, with the AUC of 0.778 (P < 0.01). The cutoff value for preoperative SVA was 30.4 mm. CONCLUSION: Although posterior fusion terminating in the thoracic spine was not superior to the cervical spine for patients with multilevel OPLL, for elderly patients (>67 years) with great preoperative SVA (>30 mm), terminating at C(6) was recommended to limit the invasion of cervical extensor muscles, provided the decompression was adequate.