Sleep Medicine Reviews

Transcription

1 Sleep Medicine Reviews 30 (2016) 63e71 Contents lists available at ScienceDirect Sleep Medicine Reviews journal homepage: CLINICAL REVIEW Sleep disorders in children with cerebral palsy: An integrative review Ana Luíza P.A. Lelis a, *, Maria Vera L.M. Cardoso a, Wendy A. Hall b a Nursing Department, Federal University of Ceara (UFC), Alexandre Baraúna, 1115 Rodolfo Teofilo, CE, Brazil b School of Nursing, University of British Columbia (UBC), T Wesbrook Mall, Vancouver, BC, Canada article info summary Article history: Received 30 March 2015 Received in revised form 25 November 2015 Accepted 26 November 2015 Available online 11 December 2015 Keywords: Child Sleep disorders Cerebral palsy Sleep disorders are more prevalent in children with cerebral palsy. The review aimed to identify and synthesize information about the nature of sleep disorders and their related factors in children with cerebral palsy. We performed an electronic search by using the search terms sleep/child*, and sleep/ cerebral palsy in the following databases: Latin American literature on health sciences, SCOPUS, medical publications, cumulative index to nursing and allied health literature, psycinfo, worldcat, web of science, and the Cochrane library. The selection criteria were studies: available in Portuguese, English or Spanish and published between 2004 and 2014, with results addressing sleep disorders in children (ages 0e18 y) with a diagnosis of cerebral palsy. 36,361 abstracts were identified. Of those, 37 papers were selected, and 25 excluded. Twelve papers were incorporated in the study sample: eight quantitative studies, three reviews, and one case study. Eleven types of sleep disorders were identified, such as difficult morning awakening, insomnia, nightmares, difficulties in initiating and maintaining nighttime sleep (night waking), and sleep anxiety. Twenty-one factors were linked to sleep disorders, which we classified as intrinsic factors associated with common comorbidities accompanying cerebral palsy, and extrinsic aspects, specifically environmental and socio-familial variables, and clinical-surgical and pharmacological interventions Elsevier Ltd. All rights reserved. Introduction Sleep disorders occur in approximately 25% of typically developing children and are often reported by parents [1]. In samples of children with disabilities, the etiologies of sleep disorders are diverse [2]. Causes may include the extent and location of brain abnormalities, severity of developmental delay, any accompanying sensory loss, epilepsy, intellectual deficits, cerebral palsy (CP), attention-deficitehyperactivity disorder (ADHD), autism, eating difficulties, and other neuropsychiatric health problems such as anxiety, hyperactivity, aggressive behavior, depression, and pain [3]. Environmental factors and caregivers' behaviors can also contribute to children's sleep disorders. Because treatments for sleep disorders associated with particular features vary it is vitally important to identify factors related to sleep disorders [4] in homogeneous samples of children with CP [5]. * Corresponding author. Filomeno Gomes Avenue, Jacarecanga Neighborhood Fortaleza, Ceara, Post Code: , Brazil. Tel.: þ address: aninhanurse123@gmail.com (A.L.P.A. Lelis). CP is a term representing a group of disorders relating to motor development due to non-progressive lesions of the developing brain, often accompanied by disturbances of sensation, cognition, and/or a seizure disorder [6]. Depending on the type of CP, 25e80% of children have additional impairments. Of those children, a large proportion has some level of intellectual impairment; the prevalence varies with the type of CP but increases when epilepsy is present [7]. Odding and colleagues indicated that epilepsy is present in 20e40% of CP cases and is most common among children with hemiplegic or quadriplegic CP. They also reported that up to 80% of children diagnosed with CP have some impairment of speech and almost 75% have low visual acuity. Half of all children diagnosed with CP have gastrointestinal and feeding problems, with inadequate growth occurring in 25%, while being under or overweight characterizes about 50% of children [7]. The higher prevalence of sleep disorders in children with CP has typically been represented by the following characteristics: sleepewake transition disorders, excessive daytime sleepiness, and arousal disorders. Active epilepsy in children with CP has had the strongest correlation with sleep disorders, specifically with excessive daytime sleepiness [5] / 2015 Elsevier Ltd. All rights reserved.

