甲状腺术后不同时间检测甲状旁腺激素预测低钙血症发生的系统评价

目的探讨甲状腺术后不同时间段检测甲状旁腺激素（PTH）对术后低钙血症的预测价值。方法检索中国生物医学文献库、万方数据库、同方数据库及维普中国科技期刊数据库的中文文献，同时检索MEDLINE、OVID、PUBMED、EMBASE等外文数据库，纳入以甲状腺术后发生低钙血症为参考标准，对比不同时间段检测的PTH值预测术后低钙血症的价值。根据检测时间不同分为2组，并分别对纳入文献进行质量评价及偏倚分析，提取相关信息后通过STATA12及Meta-Disc软件进行合并分析，计算合并的敏感度（SEN）、特异度（SPE）、阳性似然比（LR＋）、阴性似然比（LR-）及SROC曲线下面积（AUC），并对2组指标进行对比分析。结果共纳入23篇文献，其中前瞻性队列研究15篇，回顾性研究8篇；英文文献21篇，中文文献2篇。按不同时间段分为2组，第1组（术后1h内检测PTH值）共纳入患者2012例，其中术后发生低钙血症患者494例，未发生低钙血症患者1518例；第2组（术后4-12h内检测PTH值）共纳入患者693例，其中术后发生低钙血症患者266例，未发生低钙血症患者427例。通过纳入文献漏斗图发现2组文献的发表偏倚均较小。Meta分析结果显示，除合并的SEN外，2组间合并的SPE、朋＋、LR-及AUC的差异均有统计学意义（P〈0．01）；第1组的预测效果好于第2组，术后1h内检测的PTH值〈16ng／L时其AUC面积最大。结论甲状腺术后检测PTH值是一种有效预测术后发生低钙血症的方法，其中术后1h内检测PTH值低于16ng／L对术后低钙血症的预测效果最佳。

Systematic Review of Predicting Hypocalcemia by Detecting Parathyroid Hormone at Different Time after Thyroi-dectomy

Objective To evaluate the value of parathyroid hormone （PTH） in predicting hypocalcemia at different time after thyroidectomy. Methods The literatures in CBM, WanFang, CNKI, VIP in Chinese, and OVID, PUBMED, EMBASE, and MEDLINE in English were searched. Hand searches and additional searches were also conducted. The studies of predicting hypocalcemia after thyroidectomy by detecting postoperative PTH at different time were selected, and the quality and tested the heterogeneity of included articles were assessed. Then the proper effect model to calculate pooled weighted sensitivity （SEN）, specificity （SPE）, positive likelihood ratio （LR ＋）, and negative likelihood ratio （LR-） were selected. The summary receiver operating characteristic （SROC） curve was performed and the area under the curve （AUC） was computed. Results Twenty-three articles entered this systematic review, 21 articles were English and 2 articles were Chinese. Fifteen of 23 articles were designed to be prospective cohort study （PC） and 8 of 23 articles were retrospective study （Retro）. These articles were divided into two groups. Group 1 was the studies of detecting postoper- ative PTH in 1 hour, which included 2 012 cases （494 of them occurred hypocalcemia）. Group 2 was the studies of detec- ting postoperative PTH between 4-12 hours, which included 693 cases （266 of them occurred hypocalcemia）. Thepublication bias of 2 groups were smaller that founded through the literature funnel. Meta analysis showed that in addition to merge SEN, between the 2 groups with merge SPE, LR＋, LR-, and AUC differences were statistically significant （P 〈 0. 01） ; the forecast effect of group 1 was better than group 2, and the A UC was the largest area when the PTH value in 1 hour after operation was below 16 ng/L. Conclusion Detection of postoperative PTH value is an effective method for predicting postoperative hypocalcemia. The 1 hour after operation for detecting PTH value below 16 ng/L to predict postoperative hypocalcemia have the best effect.