Early assessment value of brain function prognosis in patients with traumatic brain injury by regional saturation of cerebral oxygenation combined with percentage of α variability / 中华危重病急救医学

ABSTRACT

To explore the usability of regional saturation of cerebral oxygenation (rScO2) combined with percentage of α variability (PAV) in predicting brain function prognosis in patients with traumatic brain injury (TBI). Methods A retrospective analysis was conducted. The clinical data of patients with TBI who were monitored rScO2 and bedside quantitative electroencephalogram (qEEG) admitted to intensive care unit (ICU) of Henan Provincial People's Hospital from August 2018 to July 2019 were collected. The rScO2, PAV, and Glasgow coma scale (GCS) score were recorded within 72 hours after the TBI. The primary prognostic indicator was the 3-month Glasgow outcome score (GOS) score. The differences between the two groups of poor prognosis of brain function (GOS score 1-3) and good prognosis (GOS score 4-5) were compared. Binary multivariate Logistic regression analysis was used to analyze the correlation between rScO2, PAV, GCS score and the prognosis of brain function in patients with TBI. In addition, receiver operating characteristic (ROC) curve was plotted to analyze the predicting value of rScO2 and PAV only or combination for prognosis of brain function. Results A total of 42 patients with TBI were enrolled in the study, with rScO2≥0.60 (grade Ⅰ) in 14 patients, 0.50≤rScO2 < 0.60 (grade Ⅱ) in 16 patients,and rScO2 < 0.50 (grade Ⅲ) in 12 patients. PAV 3-4 scores (grade Ⅰ) were detected in 16 patients, 2 scores (grade Ⅱ) in 17 patients, and 1 score (grade Ⅲ) in 9 patients. GCS score 9-14 (grade Ⅰ) were observed in 13 patients,4-8 (grade Ⅱ) in 23 patients, and 3 (grade Ⅲ) in 6 patients; 18 patients had poor prognosis and 24 had good one. The rScO2, PAV and GCS scores of the poor-prognosis group were significantly higher than those in the good-prognosis group [rScO2 with grade Ⅲ 55.6% (10/18) vs. 8.3% (2/24), PAV with grade Ⅲ 38.9% (7/18) vs. 8.4% (2/24), GCS score with grade Ⅲ 27.7% (5/18) vs. 4.1% (1/24)] with significant differences (all P < 0.05). There was no significant difference in other general data including gender, age, total length of hospital stay or acute physiology and chronic health evaluation Ⅱ (APACHEⅡ) score between the two groups. Binary multivariate Logistic regression analysis showed that rScO2 and PAV were independent risk factors for prognosis of brain in patients with TBI [rScO2 odds ratio (OR) = 4.656, 95% confidence interval (95%CI) was 1.071-20.233, P = 0.040; PAV OR = 3.525, 95%CI was 1.044-11.906, P = 0.042]. ROC curve analysis showed that both of rScO2 and PAV had predictive value for the prognosis of brain function in patients with TBI (AUC was 0.796 and 0.780, respectively, both P < 0.01), and rScO2 combined with PAV had higher predictive value with the AUC of 0.851 (P < 0.01) than rScO2 or PAV alone, the sensitivity was 94.4% and the specificity was 62.5%. Conclusions rScO2 and PAV were associated with early brain function prognosis in patients with TBI. The combination of two monitoring indicators can reliably assess the prognosis of brain function in patients with TBI.