TABLE IV

Important OPLL Predictors of Surgical Outcome: Results of Univariate and Multivariate Analyses*

StudyStudy Design (Level of Evidence)Summary of AnalysisUnivariate AnalysisMultivariate Analysis
Chen et al.21 (2014)Retrospective cohort study (III)Multiple regression analysis including age, duration of symptoms, smoking status, diabetes, preoperative JOA score, levels of corpectomies, occupying ratio, dural ossification, and signal intensity on T2-weighted MRI
  • Postoperative JOA Score

  • • Occupying ratio (<50%, 13.84 ± 1.84; ≥50%, 13.52 ± 1.92): p = 0.345

  • • Dural ossification (yes, 13.57 ± 2.18; no, 13.76 ± 1.39): p = 0.534

  • Postoperative JOA Score

  • • Occupying ratio: p = 0.428

  • • Dural ossification: p = 0.746

Choi et al.22 (2005)Retrospective cohort study (II)Multivariate logistic regression including snake-eye appearance, diabetes, duration of symptoms, preoperative Nurick grade, and occupying ratio
  • Nurick Outcome

  • • Occupying ratio: p = 0.4222

  • • Type of OPLL: p = 0.9531

  • • Double-layer sign: p = 1.0000

  • Nurick Outcome

  • • Occupying ratio (OR, 0.973; 95% CI, 0.85 to 1.07): p = 0.5356

Iwasaki et al.23 (2002)Retrospective cohort study (III)Stepwise multivariate regression analysis including age, preoperative JOA score, type of OPLL, occupying ratio, space available for the spinal cord, postoperative change of cervical alignment, radiographic fusion, and growth of OPLLNot reported
  • Postoperative JOA score

  • • Type of OPLL: p > 0.05

  • • Occupying ratio: p > 0.05

  • • Space available for spinal cord: p > 0.05

Iwasaki et al.7 (2007)Retrospective cohort study (II)Multiple regression analysis including age, preoperative JOA score, occupying ratio, space available for the spinal cord, preoperative alignment, and postoperative change in cervical alignment
  • Postoperative JOA Score

  • • Shape of ossification (hill-shaped, 11.7; plateau-shaped, 14.1): p < 0.01)

  • • Occupying ratio (<60%, 14.1; ≥60%, 11.0): p < 0.03

  • Recovery Rate

  • • Shape of ossification (hill-shaped, 30.0%; plateau-shaped, 59.9%): p < 0.01

  • • Occupying ratio (<60%, 58%; ≥60%, 14%): p < 0.03

  • JOA Score at Time of Maximum Recovery

  • • Shape of ossification: p < 0.0001

  • • Occupying ratio: p > 0.05

  • • Space available for spinal cord: p > 0.05

Kim et al.24 (2015)Retrospective cohort study (III)Multiple linear regression analysis including age, sex, diabetes, smoking status, BMI, shape of ossification, signal intensity grade, preoperative JOA score, occupying ratio, and sagittal alignment
  • Recovery Rate

  • • Type of OPLL (continuous, 87.12%; mixed, 80.01%; segmental, 51.79%; localized, 69.07%): p = 0.154

  • • Shape of lesion (hill-shaped, 74.05%; plateau-shaped, 70.62%): p = 0.751

  • • Occupying ratio: p = 0.196

  • Recovery Rate

  • • Occupying ratio: p = 0.041

  • • Shape of lesion: p = 0.241

Ogawa et al.25 (2005)Retrospective cohort study (III)Multiple regression analysis controlling for age, operative time, blood loss, and duration of symptomsNot reported
  • Final Follow-up (>7 yr)

  • • Patients with segmental OPLL had lower median or change in postoperative JOA and upper-extremity and trunk scores than patients with nonsegmental OPLL

  • • There was no difference in median or change in postoperative lower-extremity and bladder scores

Uchida et al.26 (2005)Retrospective cohort study (III)Multivariate and multiple regression analysisNot reported
  • Postoperative JOA score

  • • Occupying ratio (<40%, ≥40%): PCC = 0.881

  • • Type of OPLL: PCC = 0.555

  • * JOA = Japanese Orthopaedic Association, OR = odds ratio, CI = confidence interval, PCC = Pearson correlation coefficient.