摘要
目的:构建有效的预测模型,以识别慢性血栓栓塞性肺动脉高压(CTEPH)患者球囊肺动脉成形术(BPA)后血液动力学结局的预测因子,从而指导临床实践。方法:回顾性连续纳入2018年5月至2023年6月期间于中国医学科学院阜外医院接受BPA治疗的CTEPH患者99例,根据术后随访结果将平均肺动脉压<30 mmHg(1 mmHg=0.133 kPa)的患者分入良好血液动力学结局组(n=46),其余患者分入不良血液动力学结局组(n=53)。从单因素Logistic回归分析结果中筛选变量纳入多因素Logistic回归分析,通过向后逐步回归法构建预测模型,并建立列线图。采用受试者工作特征(ROC)曲线和校准曲线评估模型的区分度和准确度,采用Bootstrap法(1000次重抽样)进行内部验证。结果:99例患者的中位BPA治疗次数为3(3,5)次。与良好血液动力学结局组相比,不良血液动力学结局组患者病程更长,一氧化碳弥散量占预计值百分比更低,基线和随访时的血液动力学状态均更差(P均<0.05)。多因素Logistic回归分析显示,基线闭塞病变比例(OR=1.05,95%CI:1.02~1.09,P=0.005)、基线肺血管阻力(OR=1.24,95%CI:1.08~1.42,P=0.003)和病程(OR=1.27,95%CI:1.10~1.46,P=0.001)与BPA术后不良血液动力学结局独立相关。由这3个预测因子构建的预测模型对BPA术后血液动力学结局具有最高的预测能力(曲线下面积为0.86,95%CI:0.78~0.92)。校准曲线显示,该模型对BPA术后不良血液动力学结局的预测发生风险与实际发生风险高度一致(校准斜率为1.1)。内部验证进一步证实了该模型的预测能力(原始一致性指数为0.86,校正后一致性指数为0.85)。结论:本研究构建的预测模型可用于BPA术前评估,有助于临床医师早期识别可能需要多次BPA治疗才能获得良好血液动力学结局的患者。
Objectives:This study aims to develop a predictive model to identify predictors of hemodynamic outcomes after balloon pulmonary angioplasty(BPA)in patients with chronic thromboembolic pulmonary hypertension(CTEPH).Methods:A total of 99 consecutive patients diagnosed with CTEPH who underwent BPA between May 2018 and June 2023 were retrospectively enrolled.Patients who achieved a follow-up mean pulmonary arterial pressure<30 mmHg(1 mmHg=0.133 kPa)after BPA were categorized as having a favorable haemodynamic outcome(favorable haemodynamic outcome group,n=46),otherwise were categorized as having an unfavorable haemodynamic outcome(unfavorable haemodynamic outcome group,n=53).Variables selected from univariable logistic regression analysis were incorporated into multivariable logistic regression analysis.A predictive model was constructed using backward stepwise regression,and a nomogram was developed accordingly.The discriminative ability,calibration,and internal validation of the model were evaluated by receiver operating characteristic(ROC)curve analysis,calibration curves,and the bootstrap method(1000 resamples),respectively.Results:Among the 99 patients,a median of 3(3,5)BPA procedures were performed per patient.Compared to patients in favorable haemodynamic outcome group,patients in unfavorable haemodynamic outcome group had a significantly longer disease duration,lower percentage of predicted diffusing capacity of the lung for carbon monoxide,and worse haemodynamics both at baseline and follow-up.Multivariable logistic regression demonstrated that the proportion of occlusive lesions(OR=1.05,95%CI:1.02-1.09,P=0.005),baseline pulmonary vascular resistance(OR=1.24,95%CI:1.08-1.42,P=0.003)and disease duration(OR=1.27,95%CI:1.10-1.46,P=0.001)were independently associated with unfavorable haemodynamic outcome post BPA.The predictive model constructed from these three factors demonstrated the highest predictive capability for hemodynamic outcomes after BPA(area under the ROC curve was 0.86,95%CI:0.78-0.92).The calibration curve exhibited a high degree of concordance between the anticipated and observed incidences of unfavorable haemodynamic outcome after BPA with a slope of 1.1.Internal validation further confirmed the predictive ability of this model(original concordance index was 0.86,bias-adjusted concordance index was 0.85).Conclusions:The prediction model developed in this study can be utilized for pre-BPA assessment,assisting clinicians in early identification of patients who may require multiple BPA procedures to achieve favorable hemodynamic outcomes.
作者
王一佳
李欣
杨涛
曾绮娴
黄志华
段安琪
王齐
高璐阳
李思聪
赵青
罗勤
赵智慧
柳志红
WANG Yijia;LI Xin;YANG Tao;ZENG Qixian;HUANG Zhihua;DUAN Anqi;WANG Qi;GAO Luyang;LI Sicong;ZHAO Qing;LUO Qin;ZHAO Zhihui;LIU Zhihong(Center for Respiratory and Pulmonary Vascular Diseases,National Center for Cardiovascular Diseases and Fuwai Hospital,Chinese Academy of Medical Sciences and Peking Union Medical College,Beijing 100037,China)
出处
《中国循环杂志》
北大核心
2026年第2期174-180,共7页
Chinese Circulation Journal
基金
国家重点研发计划(2023YFC2507203)
北京市自然科学基金(L252080)
中国医学科学院医学与健康科技创新工程(2023-I2M-C&T-B-063)。
