机械按压与人工按压对心搏骤停患者自主循环恢复和预后影响的因素分析

Influence factors analysis of mechanical compression and hands-only compression on restoration of spontaneous circulation and prognosis in patients with cardiac arrest

目的探讨不同按压方式对心搏骤停患者自主循环恢复(ROSC)及预后的影响因素。方法基于全国急诊心搏骤停救治数据库,收集2015年7月至2017年7月来自7个省14家教学医院收治的517例心搏骤停患者的临床资料。根据患者是否接受机械按压分为机械按压组和人工按压组。收集患者人口学数据、复苏参数〔按压频率、监测仪显示的通气频率、复苏时间、药物使用〕及生理学参数〔呼气末二氧化碳分压(PETCO2)、脉搏血氧饱和度(SpO2)等〕,比较两组患者ROSC率及24 h、7 d、28 d生存率。依据复苏时程是否超过60 min,使用多因素Logistic回归模型分析患者ROSC的影响因素。结果 517例患者中,24例因资料不全而被排除,最终共493例患者纳入分析,其中机械按压组214例,人工按压组279例。与人工按压组比较,机械按压组患者年龄更大,合并慢性阻塞性肺疾病(COPD)的比例更高,非可除颤心律比例更低,复苏中按压频率更低,通气频率更快,PETCO2水平和肾上腺素、碳酸氢钠用量更高,心肺复苏(CPR)时间更长。机械按压组患者ROSC率高于人工按压组〔36.9%(79/214)比30.5%(85/279)〕,但差异无统计学意义〔优势比(OR)=1.10,95%可信区间(95%CI)=0.68~1.76,P=0.693〕,使用多因素Logisitc回归校正混杂因素后,两组患者ROSC率比较差异仍无统计学意义(OR=1.21,95%CI=0.54~1.88,P=0.054)。两组患者24 h、7 d和28 d存活率比较差异也无统计学意义。两组患者不同复苏时程复苏参数和生理学参数比较结果显示,CPR时间<60 min时,机械按压组按压频率较低,通气频率和肾上腺素用量较高;CPR时间≥60 min时,机械通气按压组肾上腺素用量和PETCO2较高。多因素Logistic回归分析显示,对于CPR时间<60 min的患者,非可除颤心律(OR=0.29,95%CI=0.05~0.75,P=0.015)、按压频率>120次/min(OR=0.39,95%CI=0.24~0.64,P<0.001)和通气频率>40次/min(OR=0.50,95%CI=0.31~0.84,P=0.034)是ROSC的独立危险因素;PETCO2≥20 mmHg(1 mmHg=0.133 kPa)是ROSC的保护因素(OR=2.79,95%CI=1.88~4.49,P<0.001)。而对于CPR时间≥60 min的患者,年龄≥65岁(OR=0.33,95%CI=0.15~0.67,P=0.018)、夜间收住(OR=0.74,95%CI=0.59~0.94,P=0.035)、非可除颤心律(OR=0.38,95%CI=0.25~0.65,P=0.001)、非心源性心搏骤停(OR=0.35,95%CI=0.25~0.48,P=0.013)、合并糖尿病(OR=0.46,95%CI=0.27~0.82,P=0.015)是ROSC的独立危险因素;心搏骤停发生在抢救室(OR=2.02,95%CI=1.02~2.92,P=0.023)、机械按压(OR=1.41,95%CI=1.12~1.75,P=0.043)、PETCO2≥20 mmHg(OR=2.94,95%CI=1.34~4.54,P=0.012)以及合并急性冠脉综合征(ACS;OR=2.47,95%CI=1.15~3.78,P=0.043)是ROSC的保护因素。结论对于急诊心搏骤停患者,机械按压与人工按压在ROSC及24 h、7 d、28 d存活率方面差异无统计学意义;对于CPR时间超过60 min的患者,机械按压与患者较高的ROSC率有关。

