炎症指标联合阴道pH诊断先兆流产的价值

Diagnostic Value of Inflammatory Indices Combined with Vaginal pH for Threatened Miscarriage

摘要: 目的：探讨中性粒细胞与淋巴细胞比值(NLR)、血小板与淋巴细胞比值(PLR)、淋巴细胞与单核细胞比值(LMR)联合阴道pH鉴别正常妊娠与先兆流产的价值。方法：纳入63例妊娠女性，其中正常妊娠组32例、先兆流产组31例。比较两组一般资料、阴道pH及炎症指标，采用二元Logistic回归构建炎症指标联合模型和阴道pH联合炎症指标模型，以ROC曲线评价辅助诊断效能，并采用2000次Bootstrap重抽样进行内部验证，同时将年龄、孕周和分娩次数纳入全联合模型进行敏感性校正。结果：先兆流产组NLR升高、LMR降低(P < 0.01)，PLR及阴道pH组间差异无统计学意义(P > 0.05)。阴道pH、NLR、PLR、LMR、炎症指标联合模型及阴道pH联合炎症指标模型的表观AUC分别为0.622、0.805、0.635、0.727、0.863和0.953；全联合模型敏感度为93.5%，特异度为90.6%，Bootstrap平均乐观值为0.018，校正AUC为0.935；年龄、孕周和分娩次数校正模型的表观AUC为0.966，Bootstrap校正AUC为0.922。结论：NLR升高和LMR降低与先兆流产具有一定关联；“系统性炎症/免疫状态 + 局部阴道微环境”的联合模型在本样本中表现出较好的区分度，但结果仍属小样本探索性证据，需在更大样本及独立队列中进一步验证。

Abstract: Objective: To evaluate the auxiliary diagnostic value of the neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), lymphocyte-to-monocyte ratio (LMR), and vaginal pH for distinguishing threatened miscarriage from normal pregnancy. Methods: Sixty-three pregnant women were included, comprising 32 women with normal pregnancy and 31 with threatened miscarriage. General characteristics, vaginal pH, and the inflammatory markers were compared between groups. Binary logistic regression was used to construct an inflammatory-index model and a model combining vaginal pH with the inflammatory indices. Receiver operating characteristic (ROC) curves were used to assess discrimination, with 2000 bootstrap resamples for internal validation; age, gestational age, and parity were additionally included in a sensitivity-adjusted model. Results: NLR was higher and LMR was lower in the threatened miscarriage group (P < 0.01), whereas PLR and vaginal pH did not differ significantly between groups (P > 0.05). The apparent AUCs for vaginal pH, NLR, PLR, LMR, the inflammatory-index model, and the vaginal pH plus inflammatory-index model were 0.622, 0.805, 0.635, 0.727, 0.863, and 0.953, respectively. The full combined model had a sensitivity of 93.5% and specificity of 90.6%, with a mean Bootstrap optimism of 0.018 and an optimism-corrected AUC of 0.935. The age-, gestational-age-, and parity-adjusted model had an apparent AUC of 0.966 and a Bootstrap-corrected AUC of 0.922. Conclusion: Elevated NLR and reduced LMR were associated with threatened miscarriage. The “systemic inflammation/immune status + local vaginal microenvironment” combination showed good discrimination in this sample, but the findings remain exploratory and require validation in larger and independent cohorts.