A Clinical Study On The Ability Of Integrative Weaning Index To Predict Weaning From Mechanical Ventilation

Objective: To determine the value of integrative weaning indexmeasured before extubation as a predictor of weaning success for patients whowere on mechanical ventilation for more than72hours.Methods: The study was conducted from September2012to February2013in the intensive care unit of the fourth hospital of Hebei MedicalUniversity. Fifty patients who with failure respiration mechanically-ventilatedfor more than72hours were randomly divided into2groups:①FiO₂=0.40group.②FiO₂=0.35group. Daily weaning screening test was taken andventilator support condition was gradually reduced.The oxygen concentrationof the two groups decreased to0.40and0.35respectively. A30minutes SBTwas taken with T-piece after the criteria of weaning screening test was meted.The mechanical-ventilation time of the two groups in their respective oxygenconcentration was more than30minutes. Record Age, Sexuality, the cause ofrespiratory failure, durationof mechanical ventilation, Success or failure ofweaning, APACHE II, blood pressure （Bp）, heart rate （HR）, respiratoryfrequency （f）, conscious state in the weaning process. Static compliance of therespiratory system （Cst,rs）, tracheal airway occlusion pressure（P0.1）, maximuminspiratory pressure （MIP）, P0.1/MIP, expiratory tidal volume （Vt）, intrinsicpositive end expiratory pressure （PEEPi）, arterial oxygen saturation （SaO₂）and other parameters of the arterial blood gas analysis, and the rapid shallowbreathing index （RSBI） in the first minutes of SBT were recored in all patients.We calculate IWI according to the formula IWI=Cst,rs×SaO₂/RSBI. Thevalue of IWI, RSBI, Cst,rs, P0.1, MIP, P0.1/MIP, SaO₂were evaluated bycalculating the area under the receiver operator characteristic （ROC） curvesinthe two groups and all patients. The area under the ROC curves （AUC） of IWIwas compared with other paraments. We also observed the difference of each parameter between patients who were successful in weaning and who failed ineach group. Statistical analysis was performed with SPSS13.0softwarepackage. The ROC curves were calculated and compared with Medcalc11.4software. Values were presented as mean±SD unless otherwise specified. Thecomparison of two-samples was calculated with Student’s t-tests. The meansof each group were analyzed with ANOVA. The two-sample rate comparisonwas calculated by chi-square test. A significant statistics was p-value<0.05.Results:150patients included34male and16female patients. The age was37-84years old and the average age was66.38±10.492years.44patientswere after surgery operations and6patients were not.32patients （64%） weresuccessfully liberated from mechanical ventilation while18patients failed.217patients （68%） were successfully liberated from mechanicalventilation while8patients failed in FiO₂=0.40group.15patients （68%） weresuccessfully liberated from mechanical ventilation while10patients failed inFiO₂=0.35group. There was no significant difference between the two groupsof the average of age, ratio of male and female, APACHE II, duration ofventilation, ratio of successful weaning （p>0.05）. The average of IWI, RSBI,Cst,rs, MIP, P0.1, P0.1/MIP were not significantly different between the twogroups （p>0.05）. There was significant difference between the two groups ofthe average of SaO₂（FiO₂=0.40,96.73±1.65VS FiO₂=0.35,95.47±2.23p=0.027）.3IWI>50.40ml/cmH2Obreaths/minute/liter in group FiO₂of0.40, as athreshold of predicted successful weaning, Sensitivity is82.35%, Specificity is87.50%. The AUC of IWI was0.846±0.117,95%confidence interval （CI）was0.646-0.958. There was significantly different comparing the AUC of IWIwith the AUC of MIP （0.846±0.117VS0.526±0.145, p=0.009）. There wasno significant difference in comparing the AUC of IWI with the AUC of RSBI,Cst,rs, P0.1, P0.1/MIP and SaO₂（p>0.05）. There was significant difference ofthe average of IWI, RSBI, P0.1between patients who were successful inweaning and who failed（p<0.05）while the average of Cst,rs, MIP, P0.1/MIP, SaO₂are not（p>0.05）.4IWI>39.33ml/cmH2Obreaths/minute/liter in group FiO₂of0.35, as athreshold of predicted successful weaning, Sensitivity is93.33%, Specificity is90%. The AUC of IWI was0.953±0.395,95%CI was0.7876-0.998. Therewas significantly different comparing the AUC of IWI with the AUC of P0.1,MIP, P0.1/MIP, SaO（2p<0.05）. There was no significant difference comparingthe AUC of IWI with the AUC of RSBI, Cst,rs （p>0.05）. There wassignificant difference of the average of IWI, RSBI, Cst,rs between patientswho successful weaning with who was failure （p<0.05） while the average ofP0.1, MIP, P0.1/MIP, SaO₂are not （p>0.05）.5IWI＞45.70ml/cmH2Obreaths/minute/liter in comprehensive analysisof all50patients, as a threshold of predicting successful weaning, Sensitivityis90.62%, Specificity is83.33%. The AUC of IWI was0.908±0.055,95%CIwas0.792-0.971. There was no significant different compare the AUC of IWIwith the AUC of RSBI （0.908±0.055VS0.821±0.0720, p=0.096）. Therewas significant different compare the AUC of IWI with the AUC of Cst,rs, P0.1,MIP, P0.1/MIP, SaO₂（p<0.05）. There was significant different of the averageof IWI, RSBI, Cst, P0.1, SaO₂between patients who were successful inweaning with those who failed（p<0.05）while the average of MIP, P0.1/MIPare not （p>0.05）.Conclusions:Except RSBI, Our results showed that IWI can be helpful to improve itsweaning predictive value when compared with other traditional weaningindexes As a predictor for weaning, the ability of IWI in group FiO₂=0.40waspoor. Adjusting the concentration of oxygen inhalation to0.35before SBTmay make IWI have better performance.