Alvarado Sánchez Jorge Iván, Caicedo Ruiz Juan Daniel, Diaztagle Fernández Juan José, Amaya Zuñiga William Fernando, Ospina-Tascón Gustavo Adolfo, Cruz Martínez Luis Eduardo
Department of Anaesthesiology, Centro Policlínico del Olaya, Bogotá, Colombia.
Department of Physiology Sciences, Faculty of Medicine, Universidad Nacional de Colombia, Bogotá, Colombia.
Ann Intensive Care. 2021 Feb 8;11(1):28. doi: 10.1186/s13613-021-00817-5.
Dynamic predictors of fluid responsiveness have shown good performance in mechanically ventilated patients at tidal volumes (Vt) > 8 mL kg. Nevertheless, most critically ill conditions demand lower Vt. We sought to evaluate the operative performance of several predictors of fluid responsiveness at Vt ≤ 8 mL kg by using meta-regression and subgroup analyses.
A sensitive search was conducted in the Embase and MEDLINE databases. We searched for studies prospectively assessing the operative performance of pulse pressure variation (PPV), stroke volume variation (SVV), end-expiratory occlusion test (EEOT), passive leg raising (PLR), inferior vena cava respiratory variability (Δ-IVC), mini-fluid challenge (m-FC), and tidal volume challenge (VtC), to predict fluid responsiveness in adult patients mechanically ventilated at Vt ≤ 8 ml kg, without respiratory effort and arrhythmias, published between 1999 and 2020. Operative performance was assessed using hierarchical and bivariate analyses, while subgroup analysis was used to evaluate variations in their operative performance and sources of heterogeneity. A sensitivity analysis based on the methodological quality of the studies included (QUADAS-2) was also performed.
A total of 33 studies involving 1,352 patients were included for analysis. Areas under the curve (AUC) values for predictors of fluid responsiveness were: for PPV = 0.82, Δ-IVC = 0.86, SVV = 0.90, m-FC = 0.84, PLR = 0.84, EEOT = 0.92, and VtC = 0.92. According to subgroup analyses, variations in methods to measure cardiac output and in turn, to classify patients as responders or non-responders significantly influence the performance of PPV and SVV (p < 0.05). Operative performance of PPV was also significantly affected by the compliance of the respiratory system (p = 0.05), while type of patient (p < 0.01) and thresholds used to determine responsiveness significantly affected the predictability of SVV (p = 0.05). Similarly, volume of fluids infused to determine variation in cardiac output, significantly affected the performance of SVV (p = 0.01) and PLR (p < 0.01). Sensitivity analysis showed no variations in operative performance of PPV (p = 0.39), SVV (p = 0.23) and EEOT (p = 0.15).
Most predictors of fluid responsiveness reliably predict the response of cardiac output to volume expansion in adult patients mechanically ventilated at tidal volumes ≤ 8 ml kg. Nevertheless, technical and clinical variables might clearly influence on their operative performance.
对于潮气量（Vt）>8ml/kg的机械通气患者，液体反应性的动态预测指标已显示出良好的性能。然而，大多数危重症情况需要更低的Vt。我们试图通过Meta回归和亚组分析来评估Vt≤8ml/kg时几种液体反应性预测指标的操作性能。
在Embase和MEDLINE数据库中进行了全面检索。我们检索了前瞻性评估脉压变异（PPV）、每搏量变异（SVV）、呼气末阻断试验（EEOT）、被动抬腿试验（PLR）、下腔静脉呼吸变异度（Δ-IVC）、迷你液体负荷试验（m-FC）和潮气量负荷试验（VtC）等预测指标操作性能的研究，这些研究旨在预测Vt≤8ml/kg、无呼吸努力和心律失常的成年机械通气患者的液体反应性，发表时间为1999年至2020年。使用分层分析和双变量分析评估操作性能，同时采用亚组分析来评估其操作性能的差异和异质性来源。还基于纳入研究的方法学质量（QUADAS-2）进行了敏感性分析。
共纳入33项研究，涉及1352例患者进行分析。液体反应性预测指标的曲线下面积（AUC）值分别为：PPV = 0.82，Δ-IVC = 0.86，SVV = 0.90，m-FC = 0.84，PLR = 0.84，EEOT = 0.92，VtC = 0.92。根据亚组分析，测量心输出量的方法差异以及将患者分类为反应者或无反应者的差异，显著影响PPV和SVV的性能（p<0.05）。PPV的操作性能也受到呼吸系统顺应性的显著影响（p = 0.05），而患者类型（p<0.01）和用于确定反应性的阈值显著影响SVV的预测性（p = 0.05）。同样，用于确定心输出量变化的输液量显著影响SVV（p = 0.01）和PLR（p<0.01）的性能。敏感性分析显示PPV（p = 0.39）、SVV（p = 0.23）和EEOT（p = 0.15）的操作性能无差异。
大多数液体反应性预测指标能够可靠地预测潮气量≤8ml/kg的成年机械通气患者心输出量对容量扩张的反应。然而技术和临床变量可能会明显影响其操作性能。