2 64 A.L.P.A. Lelis et al. / Sleep Medicine Reviews 30 (2016) 63e71 Abbreviations ADHD attention-deficitehyperactivity disorder AHI apnea hypopnea index AI arousal index CBCL child behavior check list CSHQ child sleep habits questionnaire CINAHL cumulative index to nursing and allied health literature CP cerebral palsy EDS excessive daytime sleepiness EMG electromyographic GMFSC gross motor function classification system HB hyperbolide LILACS NDD NTPE OSA PLMD PSG PSQ PSQI REM RLS SDB SDSC TBS TD Latin American literature on health sciences neurodevelopmental disability night-time postural equipment obstructive sleep apnea periodic limb movement disorder polysomnograms pediatric sleep questionnaire Pittsburgh sleep quality index rapid eye movement restless leg syndrome sleep disordered breathing sleep disturbances scale for children tongue base suspension typically developing When children experience sleep deprivation they evidence mood, behavioral, and cognitive impairments [8,9]. Insufficient or inefficient sleep adversely affects cognitive tasks, such as executive control, attention regulation, and working memory [10]. Lower quality of life in children with CP has been associated with insomnia and excessive daytime sleepiness. Lower quality of life is linked with impaired physical, emotional, social, and school functioning; considering effects of sleep disorders in general is important to raise awareness about managing sleep in children with CP [11]. Therefore, it is important to review and synthesize scientific evidence about the nature of children's sleep disorders associated with CP and related factors. Study question and purpose This integrative review aims to provide a comprehensive review and synthesis of the research literature about sleep disorders and factors related to those disorders in children with CP. For the purposes of the review, a child is defined as a person between 0 and 18 y. The research questions were: What is the nature of sleep disorders and what factors are related to sleep disorders in children with CP? Methods An integrative review represents a method to summarize past research literature and to provide comprehensive understanding of an area [12]. In contrast to systematic reviews that include only quantitative studies, this method includes all study designs while still adhering to rigorous review processes [12e14]. It involves identification of research questions, a literature search, categorization and assessment of studies, interpretation of results, and synthesis of knowledge [14,15]. A systematic search was conducted using eight databases in the following sequence: Latin American literature on health sciences (LILACS), Scopus medical publications (PubMed), cumulative index to nursing and allied health literature (CINAHL), PsycInfo WorldCat Web of Science and the Cochrane library. Search terms were sleep, child* and cerebral palsy. The bibliographic survey occurred from June to the end of July, The inclusion criteria for this review were studies: 1) available in Portuguese, English or Spanish; 2) addressing sleep disorders in children, with diagnosis of CP, aged between 0 and 18 y; 3) published between 2004 and 2014, and 4) examining factors associated with sleep disorders in the population of interest. Papers were excluded if: 1) the manuscripts were editorials or letters to the editor, 2) the paper was not published in a peerreviewed journal (e.g., abstracts or dissertations), and 3) there were repeated articles in subsequent databases. Quality appraisal and synthesis The articles were classified according to level of evidence: Level I (systematic review or meta-analysis of randomized controlled clinical trials or clinical guidelines based on systematic reviews of randomized controlled trials); Level II (at least one well-designed randomized controlled trial); Level III (well-designed clinical trials without randomization); Level IV (well-designed cohort and case-control studies); Level V (systematic reviews of descriptive and qualitative studies); Level VI (single descriptive or qualitative study) and; Level VII (opinions of authorities and/or report of expert committees) [16]. Results Study sample By using the broad search terms sleep and child*, and sleep and cerebral palsy, in eight databases, 36,361 abstracts were identified. Based on a review of titles and study abstracts, 36,324 were excluded because they included primarily adults (n ¼ 20,300), did not pertain to purpose of this study (n ¼ 16,000) and/or represented heterogeneous disabilities in samples (n ¼ 24). Based on the specified inclusion criteria, 37 papers were selected. Following more in-depth examination using the inclusion and exclusion criteria, 25 were excluded: 20 were duplicates and five did not include risk factors related to sleep disorders in their findings. A total of 12 papers were included in study sample. The selection process is shown in Fig. 1. The study sample set included eight quantitative studies, three reviews, and one case study. Cross-sectional design was applied in four of quantitative studies, case-control in two, a retrospective chart review in one and one was a prospective cohort study. The majority of the studies were published in pediatric journals; three were focused in the field of neurology. Most studies presented a low level of evidence: five with level VI, three level V and three level IV. Two authors conducted studies in the United Kingdom, while the remainder occurred in Ireland, The Netherlands, India, United States of America, Italy, Brazil and Malaysia. Eight studies assessed sleep including: frequency, types of sleep disorders, and their risk factors. Two studies evaluated sleep in

