基于临床指征分析人工气道分泌物聚集

Clinical-Indicator-Based Assessment of Secretion Accumulation in Artificial Airways

摘要: 目的：有创机械通气的重症肺炎患者是围手术期呼吸道管理的重点关注人群。此类患者气道分泌物清除时机尚缺乏统一的判断标准。针对此问题，本研究旨在分析气道峰压、血氧饱和度、呼吸音特征与气道分泌物聚集量的相关性，明确气道清除时机的最佳判断指标。方法；本研究在武汉协和医院综合ICU开展单中心前瞻性研究，纳入10例因重症肺炎接受有创机械通气且需要频繁吸痰的患者作为研究对象。记录每次吸痰操作前的呼吸音类型、气道峰压升高差值、血氧饱和度降低差值以及实际吸出痰液量，并采用皮尔逊或斯皮尔曼相关分析及限制性立方样条评估各临床指征与痰液量的相关性。结果：本研究共收集50次气道吸痰事件的数据。单次吸出痰液量范围为1.10~8.90 g，中位数3.35 g (IQR 2.60 g)，平均值3.71 g ± 1.84 g。呼吸音特征与痰液量呈显著高度正相关(r = 0.840, P < 0.0001)，不同类型的异常呼吸音能够直观反映气道分泌物的聚集程度。气道峰压升高差值与痰液量亦呈显著较强的正相关(r = 0.774, P < 0.0001)，但两者之间存在显著的非线性关系。血氧饱和度降低差值与痰液量的相关性相对较弱(r = 0.423, P = 0.0022)。结论：呼吸音特征可作为判断人工气道分泌物清除时机的重要指征。通过实时监测呼吸音变化，有望实现对痰液清除时机的及时预警。

Abstract: Objective: Patients with severe pneumonia receiving invasive mechanical ventilation are a priority population for perioperative respiratory management. Standardized criteria for determining the optimal timing of airway secretion clearance are lacking. This study evaluated the associations of peak inspiratory pressure (PIP), peripheral oxygen saturation (SpO₂), and breath-sound characteristics with sputum mass, aiming to identify practical indicators for suction timing. Methods: A prospective observational study was conducted, including 10 patients with severe pneumonia who were on invasive mechanical ventilation. The type of respiratory sounds, the increase in airway peak pressure, the decrease in blood oxygen saturation, and the amount of sputum were recorded before each suctioning procedure. Pearson/Spearman correlation analysis and restricted cubic spline plots were used to evaluate the associations between each indicator and sputum mass. Results: Data from 50 suctioning events were analyzed. Sputum mass per event ranged from 1.10 to 8.90 g (median 3.35 g; IQR 2.60 g; mean 3.71 ± 1.84 g). Breath-sound characteristics showed the strongest positive correlation with sputum mass (r = 0.840, P < 0.0001), with abnormal breath sounds reflecting greater secretion burden. ΔPIP was also strongly and positively correlated with sputum mass (r = 0.774, P < 0.0001) and demonstrated a significant nonlinear relationship. By contrast, ΔSpO₂ exhibited a weaker correlation (r = 0.423, P = 0.0022). Conclusion: Respiratory sound characteristics are good indicators for determining the optimal timing of artificial airway secretion clearance. By real-time monitoring of changes in respiratory sound characteristics, it is possible to achieve precise early warning of the optimal timing for airway secretion clearance.