Objective To evaluate the influence factors of different compression modes on restoration of spontaneous circulation (ROSC) and outcomes in patients with cardiac arrest. Methods Based on the national database of emergency cardiac arrest treatment, the clinical data of 517 patients with cardiac arrest admitted to 14 teaching hospitals in 7 provinces from July 2015 to July 2017 were enrolled. According to the way of compression, the patients were divided into mechanical compression group and hands-only compression group. The demographic data, resuscitation parameters [compression frequency, monitored ventilation frequency, duration of resuscitation, drug usage] and physiological parameters [end-expiratory partial pressure of carbon dioxide (PETCO2), pulse oxygen saturation (SpO2)] were collected. The ROSC rates and 24-hour, 7-day, 28-day survival rates were compared between the two groups. Multivariate Logistic regression model was used to analyze the influencing factors of ROSC according to whether the duration of resuscitation was longer than 60 minutes. Results Of 517 patients, 24 were excluded because of incomplete data. A total of 493 patients were enrolled in the analysis with 214 patients in the mechanical compression group, and 279 in the hands-only compression group. Compared with hands-only compression group, the patients in mechanical compression group had higher age, proportion of chronic obstructive pulmonary disease (COPD) and PETCO2, fewer un-shockable rhythm, less compression rate, more epinephrine and sodium bicarbonate usage, and longer duration of cardiopulmonary resuscitation (CPR). Although the rate of ROSC in the mechanical compression group was higher than that in the hands-only compression group [36.9%(79/214) vs. 30.5%(85/279)], there was no significant difference in the rate of ROSC between the two groups [odds ratio (OR)= 1.10, 95% confidence interval (95%CI)= 0.68-1.76, P = 0.693], even after adjusted for con-variables by multivariate Logistic regression (OR = 1.21, 95%CI = 0.54-1.88, P = 0.054). Furthermore, 24-hour, 7-day, and 28-day survival rate also showed no significant difference in both univariate model and multivariate model. Comparisons of resuscitation parameters and physiological parameters between the two groups showed that when the duration of CPR < 60 minutes, the pressing frequency of the mechanical compression group was lower, ventilation frequency and adrenaline dosage were higher;and when the duration of CPR ≥ 60 minutes, the adrenaline dosage and PETCO2 of the mechanical compression group were higher. Multivariate Logistic regression analysis showed that among patients with a duration of CPR < 60 minutes, un-shockable rhythm (OR = 0.29, 95%CI = 0.05-0.75, P = 0.015), compression rate > 120 times/min (OR = 0.39, 95%CI = 0.24-0.64, P < 0.001), ventilation frequency > 40 times/min (OR = 0.50, 95%CI = 0.31-0.84, P = 0.034) were independent risk factors for ROSC;while PETCO2≥20 mmHg (1 mmHg = 0.133 kPa) was protective factor for ROSC (OR = 2.79, 95%CI = 1.88-4.49, P < 0.001). However, for patients with CPR duration ≥ 60 minutes,≥ 65 years old (OR = 0.33, 95%CI = 0.15-0.67, P = 0.018), admission at night (OR = 0.74, 95%CI = 0.59-0.94, P = 0.035), un-shockable rhythm (OR = 0.38, 95%CI = 0.25-0.65, P = 0.001), non-cardiogenic cardiac arrest (OR = 0.35, 95%CI = 0.25-0.48, P = 0.013), previous history of diabetes mellitus (OR = 0.46, 95%CI = 0.27-0.82, P = 0.015) were independent risk factors for ROSC, and cardiac arrest occurred in emergency room (OR = 2.02, 95%CI = 1.02-2.92, P = 0.023), mechanical compression (OR = 1.41, 95%CI = 1.12-1.75, P = 0.043), PETCO2≥ 20 mmHg (OR = 2.94, 95%CI = 1.34-4.54, P = 0.012), previous history of acute coronary syndrome (ACS;OR = 2.47, 95%CI = 1.15-3.78, P = 0.043) were protective factors for ROSC. Conclusions Mechanical compression CPR had no significant differences in the rate of ROSC and 24-hour, 7-day, 28-day survival rates for cardiac arrest patients in the emergency departments compared with hands-only compression CPR. For those who undergone CPR duration more than 60 minutes, mechanical compression was associated with a higher rate of ROSC.