32. DJK and revision rates in multilevel posterior cervical fusions terminating at the cervicothoracic junction

Abstract

<h3>BACKGROUND CONTEXT</h3> Despite the consensus that posterior cervical fusion can be used to treat multiple cervical pathologies, questions remain as to if crossing the cervicothoracic junction (CTJ) influences the rate of revision and how radiographic outcomes may differ. Even less is known about how crossing the CTJ may predict the development of adjacent segment disease. A retrospective review of 91 posterior cervical fusion patients was performed comparing long constructs (≥3 levels) terminating at C7 vs T1-3. Revision and DJK rates were compared between the two groups. Mean change in cervical lordosis, T1 slope, cervical SVA, and T1 slope-cervical lordosis mismatch (T1S-CL) were also compared. Distal segment kyphosis at final follow up was significantly greater in the thoracic group, but no other statistically significant differences were identified. <h3>PURPOSE</h3> Compare revision rates and radiographic outcomes in patients receiving posterior cervical fusion of ≥3 levels that terminate at C7 vs T1, 2 or 3. Our hypothesis is patients with multilevel posterior cervical fusion constructs terminating at C7 will have similar rates of revision, DJK and cervicothoracic radiographic outcomes compared to patients whose constructs crossed the CTJ to T1, T2, or T3. <h3>STUDY DESIGN/SETTING</h3> Single center retrospective analysis. <h3>PATIENT SAMPLE</h3> Consecutive posterior cervical fusion patients with greater than or equal to 3 levels fused from January 2010 until December 31, 2019 at a single institution. <h3>OUTCOME MEASURES</h3> Primary outcome: 1. Need for revision. Secondary outcomes: 2. Distal junctional kyphosis; 3. Mean change in cervical lordosis (CL); 4. Mean change in T1 slope (T1S); 5. T1S - CL mismatch; 6. cervical SVA. <h3>METHODS</h3> A retrospective review of medical records identified posterior cervical fusion cases with at least one year of radiographic follow-up and fusion greater than or equal to 3 segments. Identified cases were divided into 2 groups based on location of lowest instrumented vertebra. Group 1 included fusions terminating at C7 and group 2 included fusions terminating at T1, 2, or 3. At multiple intervals, cervical lordosis (CL), T1 slope, cervical sagittal vertical axis (cSVA), distal segment kyphosis, and T1 slope-cervical lordosis mismatch (T1S-CL) were measured. Any need for revision surgery was also documented. <h3>RESULTS</h3> A total of 91 patients were included, 53 in Group 1 (mean age 58.9 years, 56.6% women) and 38 in Group 2 (mean age 64.4 yrs, 55% women). Revision rate did not reveal statistical difference between the two groups (Group 1: 9.4% vs Group 2: 2.6% P=0.39). There was no statistical difference in patients meeting criteria for DJK at final follow up (Group 1: 5.6% vs Group 2: 5.2% P=0.9). However, within each group, a statistically significant increase in mean distal segment kyphosis at final follow-up was identified in Group 2 (Group 1: 0.82°, P=0.31 vs Group 2: 2.5°, P=0.0001). Mean change in CL, T1 Slope, cSVA and T1S-CL mismatch were not statistically different between the groups at final follow up. <h3>CONCLUSIONS</h3> Our study demonstrated that revision rates, cervicothoracic radiographic parameters, and DJK were not significantly different when comparing multilevel posterior cervical fusions terminating at C7 vs the upper thoracic spine. <h3>FDA DEVICE/DRUG STATUS</h3> This abstract does not discuss or include any applicable devices or drugs.