Factors influencing ventilator weaning and predictive value of weaning criteria in children / 中华儿科杂志

ABSTRACT

<p><b>OBJECTIVE</b>To analyze the factors that influence the outcome of ventilator weaning and the predictive value of ventilator weaning criteria such as PaO2/FiO2, Vt/kg, Cst, RSBI and the changing trend of RSBI [&Delta;RSBI = (RSBI2-RSBI1)/RSBI1, RSBI1 and RSBI2 are the start and end of SBT separately] in children.</p><p><b>METHOD</b>One hundred and thirty-two children who were mechanically ventilated over 24 hours were enrolled from April 2012 to October 2013 in Shenzhen Children's Hospital. SIMV+PSV and SBT ventilator mode were used in ventilator weaning according to clinical experience. Age, gender, weight, critical illness score, duration of mechanical ventilation, PICU stays and the causes to mechanical ventilation were taken into account as the effect on the result of ventilator weaning. Using area under ROC curve of the indexes such as PaO2/FiO2, Vt/kg, Cst, RSBI and the &Delta;RSBI to evaluate the predictability of children ventilator weaning.</p><p><b>RESULT</b>One hundred and thirty-two children were enrolled and 109 succeeded in ventilator weaning, while 23 failed. Seventy-nine children using SIMV+PSV mode and 64 succeeded in ventilator weaning among them, while 23 failed. The success rate was 81.0%. The remaining 53 children using SBT mode, 45 among them succeeded and 8 failed in ventilator weaning, which success rate was 84.9%. There was no statistically significant difference in ventilator weaning mode(χ2=0.334, P=0.563). Age, gender, weight, critical illness score between the succeeded group and the failed group did not show statistically significant difference (t=-0.661; χ2=1.271; t=-0.749, -0.020; P>0.05), but duration of mechanical ventilation, PICU stays and the causes for mechanical ventilation showed significant difference (t=3.751, 3.701; χ2=11.273, P<0.05). The area under ROC curve of PaO2/FiO2, Vt/kg, Cst and RSBI in SIMV+PSV was 0.506, 0.415,0.557 and 0.512, in SBT was 0.503, 0.653, 0.546 and 0.708 separately. There was no statistically significant difference in the predictors mentioned above (P>0.05). But the area under ROC curve of &Delta;RSBI in SBT was 0.814, which was significantly different compared to RSBI in SIMV+PSV (Z=2.966, P<0.05).</p><p><b>CONCLUSION</b>To children who mechanically ventilated for over 24 hours, duration of mechanical ventilation, PICU stays and the causes for mechanical ventilation are the main influencing factors. &Delta;RSBI is a better predictor with feasibility and safety, which is also worth generalizing.</p>