Discussion
The results of this unique prospective nationwide cohort study of children with a newly diagnosed brain tumor demonstrate a high prevalence of sleep problems at brain tumor diagnosis, mostly reported by parents of patients. Actigraphic sleep outcomes were not different from those of healthy controls.
We found a high rate of parent reported child sleep disturbance and sleep related impairment, with up to half of the parents reporting moderate or severe problems. Children themselves frequently reported severe sleep disturbances, especially more shortly after brain tumor diagnosis. These findings are consistent with our expectations, indicating sleep problems are experienced regularly and already at the earliest phase of cancer treatment, possibly arising as a result of factors such as distress and neurological damage. Interestingly, alhough parents reported high rates of sleep related impairments, children did not report this and on average these scores did not differ from the general population.
Differences in self- and proxy-report are common in pediatric research and may be explained by several factors29-31. Firstly, children may underreport symptoms. This can be the result of “response shift”, meaning symptoms are judged differently during cancer treatment than how they would be judged before diagnosis32. It could also be that neurocognitive-, stress- and sleep disturbances impact children’s’ capability of adequately recalling sleep experiences33. Second, parents may overreport symptoms due to feelings of stress and concern. Earlier research suggests that parental distress, parental sleep problems and parenting problems are related to parent reported child sleep30. Hence, differences in self- and proxy-reports emphasize the importance of using both types when measuring sleep, as they are both informative and may provide complementary information in this phase of treatment.
We hypothesized that actigraphic outcomes would show lower sleep estimates compared to age-matched, healthy controls, due to the physical and psychosocial stressors associated with the period following pediatric brain tumor diagnosis. The absence of these findings is largely consistent with earlier, similar research in patients with acute lymphoblastic leukemia (ALL)30. However, children with ALL did show longer sleeping times than healthy peers, possibly because these children require more sleep, as they are ill and physically recovering. It could therefore be argued that although the number of minutes children with recently diagnosed brain tumors sleep is similar to healthy children, they might have a higher need for sleep to support optimal recovery34,35. Furthermore, this study sample was almost entirely assessed during the Covid-19 pandemic, while data of the control group was collected before. It is plausible that children sleep better during the pandemic, as they were able to sleep longer due to home schooling and were less exposed to stimuli during the day36. In the healthy population, it was found that people with insomnia complaints experienced clinically meaningful alleviations of symptoms during the pandemic37. Another factor that may influence sleep during the diagnostic period are substantial efforts and strategies of parents, such as co-sleeping and comforting activities, as illustrated in parents of children with ALL38. Possibly, as parents put in a great deal of energy, they do report sleep problems, and yet these efforts seem relatively effective in terms of sleep duration.
Generally, little research has been done with actigraphy and children with cancer during treatment. However, sleep problems are well described and measured amongst brain tumor survivors11,39. Toxic treatmens effects such as radiation therapy or endocrine disturbances may lead to those sleep problems at a later stage. Longitudinal data from this current study should provide more insight into this matter11,40.
This study has several limitations. Although the participant group is relatively large for pediatric brain tumor research, there may not have been enough power to demonstrate specific predictors for sleep problems. In addition, not all parents and children participated in all sleep measurements, due to for example treatment toxicity or study burden, increasing the risk of participation bias. Subsequently, even though participants were recruited from a national pediatric oncology hospital, specific tumor groups were underrepresented which may have lead to selection bias. Twenty-four children with low grade brain tumors who required neurosurgery only and had a favorable prognosis were primarily treated in affiliated hospitals and therefore not participating in this study. Twelve children with high grade tumors were not invited, as they were receiving palliative care with high morbidity, and the treating physician requested not to burden the family.
Actigraphy measures movement and is well validated, but does not measure sleep phases (light, deep and REM sleep). Also, children may have shifted circadian rhythms or inconsistent bedtimes, which is not reflected in the actual number of minutes asleep, but may still contribe to fatigue3,7,41. Future research should explore sleep phases and rhythms to gain more insight into sleep quality. Lastly, previous research suggests more knowledge of parents on sleep hygiene benefits child sleep, suggesting education and support for parents may be an efficient intervention42,43. Future research should therefore consider collecting more comprehensive data on parents and parenting strategies as well.
In conclusion, sleep problems in children with a brain tumor are frequently reported in the first three months after brain tumor diagnosis. Clinicians should be attentive to sleep problems, as it may induce serious, negative consequences in this already vulnerable group. Increasing our understanding of sleep is of major importance because sleep is vital for recovery and health-related quality of life.
Conflict of Interest. The authors have no conflict of interest to declare.
Acknowledgements. The authors would like to thank the psycho-oncology and neuro-oncology teams for their contribution and support.
Funding. This study was funded by The Princess Máxima Center for Pediatric Oncology.