无创正压通气治疗急性呼吸窘迫综合征失败的原因分析

Analysis of the clinical effect of noninvasive positive pressure ventilation in the treatment of acute respiratory distress syndrome

【摘要】目的观察无创正压通气（NPPV）治疗急性呼吸窘迫综合征（ARDS）患者的临床疗效，并分析导致治疗失败的危险因素。方法采用回顾性观察性研究方法，选择2013年1月至2015年12月广东省人民医院急诊重症加强治疗病房（EICU）收治且应用NPPV治疗的ARDS患者，根据治疗效果分为成功组和失败组。记录两组患者一般情况、ARDS分级、治疗前急性生理学与慢性健康状况评分系统Ⅱ（APACHEⅡ）评分、ARDS诱因，以及治疗前和治疗2h后心率（Ha）、呼吸频率（RR）、氧合指数（eaO2/FiO2）、动脉血氧分压（PaO2）、动脉血二氧化碳分压（PaCO2）、动脉血氧饱和度（SaO2）。采用logistic回归分析方法筛选导致NPPV治疗ARDS失败的独立危险因素。结果共人选137例ARDS患者，排除昏迷（6例）、血流动力学不稳定（18例）、严重低氧血症（5例）及数据不完整（5例）者后，最终103例进入统计，NPPV成功组69例、失败组34例。与成功组比较，失败组患者APACHEⅡ评分更高（分：21．4±6．2比19．7±8．9），重度ARDS和肺内源性ARDS的患者比例更大[分别为82．4％（28／34）比5．8％（4／69）、32．4％（11／34）比8．7％（6／69），均P〈0．05]。NPPV治疗前，失败组HR和RR明显高于成功组[HR（次／min）：124±13比117±12，RR（次／min）：39±5比33±4]，PaO2／FiO2、PaO2、PaCO2、SaO2明显低于成功组[PaO]FiO2（mmHg，1mmHg=0．133kPa）：104±10比156±12，PaO2（mmHg）：53±8比68±7，PaCO2（mmHg）：31±5比37±7，SaO2：0．83±0．07比0．91±0．05，均P〈0．05]。Logistic回归分析显示，导致NPPV治疗ARDS失败的独立危险因素包括：重度ARDS[优势比（OR）：10．533，95％可信区间（95％CI）：5．847—89．852，P=0．000]，肺内源性ARDS（OR=4．831，95％CI=1．688～13．825，P=0．003），治疗2h后PaO2/FiO2〈140mmHg（OR=7．049，95％CI=1．266～39．236，P=0．026）。结论NPPV对重度ARDS和肺内源性ARDS患者治疗失败率较高；NPPV治疗2h疗效不明显者应及时切换为有创通气。

Objective To evaluate the clinical efficacy of noninvasive positive pressure ventilation （NPPV） in the treatment of patients with acute respiratory distress syndrome （ARDS）, and to look for the predictors of failure of NPPV. Methods A retrospective observation was conducted. ARDS patients underwent NPPV admitted to emergency intensive care unit （EICU） of Guangdong General Hospital from January 2013 to December 2015 were enrolled. The patients were divided into success group and failure group according to the clinical efficacy. The condition of the patients in the two groups was evaluated, and ARDS classification and acute physiology and chronic health evaluationⅡ （APACHE Ⅱ ） score before treatment were recorded. Etiological composition of ARDS was analyzed. The parameters, including heart rate （HR）, respiratory rate （RR）, oxygenation index （PaO2/FiO2）, arterial partial pressure of oxygen （PaO2）, arterial partial pressure of carbon dioxide （PaCO2） and arterial oxygen saturation （SaO2）, were recorded before and 2 hours after the treatment of NPPV. Multivariate logistic regression analysis was conducted for predicting the independent factors inducing the failure of NPPV treatment of patients with ARDS. Results The date of 137 patients with ARDS were collected, excluding the followed patients, 6 with coma, 18 with hemodynamie instability, 5 with severe hypoxia, and 5 with incomplete date. Finally, a total of 103 patients entered the statistics. There were 69 patients in NPPV success group, and 34 in failure group. Compared with success group, APACHE H score in the failure group was higher （21.4 ± 6.2 vs. 19.7 ± 8.9）, the ratios of patients with severe ARDS and those induced by pulmonary infeetion were higher [82.4% （28/34） vs. 5.8% （4/69）, 32.4% （11/34） vs. 8.7% （6/69）, respeetively, both P 〈 0.05]. HR and RR before NPPV in the failure group were significantly higher than those of success group [HR （bpm）： 124 ± 13 vs. 117± 12, RR （bpm）： 39 ±5 vs. 33 ± 4], and PaO2/FiO2, PaO2, PaCO2, and SaO2 were significantly lower than those of the suecess group [PaOz/FiO2 （mmHg, 1 mmHg = 0.133 kPa）： 104± 10 vs. 156± 12, PaO2 （mmHg）： 53±8 vs. 68±7, PaCO2 （mmHg）： 31 ± 5 vs. 37 ± 7, SaO2：0.83 ± 0.07 vs. 0.91 ± 0.05, all P 〈 0.05]. It was shown by logistic regression analysis that severe ARDS [odds ratio （OR） = 10.533, 95% confidenee interval （95%C/） = 5.847-89.852, P = 0.000], pulmonary infeetion resulted ARDS （OR = 4.831, 95%CI = 1.688-13.825, P = 0.003） and PaO2/FiO2 〈 140 mmHg 2 hours after treatment （OR = 7.049, 95%CI = 1.266-39.236, P = 0.026） were the independent risk factors of NPPV failure for the treatment of patients with ARDS. Conclusions Patients with severe ARDS and pulmonary infection derived ARDS were the risk factors of failure to NPPV in ARDS. Lack of improvement in oxygenation 2 hours after NPPV is the predictor of NPPV failure and change to invasive ventilation.