2024, Number 6
Prediction of the weaning off invasive mechanical ventilation by electrical impedance tomography applying a neural network
Salvador IIJ, Cadeza AJD, Monasterios LSG, Ríos AMA, Hernández CCM, Nicolás MEL
Language: Spanish
References: 16
Page: 427-432
PDF size: 300.75 Kb.
ABSTRACT
Introduction: successful weaning from mechanical ventilation (MV) is defined by the absence of ventilatory support 48 hours after extubation; weaning time can represent up to 50% of the total ventilation time. Electrical impedance tomography (EIT) is a noninvasive, radiation-free clinical imaging tool to monitor, in real time and at the patient's bedside. Objective: to compare the differences in dynamic changes of ΔEELI and regions of interest (ROI) by EIT during spontaneous ventilation test in patients with success or failure during withdrawal of invasive mechanical ventilation. Material and methods: observational, longitudinal and analytical study. Patients requiring invasive mechanical ventilation for more than 72 hours were included. Descriptive statistics were used for quantitative variables, expressing the data as mean and standard deviation, or median and interquartile range (IQR) according to the distribution, and as frequencies and percentages for categorical data. Subsequently, multivariate logistic regression (MLR) and neural network (NN) analysis was performed, adjusted for variables with clinical and statistical significance. Statistical significance was established as a p < 0.05 or < 5%. Results: a total of 30 patients were included and divided into two groups: extubation success or failure. Statistical significance was obtained between both groups in the variables: SOFA with a p = 0.015, APACHE II with a p = 0.005, leukocytes with a p = 0.001 and magnesium with a p = 0.035. The resulting predicted probability of MLR and NN for the whole group was used to obtain ROC curves and the cut-off value of –7.5 of post-SBT (spontaneous breathing trial) loss of ΔEELI ROI1. Conclusion: patients who undergo SBT present changes in functional residual capacity associated with loss of recruitment of previously ventilated areas on MV. With the advent of EIT, these changes can be monitored dynamically and at the patient's bedside in real time, offering a prognostic tool in those patients at high risk of failure upon weaning from mechanical ventilation.REFERENCES
Longhini F, Maugeri J, Andreoni C, Ronco C, Bruni A, Garofalo E, et al. Electrical impedance tomography during spontaneous breathing trials and after extubation in critically ill patients at high risk for extubation failure: a multicenter observational study. Ann Intensive Care. 2019;9(1):88. doi: 10.1186/s13613-019-0565-0