Perioperative Risk Factors And Risk Evaluation Models For Prolonged Mechanical Ventilation And Intensive Care Unit Stay After Aortic Arch Surgery

Objective:The purpose of this study was to identify risk factors and risk evaluation models for prolonged mechanical ventilation and intensive care unit (ICU) stay in patients underwent aortic arch surgery from a single institution.Methods:All consecutive patients underwent aortic arch surgeries requiring deep hypothermic circulatory arrest plus antegrade selective cerebral perfusion between Jan 2005 and Dec 2007 were reviewed retrospectively. Those patients underwent one-stage total or subtotal aortic replacement were excluded. Prolonged mechanical ventilation was defined as postoperative mechanical ventilation time longer than 48 h, and prolonged ICU stay was defined as length of stay in ICU longer than 5 d postoperatively. Univariate analysis and multivariate logistic regression were used to analyze the association between perioperative variables and prolonged mechanical ventilation and ICU stay.Results:Two hundred and ninety-eight patients were enrolled in this study, including 236 (79.2%) total aortic arch replacement surgeries and 62 (20.8%) partial aortic arch replacement surgeries, with 270 (90.6%) ascending aortic replacement surgeries. Patients were on average 44.9±10.7 years of age, and male patients were younger than female ones (44.1±10.1 yr vs.47.5±11.8 yr, p< 0.05). Two hundred and eighty-six patients (96.0%) suffered from acute or chronic aortic dissection, and only 12 patients (4.0%) suffered from aortic arch aneurysm. Emergent surgeries, which account for 28.2% of total surgeries, were all patients with aortic dissection. Postoperative mechanical ventilation time in the cohort was 15.0 (11.5,20.1) h,38 patients (12.8%) required prolonged mechanical ventilation. Postoperative duration of ICU stay was 66.9 (41.4, 113.0)h,61 patients(20.5%)stayed longer than 5 d in the ICU. Univariate analysis found age, body mass index, history of hypertension, New York Heart Association (NYHA) classification, preoperative serum creatinine, creatinine clearance, emergency, myocardial ischemic time, cardiopulmonary bypass (CPB) time, coronary artery bypass grafting. Patients requiring prolonged mechanical ventilation had higher incidences of in-hospital mortality, stroke, renal failure requiring dialysis, and reintubation and stayed longer in the ICU and the hospital (p< 0.05). Independent risk factors for prolonged mechanical ventilation were found to be age, CPB time, body mass index, NYHA classification, serum creatinine, and emergency. The evaluation model for prolonged mechanical ventilation after aortic arch surgery was P= 1/(1+exp (-x)), in which P refers to the possibility of prolonged mechanical ventilation, and x=-13.11+(0.048×age)+(0.024×CPB time)+(0.11×body mass index)+(0.008×serum creatinine) +(1.02×0 (NYHA classificationⅠorⅡ) or 1 (NYHA classification III or IV))+ (0.96×0 (selective operation) or 1 (emergency)). Hosmer-Lemeshow goodness-of-fit test (p> 0.05) and area under receiver operating characteristic curve (C statistic= 0.84) suggested good calibration and discrimination of the model. The odds ratio of prolonged mechanical ventilation unadjusted and adjusted appears to increase consecutively with serum creatinine and after values of 50 years with age. Univariate analysis found age, body mass index, history of hypertension, NYHA classification, preoperative serum creatinine, creatinine clearance, emergency, inotropes, myocardial ischemic time, CPB time, intraoperative fresh frozen plasma transfusion. Patients requiring prolonged ICU stay had higher incidences of in-hospital mortality, stroke, renal failure requiring dialysis, reintubation, required longer mechanical ventilation, and stayed longer in the hospital (p < 0.05). Independent risk factors for prolonged ICU stay were found to be age, serum creatinine, inotropes, CPB time. The evaluation model for prolonged ICU stay after aortic arch surgery was P= 1/(1+exp (-x)), in which P refers to the possibility of prolonged ICU stay, and x=-9.41+(0.042×age)+(0.0014×serum creatinine)+(0.018×CPB time)+(1.048×0 (without inotropes) or 1 (inotropes)). Hosmer-Lemeshow goodness-of-fit test (p> 0.05) and area under receiver operating characteristic curve (C statistic= 0.78) suggested good calibration and discrimination of the model. The odds ratio of prolonged ICU stay unadjusted and adjusted appears to increase consecutively with serum creatinine and after values of 220 min with CPB time.Conclusions:The incidences of prolonged mechanical ventilation and ICU stay were high after aortic arch surgeries. Patients with prolonged mechanical ventilation and ICU stay could be predicted by multivariate predictors, which is helpful to distribute limited medical resources. Objective:To construct an amperometric method for real time and quantitative measurement of catecholamine in heart in vivo, and investigate the physiology and pathophysiology of evoked catecholamine release from cardiac sympathetic nerve terminals.Methods:Evoked release of catecholamine from cardiac sympathetic nerve in rabbit heart in vivo was recorded by electrochemical amperometry using gold fiber micro-electrodes. The catecholamine release was recorded following electrical stimuli to cardiac sympathetic nerve, hypoxia and myocardial ischemia-reperfusion. We observed the kinetics of catecholamine release in rabbit heart, modulation of cardiac sympathetic nerve by vagus, effect of hypoxia on catecholamine release induced by sympathetic stimulation and effect of myocardial ischemia-reperfusion induced catecholamine release.Results:The catecholamine release from cardiac sympathetic nerve induced by electrical stimuli was not frequency-dependent. The readily releasable pool of synaptic vesicles could be depleted by electrical stimuli with 30 Hz,3 mA and 600 pulses to cardiac sympathetic nerve. The time constant of readily releasable pool recovery was 2.5 min. Electrical stimulation to vagus partly inhibited the catecholamine release induced by stimulation to cardiac sympathetic nerve, but the inhibition effect disappeared in 20 min. The current amplitude induced by cardiac sympathetic nerve stimulation immediately after hypoxia increased by 50%, however, the enhancement effect of hypoxia disappeared in 10 min. Transient myocardial ischemia and reperfusion both induced intramyocardial catecholamine release, but a longer ischemia did not induce a consistent increase of catecholamine concentration.Conclusion:The amperometric method with gold fiber electrode is effective for real time and quantitative measurement of catecholamine in heart in vivo. We firstly investigated the kinetics of catecholamine release in rabbit heart. Inhibition effect of vagus and enhancement effect of hypoxia to cardiac sympathetic activity were verified with amperometric method. Both of these two effects were time limited. Reperfusion, besides myocardial ischemia, also can induce catecholamine release from cardiac sympathetic nerve. The catecholamine release might play an important role in myocardial injury.