目的 比较2018年中国指南与2019年欧洲心脏病学会（european society of cardiology， ESC）指南预后评分在预测急性肺栓塞（acute pulmonary embolism， APE）患者30 d全因死亡效能中的价值。 方法 回顾性收集2015年1月至2019年12月贵州省人民医院确诊APE住院患者资料，按照30 d死亡与否分为死亡组及生存组，并根据性别、血氧饱和度及感染情况进行亚组分析。使用SPSS软件绘制两个评分受试者工作特征曲线（ROC）并计算曲线下面积（AUC），应用Delong′s检验比较AUC差异，使用R软件survival包、survIDINRI包、PredictABEL包计算净重分类指数（NRI）及综合判别改善指数（IDI）。 结果 共纳入626例APE患者进行分析，使用两个评分对APE患者30 d死亡进行预测，结果显示，在总体区分度方面，2018年中国指南预后评分较2019年ESC指南预后评分表现较好，AUC分别为0.782、0.749，但差异无统计学意义（P > 0.05）；预测准确性方面，2019年ESC指南预后评分较2018年中国指南预后评分总体NRI提升44.4%（95%CI：0.091 ~ 0.753），正确重分类至死亡组NRI提升58.6%（95%CI：0.161 ~ 0.917），正确重分类至生存组NRI降低14.2%（95%CI：-0.249 ~ 0.08），IDI提高3.38%（P < 0.05）。亚组分析示2018年中国指南预后评分及2019年ESC指南预后评分对不同性别、不同氧饱和度患者均有预测能力（均P < 0.05）、合并感染人群AUC值分别为0.749、0.772（P < 0.05）、不合并感染人群AUC值分别为0.652、0.833（P > 0.05）。 结论 2018年中国指南预后评分、2019年ESC指南预后评分均可预测APE患者30 d死亡率，且对于合并感染的患者具有更好的预测能力；但前者对生存组预测准确性更高，且评分较为便捷，更方便临床应用，后者对死亡组预测能力有所改善。

Objective To compare the value of the 2018 Chinese guideline prognostic score with that of the 2019 European Society of Cardiology （ESC） in the predicting efficiency for acute pulmonary embolism （APE） in 30?day all?cause mortality. Methods The data of the hospitalized patients with confirmed APE from January 2015 to December 2019 were retrospectively collected. According to death within 30 days， the patients were divided into a death group and a survival group. Subgroup analysis was performed according to gender， oxygen saturation and infection. The SPSS software was used to establish the receiver operating characteristic curve （ROC） for the two scores and calculated the area under the curve （AUC）. The Delong′s test was applied to compare the AUC differences. The net reclassification index （NRI） and integrated discrimination improvement （IDI） were calculated using the R software packages of survival， survIDINRI， and PredictABEL. Results 626 APE patients were enrolled， and 30?day death was predicted in those patients using two scores. In terms of overall discrimination， the 2018 Chinese guideline prognostic score was better than the 2019 ESC guideline prognostic score， with an AUC of 0.782 and 0.749， respectively； but there were no statistical differences between the two AUC （P > 0.05）. In terms of prediction accuracy， the NRI of the 2019 ESC guideline prognostic score was 44.4% （95%CI：0.091 ~ 0.753）， higher than that of the 2018 Chinese guidelines prognostic score， which increased by 58.6% （95%CI：0.161 ~ 0.917） in the correct reclassification to death group， while decreased by 14.2% （95% CI：?0.249 ~ 0.08） in the correct reclassification to survival group. IDI increased by 3.38% （P < 0.05）. Subgroup analysis showed the prognostic scores of the 2018 Chinese guidelines and the 2019 ESC guidelines prognostic scores had predictive ability for patients with different gender and different oxygen saturation （P < 0.05）， and the prognostic scores for co?infected population （AUC： 0.749， 0.772） （P > 0.05）， non?coinfected population （AUC： 0.652， 0.833）. Conclusions Both the 2018 Chinese guideline prognostic score and the 2019 ESC guideline prognostic score can predict 30?day mortality in APE patients， and have a better predictive ability for the co?infected population. However， the predictive accuracy of the former is higher than that of the latter in the survival group， and the score is more rapid and convenient for clinical application， while the latter has improved the prediction ability in the death group。