文献速递关节外科

牛津膝评分与牛津髋评分能否识别不需行全膝关节或全髋关节置换术的患者

Can the Oxford Knee and Hip Score identify patients who do not require total knee or hip arthroplasty?
PubMed 文献 · Bone Joint J · 2019-01-01

该回顾性研究分析了单一位外科医生三年内连续初诊的1436膝(1016例患者)和478髋(388例患者)的初诊前牛津膝评分与牛津髋评分(OKHS),以判断其能否预测外科医生首次会诊时判定为不需手术还是需手术。结果显示不需手术组的牛津评分更优(p < 0.001),评分与手术决策相关(p < 0.001)。膝与髋的受试者工作特征曲线下面积分别为0.83(95%CI 0.81至0.85)和0.87(95%CI 0.84至0.91),膝的保守有效阈值为牛津膝评分>32分(敏感度=0.997,阴性预测值=0.992),髋为牛津髋评分>34分(敏感度=0.997,阴性预测值=0.978)。

要点

  • 初诊前牛津膝/髋评分能较好预测单一位外科医生初诊时判定为不需手术的转诊。
  • 膝的保守有效阈值为牛津膝评分>32分(敏感度=0.997,阴性预测值=0.992)。
  • 髋的保守有效阈值为牛津髋评分>34分(敏感度=0.997,阴性预测值=0.978)。

核心结论

  • 初诊前牛津膝评分与牛津髋评分在单一位外科医生的实践中能较好预测初次全膝关节置换或全髋关节置换转诊被判定为不需手术。
  • 多种牛津膝/髋评分阈值可有效筛查出不需手术的转诊。

研究设计与样本

  • 回顾性分析单一位外科医生三年内所有连续初次全膝关节置换和全髋关节置换会诊的初诊前牛津膝评分与牛津髋评分。
  • 共纳入1436膝(1016例患者)和478髋(388例患者),按外科医生首次会诊时是否提供手术分为手术组与不需手术组。
  • 采用Spearman秩相关系数和受试者工作特征曲线分析。

主要发现

  • 不需手术组的牛津评分更优(p < 0.001),且评分与手术决策相关(p < 0.001)。
  • 膝的受试者工作特征曲线下面积为0.83(95%CI 0.81至0.85),髋为0.87(95%CI 0.84至0.91)。
  • 膝的保守有效阈值为牛津膝评分>32分(敏感度=0.997,阴性预测值=0.992)。
  • 髋的保守有效阈值为牛津髋评分>34分(敏感度=0.997,阴性预测值=0.978)。
  • 研究还识别出若干潜在更低的牛津膝/髋评分阈值。

关键数值与推荐

  • 牛津膝评分>32分:敏感度=0.997,阴性预测值=0.992。
  • 牛津髋评分>34分:敏感度=0.997,阴性预测值=0.978。
  • 膝的受试者工作特征曲线下面积=0.83(95%CI 0.81至0.85)。
  • 髋的受试者工作特征曲线下面积=0.87(95%CI 0.84至0.91)。

临床怎么用

  • 原文未据此给出推荐。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

AIMS: The aim of this study was to determine if the Oxford Knee and Hip Score (OKHS) can accurately predict when a primary knee or hip referral is deemed nonsurgical versus surgical by the surgeon during their first consultation, and to identify nonsurgical OKHS screening thresholds. PATIENTS AND METHODS: We retrospectively reviewed pre-consultation OKHS for all consecutive primary total knee arthroplasty (TKA) and total hip arthroplasty (THA) consultations of a single surgeon over three years. The 1436 knees (1016 patients) and 478 hips (388 patients) included were categorized based on the surgeon's decision into those offered surgery during the first consultation versus those not (nonsurgical). Spearman's rank correlation coefficients and receiver operating characteristic (ROC) curve analysis were performed. RESULTS: Oxford Scores were better for the nonsurgical cohorts (p < 0.001) and correlated with the surgical decision (p < 0.001). ROC area under the curve values for knees (0.83, 95% confidence intervals (CI) 0.81 to 0.85) and hips (0.87, 95% CI 0.84 to 0.91) were excellent. A conservative and effective threshold for knees is Oxford Knee Score (OKS) > 32 points (sensitivity = 0.997, negative predictive value (NPV) = 0.992) and for hips is Oxford Hip Score (OHS) > 34 points (sensitivity = 0.997, NPV = 0.978). Severable potential lower OKHS thresholds were identified. CONCLUSION: Pre-consultation OKHS demonstrate good ability to predict when a primary TKA or THA referral will be deemed nonsurgical in a single surgeon's practice. Multiple OKHS thresholds can effectively screen out nonsurgical referrals. Cite this article: Bone Joint J 2019;101-B(6 Supple B):23-30.

来源:Bone Joint J,查看原文。版权归原作者所有。

引用

  1. Neufeld ME, Masri BA. Can the Oxford Knee and Hip Score identify patients who do not require total knee or hip arthroplasty? Bone Joint J. 2019. PMID: 31146569. doi:10.1302/0301-620X.101B6.BJJ-2018-1460.R1

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。