This study observed the effects of different sedation depths on sleep among children with pneumonia-related respiratory failure without organic CNS injury. This is indicative of individual differences in the severity of genetic disease, patient-ventilator synchrony during mechanical ventilation, and required sedation for organ protection. Moreover, it suggests that appropriately light sedation is conducive for normal sleep recovery during mechanical ventilation and after weaning. EEG monitoring and sleep assessment is helpful for physicians to better implement different sedation regimens for different children. There are currently no monitoring methods appropriate for children with common sleep disorders in the PICU17. Long-term Video EEG is crucial and widely used for brain function evaluation in the PICU. Moreover, the use of video EEG for sleep monitoring and evaluation in children under sedation has increased knowledge regarding the actual sleep conditions of children on mechanical ventilation and under sedation. Comparison of sleep recovery after discontinuing sedatives and weaning from mechanical ventilation has also allowed more accurate sedation implementation. There is a need for sleep monitoring equipment suitable for wide use in the PICU and sleep improvement measures for critically ill children, which could improve their intra-treatment comfort and prognosis.
To show the sedation effects on the sleep cycle of pediatric patients with severe pneumonia underwent mechanical ventilation, we excluded children with CNS infection before starting mechanical ventilation or an altered state of consciousness upon admission. Studies on adult patients on mechanical ventilation showed they lack normal sleep and have frequent sleep interruptions18. We found that children on mechanical ventilation had prolonged sleep duration and fewer awakenings while under sedation, which is consistent with the findings by Jean et al. on adults19. Although sedative use increases sleep duration in children, it alters sleep architecture, and therefore reduces sleep efficiency. In our study, children on mechanical ventilation under sedation had slower EEG activity during sleep, diffuse δ wave activity in both hemispheres, and loss of the NREM-1 and REM stages. The deep sedation group showed more significant changes in the sleep architecture, a higher proportion of REM stage loss, and a greater decrease in the NREM-2 stage than the light sedation group. This is consistent with a previous report that benzodiazepines could decrease REM and increase NREM-2 sleep, respectively, in children on mechanical ventilation20. Our participants included mostly infants and young children, whose sleep regulatory mechanisms remain to be fully developed and who may experience more pronounced sedative effects on sleep. In these children, deep sedation rapidly compressed the NREM-1 and NREM-2 stages into a deep sleep stage; moreover, post-arousal sleep interruption causes decreased REM sleep. On the other hand, we included patients with severe acute pneumonia and respiratory failure, which were mostly related to hypoxemia and hypercapnia at the start of mechanical ventilation. Both hypoxemia and hypercapnia increased exhalation effort and increased arousal and therefore reduced REM sleep. Although there was no between-group difference in the Pediatric Logistic Organ Dysfunction-2 score and overall disease severity, the deep sedation group showed a significantly longer duration of mechanical ventilation and sedation than the light sedation group. This indicates no clear causal relationship between sedation depth and the duration of mechanical ventilation. However, compared with the light sedation group, the deep sedation group showed a significantly higher CO2 level from arterial blood gas analysis before mechanical ventilation, which is indicative of more severe pulmonary dysfunction. This could have attributed to the increased need for sedation in the deep sedation group. In this cohort study, the sedation depth was determined based on the clinical needs of the patients. This could explain why the deep sedation group had poor tolerance to mechanical ventilation, more severe lung disease, and a relatively long duration of mechanical ventilation and sedation. However, this does not exclude the actual sedation depth and duration as factors for delaying respiratory function improvement. Future studies should address these questions.
In addition to frequent awakenings and abnormal sleep architecture, abnormal circadian rhythms are among the sleep disorder manifestations in PICU patients. EEG monitoring of children on mechanical ventilation in the PICU has revealed lower daytime than nighttime slow-wave activity and reduced nighttime sleep efficiency21. In our study, night sleep duration in children on mechanical ventilation under sedation was significantly reduced. The total sleep duration normalized within 1day after weaning from mechanical ventilation and sedative discontinuation; however, the night sleep duration accounted for only 40–50% of the total sleep duration. Circadian rhythm disturbances in children can cause decreased ventilation and dysfunction in the immune and endocrine systems22. Critically ill adult patients with severe REM sleep deprivation and interrupted circadian sleep cycles are more likely to develop delirium. Abnormal sleep cycles and structures in critically ill patients prolong the duration of mechanical ventilation and increase sedative-related short- and long-term psychiatric abnormalities22.
