Abstract INTRODUCTION For patients undergoing posterior cervical laminectomy and fusion (PCLF) with an upper thoracic lower instrumented vertebra (LIV), the evidence is unclear as to whether C2 vs C3 is the ideal upper instrumented vertebra (UIV). This study analyzes short-term postoperative outcomes for PCLF comparing cervicothoracic constructs with UIV at C2 and C3. METHODS Adult patients with degenerative cervical spine disease undergoing PCLF from 2012 to 2018 at a single center were identified. Patients with UIV at C2 or C3 and a lower instrumented vertebra (LIV) at T1 or T2 were included. Univariate analysis was performed using Student's t-test, Chi-squared test, or Fisher's exact test. Multivariate regression including variables significant at P = .20 determined the effect of UIV on clinical and radiographic outcomes. RESULTS A total of 74 patients were included, of whom 30 (40.5%) and 44 (59.5%) had a UIV at C2 and C3, respectively. Mean follow-up time was 13.8 mo. Baseline characteristics including age (C2 62.4 vs C3 63.0, P = .80) and preoperative Nurick score (C2 2.3 vs C3 2.2, P = .57) were similar across groups, though the C2 UIV group had a higher rate of revision surgery (23.3% vs 2.7%, P = .006) and diabetes (40.0% vs 9.1%, P = .003). Multivariate analysis of clinical outcomes showed no significant differences, including in the rate of postoperative complications (OR: 1.5, P = .62), discharge home (OR: 1.2, P = .35), hospital length of stay (ß = −1.17 d, P = .28) or Nurick score (ß = −0.46, P = .22). Differences in radiographic outcomes including postoperative cervical lordosis (ß = −7.4 degrees, P = .09) and cSVA (ß = −8.9 degrees, P = .08) approached but did not reach significance. CONCLUSION In a study of PCLF with upper thoracic LIV, there were no significant short-term clinical or radiographic differences between a C2 and C3 UIV when adjusting for baseline characteristics. These results may aid surgeons considering PCLF for degenerative cervical spinal disease.