No abstract

Background: There is significant variability in clinically important improvement (CII) criteria for spinal surgery that suggest population-and diagnosis-specific thresholds are required to determine surgical success using patient-reported outcome measures (PROMs). This study establishes surgical CII thresholds for 4 common lumbar degenerative spinal diagnoses using accepted anchor-based methodology and commonly used PROMs. Methods: CII analysis was conducted using baseline and 1-year data from participants in the Canadian Spine Outcomes and Research Network (CSORN) registry who underwent surgery for lumbar spinal stenosis (LSS), degenerative spondylolisthesis (DS), disc herniation (DH) or degenerative disc (DD) from 2015 to 2018. One-year CII thresholds were determined for the Oswestry Disability Index (ODI), and back and leg Numeric Pain Rating Scales (NPRS). At 1 year, patients reported whether they were much better, better, the same, worse or much worse compared with before their surgery. This was used as the anchor (improved: ≥ "better" v. not improved: ≤ "same") to determine CII thresholds for absolute change and percentage change for PROMs using a receiver operating characteristic (ROC) curve approach, with maximization of the Youden index as primary criterion. Correct classification rates were determined. Results: There were 856 participants with LSS (39.1% female, mean age 65.8 yr), 591 with DS (64.1% female, mean age 65.8 yr), 520 with DH (47.5% female, mean age 46.8 yr) and 185 with DD (43.8% female, mean age 50.9 yr). CII for ODI change ranged from -10.0 (DD) to -16.9 (DH). CII for back and leg NPRS change was -2 to -3 for each group. CII for percentage change varied by PROM and pathology group, ranging from -11.1% (ODI for DD) to -50.0% (leg NPRS for DH). Correct classification rates for all CII thresholds ranged from 72.1% to 89.4%. Conclusion: This work quantifies Canadian CII thresholds for the ODI and back and leg NPRS for 4 common lumbar spinal surgery cohorts, with high classification accuracy. Our results suggest that use of generic CII across different diagnoses in spine surgery is not advised. This study establishes the first comprehensive set of responder criteria in Canada for broader application and specificity in clinical and research settings and for surgical prognostic work.

The aim of this experimental study was to assess the biomechanical performance of a novel C1 posterior arch (C1PA) clamp compared with C1 lateral mass (C1LM) screws in constructs used to treat atlantoaxial instability. These constructs had either C2 pedicle (C2P) screws or C2 translaminar (C2TL) screws. Eight fresh-frozen human cadaveric ligamentous spine specimens (C0-C3) were tested under six conditions: the intact state, the destabilized state after a simulated odontoid fracture, and when instrumented with four constructs (C1LM-C2P, C1LM-C2TL, C1PA-C2P, C1PA-C2TL). Each specimen was tested in a spinal loading simulator that separately applied axial rotation, flexion-extension and lateral bending. In each test condition, displacement controlled angular motion was applied in both directions at a speed of 2 deg/s until a resulting moment of 1.5 Nm was achieved. The measured ranges of motion (ROM) of the C1-C2 segments were compared for each test condition using nonparametric Friedman tests. The destabilized state had significantly more C1-C2 motion ( p < 0.05) than the intact state in all cases, and all constructs greatly reduced this motion. C2 pedicle screw constructs that used the C1PA clamp had significantly less C1-C2 motion ( p < 0.05) than those with C1LM screws in flexion-extension as well as axial rotation and no statistically significant difference was detected in lateral bending. C2 translaminar screw constructs that used the C1PA clamp had significantly less C1-C2 motion ( p < 0.05) than those with C1LM screws in flexion-extension and no statistically significant difference was detected in axial rotation or in lateral bending. Data from the current study suggested that constructs using the novel C1PA clamp would provide as good, or improved, biomechanical stability to the C1-C2 segment compared with constructs using C1LM screws.