摘要目的:探讨压力反射敏感性（baroreflex sensitivity，BRS）对美托洛尔治疗儿童体位性心动过速综合征（postural orthostatic tachycardia syndrome，POTS）的疗效预测作用。方法:采用单中心回顾性病例分析方法，选择2013年3月1日至2024年3月31日在北京大学第一医院儿童医学中心住院并接受美托洛尔治疗的POTS患儿作为研究对象，将其分为训练集（2013年3月1日至 2022年12月31日）和验证集（2023年1月1日至2024年3月31日），记录所有患儿的基线资料。采用单因素分析及Logistic回归分析确定与美托洛尔疗效相关的因素。应用受试者工作特征曲线、校准曲线和临床决策曲线来进行评估，并在验证集中进行验证。结果:共纳入101例POTS患儿，其中训练集71例，验证集30 例。训练集中，Logistic回归分析发现，平卧位 BRS 与美托洛尔治疗的疗效显著相关（ OR=2.107，95% CI：1.234～3.596， P=0.006）。以平卧位BRS ≥ 8.045 ms/mmHg为界值，预测美托洛尔治疗儿童 POTS 有效的灵敏度为82.1%，特异度为81.4%。校准曲线和临床决策曲线提示预测值和实际值拟合较好，可取得好的临床净收益。在验证集中，预测概率混淆矩阵显示，预测准确率是93.3%，灵敏度是92.8%，特异度是93.7%。 结论:以基线平卧位BRS ≥ 8.045 ms/mmHg为界值的BRS评估可用于预测美托洛尔治疗儿童POTS的有效性。

abstractsObjective:To investigate whether baroreflex sensitivity（BRS）can predict the therapeutic response to metoprolol in children with postural orthostatic tachycardia syndrome（POTS）.Methods:A single-center retrospective case analysis method was conducted. This study enrolled pediatric patients diagnosed with POTS who received metoprolol at the Children's Medical Center，Peking University First Hospital between March 1st，2013 to March 31st，2024，and then divided these patients into training set（from March 1st，2013 to December 31st，2022）and validation set（from January 1st，2023 to March 31st，2024）. Baseline characteristics were collected for all participants. Univariate analysis followed by binary Logistic regression was used to identify predictors of metoprolol response. Model performance was assessed by receiver operating characteristic curves，calibration plots，and decision curve analysis，and subsequently confirmed in the validation cohort.Results:A total of 101 pediatric POTS patients were enrolled，including 71 patients in the training set and 30 patients in the validation set. In the training set，binary logistic regression identified supine BRS as an independent predictor of metoprolol treatment response（ OR=2.107，95% CI：1.234-3.596， P=0.006）. A supine BRS ≥ 8.045 ms/mmHg predicted response with 82.1% sensitivity and 81.4% specificity. The calibration plot and decision curve analysis suggested that the predicted values fit well with the actual values，which could achieve good clinical net benefits. In the validation set，the prediction probability confusion matrix showed that the model achieved 93.3% overall accuracy，92.8% sensitivity，and 93.7% specificity. Conclusion:Baroreflex sensitivity assessment with baseline supine BRS ≥ 8.045 ms/mmHg could be used to predict metoprolol responsiveness in children with POTS.