摘要目的 对比每搏连续无创血压监测系统(CNAP)与脉搏指示连续心排血量监测(PiCCO)技术测定机械通气重症患者血流动力学参数的差异及一致性,评估CNAP无创监测血流动力学的可行性.方法采用前瞻性观察性自身对照研究方法,选择福建省立医院重症医学四科2018年6月至2019年3月收治的机械通气且需行血流动力学监测的重症患者,在入科后立刻置入PiCCO导管,先采用热稀释法监测血流动力学参数,分别在0 h和24 h获得平均动脉压(MAPPiCCO)、心排血指数(CIPiCCO)、脉压变异度(PPVPiCCO)及外周血管阻力指数(SVRIPiCCO),然后再用CNAP监测获取相同参数(MAPCNAP、CICNAP、PPVCNAP及SVRICNAP),两种方法均连续测量3次取平均值;比较两种方法监测各血流动力学参数的差异.采用Pearson检验分析两种方法获取血流动力学参数的相关性;采用Bland-Altman法分析指标的一致性.结果 共有38例患者入选,排除24 h内死亡1例和自动出院2例、心房颤动2例、数据采集丢失1例,最终纳入32例患者.32例患者中女性12例,男性20例;年龄26～84岁,平均(66.8±19.1)岁;体重指数(23.7±3.9)kg/m2 ;急性生理学与慢性健康状况评分Ⅱ(APACHEⅡ)为(19.5±5.3)分;序贯器官衰竭评分(SOFA)为(9.7±4.1)分.两种方法监测结果显示, CICNAP与CIPiCCO、PPVCNAP与PPVPiCCO差异无统计学意义〔CI(mL·s-1·m-2):59.8±12.6比58.5±14.2,PPV :(14.7±6.8)%比(14.0±6.8)%,均P＞0.05〕;MAPCNAP、SVRICNAP均明显高于MAPPiCCO、SVRIPiCCO〔MAP(mmHg, 1 mmHg＝0.133 kPa):65.6±9.4比60.1±9.2,SVRI(kPa·s·L-1·m-2):206.2±53.9比179.5±57.8,均P＜0.01〕.相关性分析显示,两种方法测得的MAP、CI、PPV及SVRI均呈显著正相关(r值分别为0.624、0.864、0.835及0.655,均P＜0.05).一致性分析显示,CICNAP与CIPiCCO、PPVCNAP与PPVPiCCO的一致性较好,平均差值分别为1.2 mL·s-1·m-2、0.5%,95%可信区间(95%CI)分别为-12.8～15.3 mL·s-1·m-2、-7.1%～8.2% ;MAPCNAP与MAPPiCCO、SVRICNAP与SVRIPiCCO的一致性较差,平均差值分别为5.5 mmHg、26.8 kPa·s·L-1·m-2,95%CI分别为-10.4～21.3 mmHg、-64.5～118.0 kPa·s·L-1·m-2.结论 监测机械通气重症患者的CI和PPV时,CNAP与PiCCO可以相互替代;而两种方法监测的MAP、SVRI可能存在差异,需正确谨慎解读.

abstractsObjective To evaluate the difference and correlation between continuous non-invasive arterial pressure (CNAP) monitor and pulse indicated continuous cardiac output (PiCCO) monitor on determination of hemodynamic parameters in mechanically ventilated critically ill patients, and to assess the feasibility of non-invasive monitoring of hemodynamics with CNAP. Methods A prospective observation self-control study was conducted. The critically ill patients with mechanical ventilation who needed hemodynamics monitoring, and admitted to the fourth department of intensive care unit (ICU) of Fujian Provincial Hospital from June 2018 to March 2019 were enrolled. PiCCO catheter were inserted immediately after admission, the hemodynamic indexes were measured by thermodilution method, and mean arterial pressure (MAPPiCCO), cardiac index (CIPiCCO), pulse pressure variation rate (PPVPiCCO) and systemic vascular resistance index (SVRIPiCCO) were obtained at 0 hour and 24 hours respectively. Meanwhile, the above indexes (MAPCNAP, CICNAP, PPVCNAP and SVRICNAP) were measured with CNAP. All measurements were repeated thrice and average values were reported. The differences in above parameters between the two methods were evaluated. Pearson test was used for the correlation analysis and Bland-Altman analysis method was used for consistency test. Results Thirty-eight patients were enrolled into this study. One patient died within 24 hours was excluded, 2 patients were excluded due to withdrawing treatment within 24 hours, 2 patients were excluded because of atrial fibrillation, and 1 patient's data was lost due to technical problems. Thus, data from 32 patients were available for final analysis. There were 12 females and 20 males, aging 26-84 years old with the mean of (66.8±19.1) years old, body mass index (BMI) of (23.7±3.9) kg/m2, acute physiology and chronic health evaluationⅡ (APACHEⅡ) score of 19.5±5.3, sepsis-related organ failure assessment (SOFA) score of 9.7±4.1. There were no significant differences in CI or PPV between CNAP and PiCCO groups [CI (mL·s-1·m-2): 59.8±12.6 vs. 58.5±14.2, PPV: (14.7±6.8)% vs. (14.0±6.8)%, both P > 0.05]. MAP and SVRI measured by CNAP were significantly higher than those measured by PiCCO [MAP (mmHg, 1 mmHg = 0.133 kPa):65.6±9.4 vs. 60.1±9.2, SVRI (kPa·s·L-1·m-2): 206.2±53.9 vs. 179.5±57.8, both P < 0.01]. The correlation analysis showed that MAP, CI, PPV and SVRI measured by the two methods were significantly positively correlated (r value was 0.624, 0.864, 0.835 and 0.655 respectively, all P < 0.05). Bland-Altman analysis showed that CNAP and PiCCO had a good consistency for the measurement of CI and PPV, the average differences were 1.2 mL·s-1·m-2 and 0.5% respectively, while the 95% confidence interval (95%CI) were -12.8-15.3 mL·s-1·m-2 and -7.1%-8.2% respectively. However, the consistency of MAP and SVRI measured by those two methods was poor, the average differences were 5.5 mmHg and 26.8 kPa·s·L-1·m-2 respectively, while the 95%CI was -10.4-21.3 mmHg and -64.5-118.0 kPa·s·L-1·m-2 respectively. Conclusion CNAP was comparable with PiCCO when monitoring CI and PPV in mechanically ventilated critically ill patients; while the results of MAP and SVRI might be inaccurate, which should be interpreted correctly and carefully.