文献速递关节外科创伤骨科

总淋巴细胞计数无法预测全髋或全膝关节置换术后急性假体周围感染

Total Lymphocyte Count Fails as a Predictor of Acute Periprosthetic Joint Infection for Total Hip or Knee Arthroplasty.
PubMed 文献 · J Arthroplasty · 2026-04-01

本研究基于美国全国医疗数据库,纳入2016至2023年间54,019例术前1至28天有总淋巴细胞计数(TLC)的初次全髋或全膝关节置换成人患者,评估TLC作为连续变量与90天假体周围感染(PJI)风险的关系。限制性立方样条显示阈值1,450 cells/μL但置信区间跨越校正比值比1,ROC曲线下面积0.495,提示预测能力不优于随机。结论认为TLC对PJI无预测效用,不建议用于术前风险分层。

要点

  • TLC阈值1,450 cells/μL的置信区间跨越校正OR=1,无统计学意义。
  • ROC曲线下面积0.495,TLC预测PJI能力不优于随机。
  • 不建议使用TLC对初次全髋或全膝关节置换患者进行术前风险分层。

核心结论

  • 总淋巴细胞计数对初次全髋和全膝关节置换术后PJI无预测效用,不建议用于术前风险分层。

研究设计与样本

  • 利用覆盖美国约25%住院手术的全国医疗数据库,识别2016至2023年间术前1至28天有TLC值的成人初次全膝关节置换(TKA)和全髋关节置换(THA)患者,共54,019例(TKA 64.1%,THA 35.9%),其中0.58%发生PJI。
  • 采用逻辑回归生成限制性立方样条评估TLC与90天PJI风险的关系,并构建ROC曲线及Youden指数敏感性分析。

主要发现

  • 限制性立方样条识别阈值为1,450 cells/μL,但置信区间跨越校正比值比1。
  • ROC模型曲线下面积为0.495,预测能力不优于随机。
  • 在TLC 1,450 cells/μL时,敏感性和特异性分别为70.9%和26.5%。
  • Youden分析得出PJI阈值为3,580 cells/μL,J统计量为0.027。
  • 按手术类型亚组分析显示相似趋势。

关键数值与推荐

  • TLC低于1,500 cells/μL常被用作营养不良标志,但本研究未支持其作为PJI预测指标。
  • 限制性立方样条阈值1,450 cells/μL,置信区间跨越校正OR=1。
  • ROC曲线下面积0.495,敏感性和特异性分别为70.9%和26.5%(TLC 1,450 cells/μL时)。
  • Youden分析阈值3,580 cells/μL,J统计量0.027。

临床怎么用

  • 不应使用TLC对初次THA或TKA患者进行PJI风险分层。
  • 需寻找其他有效指标或综合评估方法来预测PJI风险。

局限性

  • 研究基于行政数据库,可能受编码准确性影响。
  • 仅纳入术前1至28天有TLC值的患者,存在选择偏倚可能。
  • PJI发生率较低(0.58%),可能影响统计效能。

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

BACKGROUND: Malnutrition is associated with an increased risk of periprosthetic joint infection (PJI) following total joint arthroplasty (TJA). Total lymphocyte count (TLC) has been employed as a measure of nutritional status, with a value of less than 1,500 cells/μL often used as a marker of malnutrition. This study sought to evaluate TLC as a continuous variable to determine a threshold below which the risk of PJI increases. The secondary aim was to quantify the predictive utility of TLC for the risk of PJI. METHODS: Utilizing a national health care database that captures approximately 25% of inpatient procedures performed in the United States, we identified all adult patients undergoing total knee arthroplasty (TKA) and total hip arthroplasty (THA) between 2016 and 2023 who had a TLC value obtained one and 28 days, preoperatively. Restricted cubic splines were generated using logistic regression to assess the relationship between TLC and 90-day risk of PJI among the TJA, TKA, and THA cohorts. A receiver operating characteristic curve was created to further analyze TLC's predictive utility. A Youden index analysis was subsequently performed as a sensitivity analysis. In total, 54,019 patients (TKA: 64.1%; THA: 35.9%) were identified with preoperative TLC values, of whom 0.58% developed a PJI. RESULTS: The restricted cubic splines identified a threshold at 1,450 cells/μL, with confidence intervals spanning above and below an adjusted odds ratio of 1. The area under the curve of the receiver operating characteristic model was 0.495, indicating predictive power no better than chance. At a TLC of 1,450 cells/μL, sensitivity and specificity were 70.9 and 26.5%, respectively. Youden analysis for PJI yielded a threshold of 3,580 cells/μL, with an associated J statistic of 0.027. Procedure-specific subanalyses demonstrated a similar trend. CONCLUSIONS: Total lymphocyte count has no predictive utility for PJI following primary THA and TKA. We recommend against using TLC to risk-stratify patients prior to THA or TKA.

来源:J Arthroplasty,查看原文。版权归原作者所有。

引用

  1. Telang SS, Lim MA, Culler MW, et al. Total Lymphocyte Count Fails as a Predictor of Acute Periprosthetic Joint Infection for Total Hip or Knee Arthroplasty. J Arthroplasty. 2026. PMID: 42000215. doi:10.1016/j.arth.2026.03.047

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。