Nursing care procedures are crucial for treating critically ill patients in the PICU; however, frequent nursing care procedures increase pain and stress and affect the sleep architecture. Studies have shown that adult ICU inpatients suffer from sleep interruptions while under nursing care, with increased NREM-1 stage sleep and decreased NREM-3 and REM stage sleep. In our study, the deep sedation group had more nursing care events and fewer awakenings during mechanical ventilation. In both groups, sedation depth was inversely associated with the effects of nursing care stimulation on sleep. Deep sedation in the acute phase can reduce stress response and sleep interruptions caused by stimulation. Additionally, it inhibits airway protective responses, reduces expectoration, increases airway secretions, and increases the requirement for sputum suctioning and other nursing care procedures16, 23–25. This explains the higher number of nursing care events in the deep sedation group than in the light sedation group. Although light sedation facilitates brain function evaluation in patients, it is more likely to cause patient-ventilator dyssynchrony and accidental extubation during mechanical ventilation. For critically ill patients requiring frequent nursing care procedures, a corresponding level of deep sedation is recommended. Although deep sedation has a greater effect on sleep architecture, it can reduce the stress response and decrease sleep interruptions caused by nursing care stimulation.
The correlation between sleep and prognosis in critically ill patients remains unclear. Sleep disorders in critically ill patients may lead to an increased delirium incidence, prolonged mechanical ventilation, and immune dysfunction. Friese et al. suggested that sleep deprivation can increase mortality after severe infections in mice26. In critically ill children whose neurocognition is still developing, the adverse effects of sleep disorders are dependent on the maturity of the sleep arousal regulation system and the overall neurological development stage. Moreover, there is more persistent damage to myelination, synapse formation, or emotional control regulation at younger ages. Sleep spindles and K-complexes appear in NREM-2 sleep, which is a crucial stage for memory integration and maintenance. Boyko et al.7 confirmed that loss of sleep spindles in adults can increase the mortality risk. Most children in this study resumed their sleep cycles after weaning from mechanical ventilation and sedative discontinuation; however, their circadian rhythms remained abnormal. This is related to disease and environmental factors in the ICU; moreover, sleep deprivation and abnormal sleep cycles that occur in the ICU can persist for a few post-discharge months and affect prognosis. Whether children on mechanical ventilation in the PICU develop sleep disorders after discharge remains unclear.
Compared with sleep disorders in adults, those in children are often overlooked. There has been a recent increase in attention on these sleep disorders. including the use of melatonin for sleep disorders in children with neuromuscular diseases27 and respiratory sleep disorders28 and sleep changes during muscle relaxant treatment in the PICU. It is difficult to identify sleep changes in children with pneumonia and comorbid organic nervous system damage. Sleep changes in children without neurological disease, especially after sedation during mechanical ventilation, remain unclear; therefore, we conducted this study. Many ICU patients are on mechanical ventilation; however, given the relatively young age of our patients, the leads fell off during the trial recording period for some patients, or the monitoring was interrupted for other reasons, which resulted in mid-study withdrawal and a relatively small sample size. Additionally, sleep may be affected by procedures conducted by medical and nursing personnel in the PICU, the light and sounds in the ward, the ventilator mode, and other factors. Follow-up studies should investigate the effects of different ventilator modes on patients and effects of mild hypothermia treatments on sleep, as well as expand the scope of application and monitoring.
In this study, children with severe acute bronchopneumonia who received invasive mechanical ventilation under sedation showed EEG abnormalities. These were characterized by decreased brain electrical activity, sleep periodicity, and reduced/absent physiological sleep waves. Deep sedation during mechanical ventilation can cause longer total sleep duration, shorter awakenings, and increased proportion of deep sleep; however, it may worsen abnormal sleep architecture. After weaning from mechanical ventilation and sedative discontinuation, lightly sedated children exhibited better sleep cycle recovery. Future studies should elucidate the long-term physical and psychiatric effects.