RESUMO
BACKGROUND: The gut has been hypothesized to be a protagonist tissue in multiple organ dysfunction syndrome (MODS) for the past three decades. Gastric reactance (XL) is a potential perfusion marker derived from gastric impedance spectroscopy (GIS), which is an emerging tool through which living tissue can be continuously measured to determine its pathophysiological evolution. This study aimed to compare the performance of XL [positive predictive values (PPV), negative predictive values (NPV), and area under the curve (AUC)] against commonly used perfusion markers before and during hypovolemic shock in swine subjects. METHODS: Prospective, controlled animal trial with two groups, control group (CG) N = 5 and shock (MAP ≤ 48 mmHg) group (SG) N = 16. Comparison time points were defined as T-2 (2 h before shock), T-1 (1 h before shock), T0 (shock), T1 (1 h after shock), and T2 (2 h after shock). Shock severity was assessed through blood gases, systemic and hemodynamic variables, and via histological examination for assessing inflammation-edema and detachment in the gastric mucosa. Macroscopic assessment of the gastric mucosa was defined in five levels (0-normal mucosa, 1-stippling or epithelial hemorrhage, 2-pale mucosa, 3-violet mucosa, and 4-marmoreal mucosa). Receiver Operating Characteristic (ROC) curves of perfusion markers and XL were calculated to identify optimal cutoff values and their individual ability to predict hypovolemic shock. RESULTS: Comparison among the CG and the SG showed statistically significant differences in XL measurements at T-1, T0, T1, and T2, while lactate showed statistically significant differences until T1 and T2. Statistically significant differences were detected in mucosa class (p < 0.001) and in inflammation-edema in the gastric body and the fundus (p = 0.021 and p = 0.043). The performance of the minimum XL value per subject per event (XL_Min) was better (0.81 ≤ AUC ≤ 0.96, 0.93 ≤ PPV ≤ 1.00, 0.45 ≤ NPV ≤ 0.83) than maximum lactate value (Lac_Max) per subject per event (0.29 ≤ AUC ≤ 0.82, 0.82 ≤ PPV ≤ 0.91, 0.24 ≤ NPV ≤ 0.82). Cutoff values for XL_Min show progressive increases at each time point, while cutoff values for Lac_Max increase only at T2. CONCLUSIONS: XL proved to be an indirect and consistent marker of inadequate gastric mucosal perfusion, which shows significant and detectable changes before commonly used markers of global perfusion under the hypovolemic shock conditions outlined in this work.
RESUMO
Gastric impedance spectroscopy has been proposed as a method of monitoring mucosal injury due to hypoperfusion and ischemia in the critically ill. During validation tests for this procedure, it was found that 60% of the measurements had errors by factors inherent to the clinical setting, indicating that some kind of automatic error detection should be incorporated to potentially avoid the loss of measurements. This paper presents an algorithm developed to detect errors due to bad connection, bad location or bad contact of the electrode probe. A labeled database with 20,908 sets of 92 spectral measurements each, obtained from critically ill patients was used as training/testing data. To reduce the dimensionality, the database was resized by dividing the spectral range into four bands, and then by computing mean and standard deviation in magnitude, phase, resistance and reactance for each band and measurement. Initial exploration into the data space was performed by k-means clustering, establishing the number of classes. Sequential Forward Selection was performed to determine best features from the reduced data set. Finally, Support Vector Machine classifiers were designed in a one-vs-rest hierarchical scheme to classify the quality of the spectra. Each classifier gave a hit rate greater than 95% and an area under the relative operating characteristic curve of 0.99. In a validation run with cardiac surgery and intensive care unit patient spectra, the error rates were 2.3% and 8.4% respectively.
