Charité - Universitätsmedizin Berlin, Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Center for Musculoskeletal Surgery (CMSC), Berlin, Germany; Federal Center of Traumatology, Orthopedics and Arthroplasty, Barnaul, Russia; Berlin-Brandenburg School for Regenerative Therapies (BSRT), Charité - Universitätsmedizin Berlin, Berlin, Germany.
Charité - Universitätsmedizin Berlin, Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Center for Musculoskeletal Surgery (CMSC), Berlin, Germany.
J Arthroplasty. 2020 Aug;35(8):2223-2229.e2. doi: 10.1016/j.arth.2020.03.016. Epub 2020 Mar 16.
Synovial fluid d-lactate may be useful for diagnosing periprosthetic joint infection (PJI) as this biomarker is exclusively produced by bacteria. We evaluated the performance of synovial fluid d-lactate using 2 definition criteria and determined its optimal cutoff value for diagnosing PJI.
Consecutive patients undergoing joint aspiration before prosthesis revision were prospectively included. Synovial fluid was collected for culture, leukocyte count, and d-lactate concentration (by spectrophotometry). Youden's J statistic was used for determining optimal d-lactate cutoff value on the receiver operating characteristic curve by maximizing sensitivity and specificity.
A total of 224 patients were included. Using Musculoskeletal Infection Society criteria, 71 patients (32%) were diagnosed with PJI and 153 (68%) with aseptic failure (AF), whereas using institutional criteria, 92 patients (41%) were diagnosed with PJI and 132 (59%) with AF. The optimal cutoff of synovial fluid d-lactate to differentiate PJI from AF was 1.3 mmol/L, independent of the used definition criteria. Synovial fluid d-lactate had a sensitivity of 94.3% (95% confidence interval [95% CI], 86.2-98.4) and specificity of 78.4% (95% CI, 66.8-81.2) using Musculoskeletal Infection Society criteria, whereas its sensitivity was 92.4% (95% CI, 84.9-96.9) and specificity 88.6% (95% CI, 81.9-93.5) using institutional criteria. The concentration of d-lactate was higher in infections caused by Staphylococcus aureus (P < .001) and streptococci (P = .016) than by coagulase-negative staphylococci or in culture-negative PJI.
The synovial fluid d-lactate showed high sensitivity (>90%) for diagnosis of PJI using both definition criteria and correlated with the pathogen virulence. The high sensitivity makes this biomarker useful as a point-of-care screening test for PJI.
Diagnostic level I.
滑液中的 d-乳酸可用于诊断假体周围关节感染（PJI），因为这种生物标志物仅由细菌产生。我们使用 2 种定义标准评估了滑液 d-乳酸的性能，并确定了其用于诊断 PJI 的最佳截断值。
前瞻性纳入了连续因假体翻修而行关节抽吸术的患者。收集滑膜液进行培养、白细胞计数和 d-乳酸浓度（分光光度法）检测。采用约登指数确定最佳 d-乳酸截断值，通过最大化灵敏度和特异性来优化受试者工作特征曲线。
共纳入 224 例患者。采用肌肉骨骼感染学会标准，71 例（32%）患者被诊断为 PJI，153 例（68%）患者为无菌性失败（AF）；采用机构标准，92 例（41%）患者被诊断为 PJI，132 例（59%）患者为 AF。区分 PJI 与 AF 的滑液 d-乳酸最佳截断值为 1.3 mmol/L，与使用的定义标准无关。采用肌肉骨骼感染学会标准，滑液 d-乳酸诊断 PJI 的灵敏度为 94.3%（95%置信区间[95%CI]，86.2%-98.4%），特异性为 78.4%（95%CI，66.8%-81.2%）；采用机构标准，灵敏度为 92.4%（95%CI，84.9%-96.9%），特异性为 88.6%（95%CI，81.9%-93.5%）。金黄色葡萄球菌（P<0.001）和链球菌（P=0.016）引起的感染的 d-乳酸浓度高于凝固酶阴性葡萄球菌或培养阴性的 PJI。
滑液 d-乳酸的诊断 PJI 的灵敏度均>90%（两种定义标准），与病原体毒力相关。高灵敏度使该生物标志物成为 PJI 的一种有用的即时检测筛查试验。
诊断水平 I。
