Abstract: Background　Although the main indications of stroke volume variability（SVV）have been identified，the usage of SVV to predict fluid responsiveness in one-lung ventilation has raised some controversy.Objective　To evaluate the predictive value of SVV for fluid responsiveness in patients undergoing one-lung ventilation（OLV）with different tidal volumes in video-assisted thoracic surgery.Methods　A total of 72 patients diagnosed with left lung mass and scheduled for video-assisted thoracoscopic lobectomy or radical resection of pulmonary carcinoma in left lung were enrolled in this study from July 2016 to July 2017 at Jinhua Municipal Central Hospital.The patients were randomly divided into two groups using random sequence numbers generated by SPSS 16.0，each with 36 cases：the high tidal volume group（group A）and the low tidal volume group（group B）．After anesthesia induction，patients received two lung ventilation at first，followed by the initiation of the whole left lung isolation，then ventilated with right lung ventilation.All patients received mechanical ventilation on volume-controlled mode.Ventilation parameters：group A：tidal volume was 8 ml/kg，ventilation rate 12 breaths/min，oxygen concentration 100%，respiratory ratio 1∶2，and positive end-expiratory pressure（PEEP）5 cm H2O（1 cm H2O=0.098 kPa）；group B：the tidal volume was 7 ml/kg，and other parameters were as same as that of the group A.Volume loading test was performed at 30 min after starting OLV and when hemodynamic parameters were stable.General information of patients were collected，including sex，age，body mass index，one-lung ventilation time，operation time and intraoperative bleeding volume.Hemodynamic variables，including heart rate（HR），mean arterial pressure（MAP），central venous pressure（CVP），SVV and cardiac index（CI），were measured before（T1）and 5 min after volume loading（T2）．The changing rate of SVV（ΔSVV）and CI（ΔCI） were calculated.A demonstrated increase of ≥15% in cardiac index（ΔCI）was regarded as positive standard，i.e. effective volume expansion and fluid responsiveness of patients.Receiver operating characteristic（ROC）curves for predicting fluid responsiveness of SVV were generated.Results　In the study，data collected from sixty patients who finished volume loading test were analyzed（30 for group A and 30 for group B）．SVV in group B was lower than that in group A at T1（P<0.05）．HR and SVV in group A were lower at T2 than those at T1，CI in group A was higher at T2 than that at T1（P<0.05）；CVP and CI in group B were higher at T2 than those at T1，and SVV in group B was lower at T2 than that at T1（P<0.05）．ΔSVV was negatively correlated with ΔCI in group A and group B（r=-0.838，-0.730，P<0.001）．The results of ROC curve analysis in group A showed that the area under the ROC curve of SVV for predicting fluid responsiveness was 0.744〔95%CI（0.567，0.921）〕，and the sensitivity and specificity were 57.9% and 90.9%，respectively，when the threshold value was 14.5%.The results of ROC curve analysis in group B showed that the area under the ROC curve of SVV for predicting fluid responsiveness was 0.647〔95%CI（0.444，0.850）〕，and the sensitivity and specificity were 70.6% and 61.5%，respectively，when the threshold value was 11.5%.The area under ROC curve of SVV in group A was larger than that in group B（P<0.05）．Conclusion　We found that SVV could predict fluid responsiveness in patients undergoing OLV with different tidal volume.The threshold value was 14.5% for SVV to predict fluid responsiveness when tidal volume was 8 ml/kg.When tidal volume was 7 ml/kg，the accuracy and usage of SVV became weak.

Key words: Pulmonary ventilation, Thoracoscopes, Stroke volume, Tidal volume, Stroke volume variation, Fluid responsiveness

摘要： 背景　目前，每搏量变异度（SVV）的主要适应证虽已有定论，但SVV是否适用于胸腔镜单肺通气手术仍存在争论。目的　探讨在胸腔镜单肺通气不同潮气量患者中运用SVV判断液体反应性的准确性。方法　选择2016年7月—2017年7月于金华市中心医院心胸外科诊断为左肺肿块择期行胸腔镜下左肺叶切除术或左肺癌根治术的患者72例。根据SPSS 16.0软件产生的随机序列表将其分为A组和B组，各36例。两组均麻醉后先行双肺通气，后行左全肺隔离，之后行右肺通气。两组均采用定容呼吸模式，A组潮气量8 ml/kg，通气频率12次/min，吸呼比1∶2，氧浓度100%，呼气末正压（PEEP）5 cm H2O（1 cm H2O=0.098 kPa）；B组潮气量7 ml/kg，余同A组。两组患者于单肺通气30 min、血流动力学指标稳定时进行容量负荷试验。收集患者一般资料，包括性别、年龄、体质量、单肺通气时间、手术时间及术中出血量。记录进行容量负荷试验前（T1）、完成容量负荷试验后5 min（T2）时血流动力学参数〔心率（HR）、平均动脉压（MAP）、中心静脉压（CVP）、SVV、心脏指数（CI）〕。计算SVV变化率（ΔSVV）和CI变化率（ΔCI）。以ΔCI≥15%为容量负荷试验阳性标准，即患者扩容有效、有液体反应性，绘制SVV预测液体反应性的ROC曲线。结果　本研究最后完成容量负荷试验纳入统计分析的患者共有60例，其中A组30例、B组30例。B组T1时SVV低于A组（P<0.05）。A组T2时HR、SVV低于T1时，T2时CI高于T1时（P<0.05）；B组T2时CVP、CI高于T1时，T2时SVV低于T1时（P<0.05）。A组、B组ΔSVV与ΔCI呈负相关（r=-0.838、-0.730，P<0.001）。A组ROC曲线分析结果示：SVV预测液体反应性的ROC曲线下面积为0.744〔95%CI（0.567，0.921）〕，截断值为14.5%时，灵敏度为57.9%，特异度为90.9%。B组ROC曲线分析结果示：SVV预测液体反应性的ROC曲线下面积为0.647〔95%CI（0.444，0.850）〕，截断值为11.5%时，灵敏度为70.6%，特异度为61.5%。A组SVV预测液体反应性的ROC曲线下面积大于B组（P<0.05）。结论　SVV对单肺通气不同潮气量患者的液体反应性均有预测价值，其中单肺通气潮气量为8 ml/kg时，SVV可以用于指导液体反应性的判断，截断值为14.5%；单肺通气潮气量为7 ml/kg时，SVV对判断液体反应性有一定的指导作用，但准确性欠佳。

关键词: 肺通气, 胸腔镜, 每搏输出量, 潮气量, 每搏量变异度, 液体反应性