This prospective cohort observational study found that neurodevelopmental delay was related to the weight of recipients at the time of liver transplantation. After receiving liver transplantation, the neurological development of the patients improved significantly. Moreover, reducing maximum SctO2 was an independent risk factor of neurodevelopmental delay in communication and problem-solving skills.
The neurological development status of recipients after pediatric transplantation has always been an important issue for clinicians or family members of patients. A previous study 15 found that the recipients had defects in visual–spatial and reasoning abilities and verbal and operational intelligence after pediatric liver transplantation. A multicenter observational study 16 revealed that 60.0% and 50.8% of parents thought their children were much poorer in terms of cognition and performance, respectively, after pediatric liver transplantation. Another study 17 found that 12 (26%) children received special education among 46 patients with biliary atresia who underwent pediatric liver transplantation, which was significantly higher than the average level (2.4%). Moreover, the motor abilities and intelligence quotient of these patients significantly reduced compared with those of healthy children. These results were similar to our findings that a proportion of patients suffered a certain degree of neurological dysfunction, which led to long-term neurodevelopmental delay.
The neurodevelopmental delay in our study was related to the weight of recipients (P = 0.018). It is easy to understand that lower weight may represent poor preoperative conditions or nutritional status. Therefore, we assumed that lower-weight patients might be more vulnerable regarding neurological function. Qi et al. also found that low-birth-weight newborns had increased risks of brain injuries, motor difficulties, and developmental delay 18. Furthermore, we also discovered that the neurodevelopmental delay in terms of communication and problem-solving skills was associated with intraoperative maximum SctO2 19. SctO2 monitoring has been proved to reduce the incidence and severity of neurological impairment and other complications 20. Hyttel-Sorensen et al. discovered that using SctO2 in premature infants could significantly reduce the incidence of cerebral hypoxia 21. Another study 22 revealed that postoperative cognitive impairment in elderly patients could be reduced by the close monitoring of SctO2 and BIS. The incidence of postoperative neurological defect or cognitive dysfunction was effectively reduced by increasing blood pressure, improving hemoglobin level, and adjusting respiratory parameters and oxygen concentration according to SctO2. In addition, Wagenaar et al. 23 found that the neurodevelopmental status was reflected by SctO2 levels in term neonates with perinatal arterial ischemic stroke. These studies supported our result that the decline in SctO2 might be associated with neurological impairment and long-term neurodevelopmental delay.
Several factors might lead to decreased SctO2 during pediatric living-donor liver transplantation and explain why the decline in SctO2 would cause neurodevelopmental delay. The first possible cause of neurodevelopmental abnormalities was hypovolemia. Intraoperative volume management has always been an issue in anesthesia management, especially for pediatric surgery. Hypovolemia is extremely common in this sophisticated surgery due to the poor preoperative condition of patients and possible hemorrhage during surgery and may result in insufficient organ perfusion including cerebral hypoperfusion. Second, preoperative comorbidities such as pleural effusion, pulmonary infection, and massive ascites often occurred. These comorbidities caused hypoxia and led to a decline in SctO2 24. Third, during the anahepatic phase of liver transplantation, the cardiac output was significantly reduced because of the clamping of the portal vein and inferior vena cava. This caused the reduction of cerebral blood flow25, resulting in decreased SctO226. Fourth, the cerebral autoregulation ability of young children was limited, especially under sevoflurane anesthesia27. Pether et al. revealed that cerebral perfusion is pressure dependent during sevoflurane anesthesia in the youngest pediatric patients, suggesting that the efficiency of the cerebral blood flow autoregulation is limited28. This makes the cerebral tissue more susceptible to hypovolemia or hypoperfusion.
Besides the weight of recipients and SctO2, we also found that the neurodevelopmental delay in problem-solving skills was related to the age of patients (P = 0.026, OR = 0.501, 95% CI: 0.251–0.861). Gungor et al. also found that neurological impairment was more serious in younger recipients after pediatric liver transplantation 29. In our study, the high level of postoperative albumin during hospitalization was statistically associated with poor communication skills (P = 0.049, OR = 1.183, 95% CI: 1.010–1.418). We attribute it to the subjective error of the ASQ scale done by parents and the relative small sample size.
This study had several limitations. First, this was a single-cohort observational study. The total sample size was small, although it was much larger than that of other pediatric liver transplantation studies. The insufficient sample size might have led to bias in statistical results. Second, the duration of follow-up was just 1 year, and hence might not fully reflect the long-term neurological developmental status of patients. Moreover, the neurological developmental status in the year after surgery might be influenced by many other factors such as nutritional intake 30 and parental care 31, which we did not include in our study. Third, the ASQ scale used to evaluate the neurodevelopment status in this study was testified to be consistent with other measurements 32. However, it was still a subjective evaluation scale, which depended on the attention of parents to their children and the cognitive abilities of the parents themselves. Therefore, the results of the ASQ scale might also have some deviation. Although our study had some inevitable limitations, these limitations might promote more excellent design for future prospective multicenter studies.