In this study, 135 children with TBI (78.8%) had gastrointestinal dysfunction on the first day of admission, suggesting that gastrointestinal dysfunction in children with TBI is very common. The incidence of stress ulcers in children with TBI was 48.5% and gradually increased with TBI severity, with a rate of 85.9% in severe cases, suggesting that the risk of gastrointestinal mucosal ischemia is high, which is consistent with previous reports in adult TBI patients [10]. The gastrointestinal tract is the only system that is jointly controlled by the central nervous system, enteric nerves, and autonomic nerves; thus, the complex neuroendocrine network named the brain-gut axis plays an important role in regulating gastrointestinal function. After the central nervous system is damaged, various pathways in the brain-gut axis are blocked, and many brain-gut peptides are secreted abnormally. These pathways and peptides cannot transmit information normally or stimulate gastrointestinal motility [11-13]. The gastrointestinal mucosa is in a state of hypoperfusion, resulting in the spread of focal small ulcers, which usually manifests as stress ulcers. In addition, gastrointestinal hormone levels and secretion are disordered, and intestinal flora are imbalanced. Feedback from the gastrointestinal tract to the central nervous system is abnormal, causing gastrointestinal dysfunction [14-16].
Approximately one-third of the children developed FI on the first day of ICU admission. These children were significantly more severely ill (lower GCS scores and higher SOFA scores and PCISs) and exhibited greater morality than those without FI. The prevalence of FI was lower than that in previous studies on adult critically ill patients [17, 18]. The reasons for this result are not clear. We speculate that 1) the included children did not have primary gastrointestinal injury. Compared with elderly critically ill patients with cardiovascular disease or diabetes, children's organ function is basically normal; 2) children's nervous systems are still developing; therefore, the stress response may not be completely elicited, and the abnormal release of brain-gut peptides is weakened [19]; 3) the gastrointestinal system of children can adapt to changes in the structure and function of neuronal circuits, and mucosal repair and functional reconstruction mechanisms are stronger in children than in adults [20]. Although we defined FI using objective measurements for greater precision, FI is a subjective variable and a universally used clinical characteristic covering the entire spectrum of gastrointestinal symptoms. FI allows a functional assessment with some clinical relevance [21], as shown in our study. IAH did not occur as frequently as FI in our study, and ACS occurred less often. Reference to adult evaluation criteria is of somewhat limited value. Further studies are needed to standardize the evaluation criteria.
The incidence of secondary gastrointestinal dysfunction increased with injury severity, and the GIF score on the first day and the mean GIF score for the first three days were significantly different among the severe, moderate and mild groups. The intestinal mucosa morphology was found to change within a short time after trauma, including epithelial cell detachment and apoptosis, rupture of the villi, edema of interstitial tissue and the lamina propria, interruption of tight junctions, etc., and the mucosal barrier was observed to have lost its protective function [11]. Additionally, the intestinal flora was completely disordered within a few hours after injury, and the microbial composition and relative abundance changed significantly. The number of beneficial microbiota decreased, while pathogenic flora, which showed relatively increased invasiveness and virulence, dominated the intestinal tract, and the diversity and stability of the microbial ecological system were destroyed. More severe trauma corresponded to, worse dysbiosis and a greater effect on gastrointestinal function. Thus, the GIF score is high. The infection risk and the mortality rate of pediatric patients were substantially increased [12].
The comparison of the relevant clinical indicators in the deceased and survival groups indicated that the GIF score on the first day and the mean GIF score for the first three days in the deceased group were significantly higher than those in the survival group, as were the SOFA scores and PCISs. The multivariate logistic regression analysis suggested that a high mean GIF score for the first three days was closely related to mortality as an independent risk factor. The GIF score reflects gastrointestinal function, which can be classified into different levels similar to other scoring systems for organ function failure. The clinical value and high reliability of the GIF score for predicting outcomes has been verified in intensive care patients and in digestive system diseases [10, 16, 22]. Although the GIF score can be used as an independent risk factor for predicting the risk of death in critically ill patients [23], it focuses on gastrointestinal function at the time of injury. However, the condition of a child develops and changes during hospitalization. The reliability of the GIF score on the first day was relatively low, and its role in predicting death during the entire ICU stay is limited [24]. The mean GIF score for the first three days can be used to dynamically observe and assess changes in gastrointestinal dysfunction during peak disease development, providing better continuity. Reintam et al. [8] found that the mean GIF score for the first three days was more important in predicting death than was the GIF score on the first day. In this study, the mean GIF score for the first three days was considered an independent risk factor rather than the GIF score for the first day. The reliability of the latter score was relatively low, suggesting that although the GIF score can be used as an objective indicator, the effectiveness and accuracy of dynamic observation and scoring are even higher. The mean GIF score for the first three days is better than the GIF score on the first day for evaluating the gastrointestinal function of children with TBI. One limitation of this study is that the SOFA score may be inapplicable to young infants and toddlers in terms of the items assessed [25-27]. The PCIS fully integrates the physiological and morbidity characteristics of children at different ages, which were introduced to predict the risk of death [28, 29]. In this study, the mean GIF score for the first three days had a predictive ability for death comparable to that for the PCIS. They both had good predictive abilities for the risk of death, once again confirming the clinical significance of the GIF score in diagnosing gastrointestinal dysfunction in children with TBI and further emphasizing the importance of continuous monitoring and dynamic observation of the gastrointestinal status of children at different time points. Organ dysfunction in critically ill patients should be scored dynamically [30].