A multivariate prognostic model for pain and activity limitation in people undergoing lumbar discectomy

Purpose: The purpose of this study was to identify a multivariate predictive model for 6-month outcomes on overall pain, leg pain and activity limitation in patients undergoing lumbar discectomy. Identification of predictors of outcome for lumbar discectomy has the potential to assist identifying treatment targets, clinical decision making and disease understanding. Materials and methods: Prospective cohort design. Ninety-seven patients deemed by study surgeons to be suitable for lumbar discectomy completed a comprehensive clinical and radiological baseline assessment. At 6-months post surgery outcome measures of overall and leg pain (visual analogue scale) as well as activity limitation (Oswestry Disability Index) were completed. Univariate and multivariate analyses were conducted to determine the best multivariate predictive model of outcome. Results: In the multivariate model, presence of a compensation claim, longer duration of injury and presence of below knee pain and/or parasthesia were negative prognostic indicators for at least two of the outcomes. Peripheralization in response to mechanical loading strategies was a positive prognostic indicator for overall pain and leg pain. A range of other prognostic indicators for one outcome were also identified. The prognostic model explained up to 32% of the variance in outcome. Conclusions: An 11-factor prognostic model was identified from a range of clinically and radiologically assessed variables in accordance with a biopsychosocial model. The multivariate model has potential implications for researchers and practitioners in the field. Further high quality research is required to externally validate the prognostic model, evaluate effect of the identified prognostic factors on treatment effectiveness and explore potential mechanisms of effect.

研究目的：本研究旨在为接受腰椎间盘切除术（lumbar discectomy）的患者构建用于预测其6个月总体疼痛、腿部疼痛及活动受限结局的多变量模型。明确腰椎间盘切除术结局的预测因素，有助于确定治疗靶点、辅助临床决策及深化疾病认知。
材料与方法：采用前瞻性队列研究设计。97名经本研究手术医师评估适宜接受腰椎间盘切除术的患者，完成了全面的临床与影像学基线评估。术后6个月时，受试者完成了总体疼痛、腿部疼痛（采用视觉模拟评分法（Visual Analogue Scale, VAS））以及活动受限（采用Oswestry功能障碍指数（Oswestry Disability Index, ODI））的结局评估。通过单变量与多变量分析，确定最优的结局多变量预测模型。
结果：多变量模型显示，存在工伤保险索赔、损伤病程更长、以及出现膝下疼痛和/或感觉异常，为至少两项结局的不良预后因素。机械负荷策略诱导的疼痛外周化现象为总体疼痛与腿部疼痛的良好预后因素。此外还识别出多项针对单一结局的预后指标。该预后模型可解释高达32%的结局变异度。
结论：本研究依据生物-心理-社会医学模式（biopsychosocial model），从一系列临床与影像学评估变量中筛选得到由11项指标构成的预后模型。该多变量模型对该领域的研究人员与临床从业者均具有潜在应用价值。未来仍需开展高质量研究，对该预后模型进行外部验证，评估已识别的预后因素对治疗效果的影响，并探索潜在的作用机制。