Zhou Qian, Xu Ying, Chen Jun, Gu Qin, Kong Wentao
Department of Ultrasound, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, China.
Intensive Care Unit, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, China.
Quant Imaging Med Surg. 2025 Sep 1;15(9):7697-7710. doi: 10.21037/qims-2025-167. Epub 2025 Aug 18.
Postextubation distress in critically ill patients with successful spontaneous breathing trials (SBTs) is unfavorable for prognosis. This study aimed to determine whether the combined application of multimodal ultrasound parameters of the heart, lung, and diaphragm can predict the mechanical ventilation weaning outcome among critically ill patients.
From December 2022 to December 2023, a total of 74 patients (aged over 18 years old) mechanically ventilated for more than 48 hours and prepared for an SBT were selected from the Department of Critical Care Unit, Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School. Patients who met the criteria for weaning were prepared to undergo a 30-minute SBT, during which the heart, lungs, and diaphragm of the patients were examined via ultrasound. According to the weaning outcomes, patients were divided into a success group and a failure group. Ultrasound parameters with statistical significance in univariate analysis were incorporated into a binary logistic regression model to identify the independent influencing factors of weaning outcomes. The receiver operating characteristic (ROC) curve was plotted, and the area under the curve (AUC) was calculated for statistical analysis.
Out of the 74 included patients, 23 failed weaning. There were significant differences in diaphragm excursion (DE), diaphragm thickening fraction (DTF), bilateral lung ultrasound score (LUS), total LUS, and left ventricular ejection fraction (LVEF) between the success and failure groups (P<0.05). The predictive performance of individual ultrasound parameters was as follows: LVEF achieved an AUC of 0.742 [95% confidence interval (CI): 0.642-0.842; P<0.001], with optimal thresholds of 58.05% (sensitivity: 76.5%; specificity: 69.6%); LUS achieved an AUC of 0.837 (95% CI: 0.738-0.936; P<0.001), with an 80.4% sensitivity and 82.6% specificity at a cutoff value of 17.50. DE yielded an AUC of 0.895 (95% CI: 0.821-0.969; P<0.001), with an 82.4% sensitivity and a 95.7% specificity at a cutoff value of 1.205 cm; DTF reached an AUC of 0.896 (95% CI: 0.827-0.965; P<0.001), with a 68.6% sensitivity and a 100% specificity at a cutoff value of 22.75%. A composite model integrating LVEF, LUS, DE, and DTF achieved an AUC of 0.951 (95% CI: 0.907-0.996; P<0.001), with an 88.2% sensitivity and a 95.7% specificity.
Ultrasound parameters of the heart, lungs, and diaphragm provide critical information on cardiopulmonary and diaphragmatic function during SBT. Weaning failure is more common when LUS is increased and LVEF, DE, and DTF are decreased. The combination of these three aspects can improve the accuracy of predicting weaning outcomes.
成功进行自主呼吸试验（SBT）的重症患者拔管后出现不适对预后不利。本研究旨在确定心脏、肺和膈肌的多模态超声参数联合应用能否预测重症患者机械通气撤机结局。
2022年12月至2023年12月，从南京大学医学院附属鼓楼医院重症监护病房选取74例机械通气超过48小时且准备进行SBT的患者（年龄超过18岁）。符合撤机标准的患者准备进行30分钟的SBT，在此期间通过超声检查患者的心脏、肺和膈肌。根据撤机结局，将患者分为成功组和失败组。将单因素分析中有统计学意义的超声参数纳入二元逻辑回归模型，以确定撤机结局的独立影响因素。绘制受试者工作特征（ROC）曲线，并计算曲线下面积（AUC）进行统计分析。
74例纳入患者中，23例撤机失败。成功组与失败组在膈肌移动度（DE）、膈肌增厚分数（DTF）、双侧肺超声评分（LUS）、总LUS及左心室射血分数（LVEF）方面存在显著差异（P<0.05）。各超声参数的预测性能如下：LVEF的AUC为0.742[95%置信区间（CI）：0.642 - 0.842；P<0.001]，最佳阈值为58.05%（灵敏度：76.5%；特异度：69.6%）；LUS的AUC为0.837（95%CI：0.738 - 0.936；P<0.001），截断值为17.50时，灵敏度为80.4%，特异度为82.6%；DE的AUC为0.895（95%CI：0.821 - 0.969；P<0.001），截断值为1.205 cm时，灵敏度为82.4%，特异度为95.7%；DTF的AUC为0.896（95%CI：0.827 - 0.965；P<0.001），截断值为22.75%时，灵敏度为68.6%，特异度为100%。整合LVEF、LUS、DE和DTF的复合模型AUC为0.951（95%CI：0.907 - 0.996；P<0.001），灵敏度为88.2%，特异度为95.7%。
心脏、肺和膈肌的超声参数为SBT期间的心肺及膈肌功能提供了关键信息。当LUS升高而LVEF、DE和DTF降低时，撤机失败更为常见。这三个方面的联合应用可提高预测撤机结局的准确性。
