Departments of Orthopaedic Surgery (J.J.W, C.W.P.G.v.d.W., and P.D.S.D.), Radiotherapy (Y.M.v.d.L.), and Medical Statistics and Bioinformatics (M.F.), Leiden University Medical Centre, Leiden, the Netherlands.
Department of Orthopaedic Surgery, University Medical Centre Groningen, Groningen, the Netherlands.
J Bone Joint Surg Am. 2018 Feb 7;100(3):196-204. doi: 10.2106/JBJS.16.01514.
A survival estimation for patients with symptomatic long bone metastases (LBM) is crucial to prevent overtreatment and undertreatment. This study analyzed prognostic factors for overall survival and developed a simple, easy-to-use prognostic model.
A multicenter retrospective study of 1,520 patients treated for symptomatic LBM between 2000 and 2013 at the radiation therapy and/or orthopaedic departments was performed. Primary tumors were categorized into 3 clinical profiles (favorable, moderate, or unfavorable) according to an existing classification system. Associations between prognostic variables and overall survival were investigated using the Kaplan-Meier method and multivariate Cox regression models. The discriminatory ability of the developed model was assessed with the Harrell C-statistic. The observed and expected survival for each survival category were compared on the basis of an external cohort.
Median overall survival was 7.4 months (95% confidence interval [CI], 6.7 to 8.1 months). On the basis of the independent prognostic factors, namely the clinical profile, Karnofsky Performance Score, and presence of visceral and/or brain metastases, 12 prognostic categories were created. The Harrell C-statistic was 0.70. A flowchart was developed to easily stratify patients. Using cutoff points for clinical decision-making, the 12 categories were narrowed down to 4 categories with clinical consequences. Median survival was 21.9 months (95% CI, 18.7 to 25.1 months), 10.5 months (95% CI, 7.9 to 13.1 months), 4.6 months (95% CI, 3.9 to 5.3 months), and 2.2 months (95% CI, 1.8 to 2.6 months) for the 4 categories.
This study presents a model to easily stratify patients with symptomatic LBM according to their expected survival. The simplicity and clarity of the model facilitate and encourage its use in the routine care of patients with LBM, to provide the most appropriate treatment for each individual patient.
Prognostic Level IV. See Instructions for Authors for a complete description of levels of evidence.
对于有症状的长骨转移（LBM）患者，生存估计对于避免过度治疗和治疗不足至关重要。本研究分析了总生存的预后因素，并开发了一个简单易用的预后模型。
对 2000 年至 2013 年在放射治疗和/或骨科部门接受治疗的 1520 例有症状的 LBM 患者进行了多中心回顾性研究。根据现有的分类系统，将原发性肿瘤分为 3 种临床特征（有利、中等或不利）。采用 Kaplan-Meier 法和多变量 Cox 回归模型分析预后变量与总生存的关系。采用 Harrell C 统计量评估所开发模型的判别能力。根据外部队列比较每个生存组的观察和预期生存。
中位总生存期为 7.4 个月（95%置信区间[CI]：6.7 至 8.1 个月）。根据独立的预后因素，即临床特征、卡诺夫斯基绩效评分以及是否存在内脏和/或脑转移，创建了 12 个预后类别。Harrell C 统计量为 0.70。制定了流程图以方便分层患者。使用用于临床决策的截断值，将 12 个类别缩小为 4 个具有临床意义的类别。4 个类别的中位生存期分别为 21.9 个月（95%CI：18.7 至 25.1 个月）、10.5 个月（95%CI：7.9 至 13.1 个月）、4.6 个月（95%CI：3.9 至 5.3 个月）和 2.2 个月（95%CI：1.8 至 2.6 个月）。
本研究提出了一种模型，可以根据预期生存轻松对有症状的 LBM 患者进行分层。该模型的简单性和清晰度有助于并鼓励其在 LBM 患者的常规护理中使用，为每个患者提供最合适的治疗。
预后 IV 级。有关证据水平的完整说明，请参见作者说明。