3 A.L.P.A. Lelis et al. / Sleep Medicine Reviews 30 (2016) 63e71 65 LILACS 201 CINHAL 3,872 PsycInfo 2,825 Scopus 11,789 PubMed 8,524 Web of Science 8,696 WorldCat 25 Cochrane Excluded 25 Included sample 12 Fig. 1. Flow chart of study selection process. children using specific treatment devices (night-time postural equipment and night orthoses), and two investigated sleep improvement after using a surgical intervention of tongue base suspension (TBS) to treat obstructive sleep apnea (OSA), and hyperbolide (HB) mastication to reduce sleep bruxism. Seven studies used scales or questionnaires to measure sleep disorders, one assessed sleep by questionnaire and polysomnography, and one used only polysomnography. The sleep disturbances scale for children (SDSC) was used for three studies and the pediatric sleep questionnaire (PSQ) and child sleep habits questionnaire (CSHQ) were used in the other studies. The Pittsburgh sleep quality index (PSQI) was used to assess mothers' sleep. Two studies did not specify the nature of the questionnaires employed to assess sleep. Authors defined sleep disorders or sleep problems according to cut-off score criteria for each measure. A total of 1385 subjects (732 children; 653 parents) participated in the studies (mean subjects per study ¼ 115; range of subjects per study ¼ 2e327). Children's ages were divided into developmental groups as follows: infants (0e11.9 mo), toddlers (12e35.9 mo), preschoolers (3e5 y), school-aged (6e12 y), and adolescents (13e18 y). Three studies included subjects from three age groups (preschoolers, school-aged and adolescents). Three studies included subjects from one age group (school-aged), two studies included subjects across age groups, and one study included subjects from two age groups (preschoolers and school-aged). Schoolaged children were most commonly studied (nine studies) followed by preschoolers (6), adolescents (4), toddlers (3), and infants (2). In seven studies, parents or main caregivers completed sleep questionnaires. The two case control studies compared sleep for children with CP to that of typically developing (TD) children. The case study involved a seven-year-old boy with severe spastic CP and sleep bruxism. Authors defined typically developing children as having no chronic illnesses that might interfere with sleep (e.g., asthma requiring medication or cardiac failure) [17], and as children whose development was within the normal range [18]. Authors of the three review papers neither reported their search methods nor numbers of included and excluded articles. In the eight studies that classified the type of CP, 212 children had spastic diplegia, 195 spastic quadriplegia, 181 spastic hemiplegia, 51 dyskinetic, 21 hypotonic, six ataxic, three athetoid and three unknown [5,17e20,25e27]. Analytic approaches for each study are described in Table 1. The nature of sleep disorders described in Table 2 included: sleep hyperhidrosis [17,19], disorders of arousal [17], difficult morning awakening [20], sleep disordered breathing [18,20e24], bruxism [20,25], insomnia [20], nightmares [20], parasomnias, e.g., sleepwalking [18,20], difficulties in initiating and maintaining nighttime sleep (night waking) [5,18,26], sleepewake transition disorders [19], and sleep anxiety [18]. Twenty-one factors were related to sleep disorders in children with CP (Table 2). The factors were organized according to categorical groups of comorbidities and factors related to sleep in children with CP (Table 3): intrinsic factors were related to pathophysiological features and extrinsic factors classified as environmental issues. Pathophysiological factors were identified in a number of publications. Six papers linked sleep disorders to epilepsy/nocturnal seizures [5,18e20,23,24] and six to visual impairment [5,18,20e24]. Three papers implicated mental/cognitive impairment [18,19,23]. Three publications linked sleep disorders to type of CP [5,19,20]. Three publications linked sleep disorders to upper airway obstruction [21,23,24]. Two papers implicated gastro-esophageal reflux [21,24], with two others pointing to postural limitations/ muscle spasticity [21,24]. Two papers referenced physical discomfort/muscle pain [21,24]. One paper in each case implicated brainstem dysfunction [21], psychiatric problems [20], pulmonary aspiration [21], hearing impairment [20], abnormal melatonin secretion [23], and behavioral problems [19]. For environmental factors, three publications related sleep disorders to the use of devices [25e27], while three others implicated anticonvulsant medication [19,21,23]. One paper implicated surgical intervention [22]. Parental characteristics linked to sleep disorders were identified in two papers that referred to bed-sharing [5,19], one article that implicated single-parent status [5], one paper that referred to caregiver employment status [18], and one that implicated caregivers' sleep disorders [17]. Our results are organized by intrinsic and extrinsic factors related to sleep disorders as presented in Table 3, as well as by the strength of the evidence supporting a factor. Discussion This review provided a comprehensive synthesis of the research literature about the nature of sleep disorders for children with CP and factors related to their sleep disorders. We have identified eleven types of sleep disorders and twenty-one factors linked to sleep disorders for children with CP. Our findings extend previous descriptions of sleep disorders experienced by children with CP [5,17e27], thereby potentially increasing awareness of care providers about sleep problem management. Unfortunately, sleep disorders associated with CP are difficult to categorize. The majority of the studies used questionnaires that examined symptoms of sleep problems. The data included a mixture of symptoms of poor sleep and classification by types of sleep disorders; in some studies

4 66 A.L.P.A. Lelis et al. / Sleep Medicine Reviews 30 (2016) 63e71 Table 1 Methodological characteristics of included original publications. Authors Objective of study Nature of sample Analytic statistical techniques Limitations Atmawidjaja et al., 2014 [17] Elsayed et al., 2013 [20] Fitzgerald et al., 2009 [21] Giannasi et al., 2014 [26] Hartzell 2013 [22] Hill et al., 2009 [33] Mol et al., 2012 [27] Newman et al., 2006 [5] Romeo et al., 2014 [19] Simard-Tremblay et al., 2011 [23] Singhi, Sankhyan 2012 [25] Wayte et al., 2012 [18] Caseecontrol study to determine if a CP population was at increased risk of sleep disturbances, and factors associated with risk. Cross-sectional study assessing sleep of children with cerebral palsy Review highlighting problems and potential therapies for suppurative lung disease and sleep disordered breathing in children with CP. Case study evaluating the effect of the HB mastication on EMG activity in the jawclosing muscles and the reduction in sleep bruxism in a child with CP using surface EMG analysis before and after nine months of treatment Retrospective chart review determining the benefit of the TBS technique, its safety, and appropriate use. Cross-sectional pilot study to explore the influence of NTPE on sleep quality and respiratory function in children with severe CP. Cross-sectional study investigating whether children with CP using night orthoses experience more sleep disturbance than children not using night orthoses Cross-sectional design to determine the frequency of sleep disorders in children with CP and identify factors associated with these problems. Prospective cohort study estimating the frequency of sleep disorders in children with CP using the sleep disturbance scale for children (SDSC) and evaluating relationships between sleep disorders and motor, cognitive, and behavioral problems. Review paper exploring types of sleep problems, risk factors, and consequences in children with CP. Review paper focusing on common co-morbidities for children with CP Case-control study examining relationships between children's sleep, and maternal sleep and mood for children with CP. 109 children with CP (aged 4e18y) with their siblings and their main caregivers 100 children with CP (2 e12y) and their parents Univariate analysis; Chi-square; Wilcoxon rank-sum test; multiple regression analysis Mean; Median; Percentiles; Chi-square Population more severely affected and sample size may not have been sufficient. Questionnaire-based studies are subject to reporter bias, even when healthy siblings are used as comparisons. None identified Not specified Not applicable None identified A seven-year-old boy with severe spastic CP and sleep bruxism and his mother 14 charts of children with CP (ages 0e18y) treated for OSA. 10 Children (3e16y) and their parents 82 children 6 to 15y (55 using night orthoses and their parents; 27 not using night orthoses) 173 parents of children (6 e10 y) 165 children with CP (aged 6e16 y) and their main caregiver Not applicable Student's paired t-test. Wilcoxon-signed rank tests; Paired-sample t-tests; Bonferroni adjustment; Chisquared test; ManneWhitney U-test; one-way t-test Descriptive statistics; Pearson multivariate; analysis of covariance correlations; Wilk's lambda Univariate analyses; multivariable analyses Mean; median; KruskaleWallis equality of populations rank test; Fisher exact test; Spearman rank order correlation test; multivariate analysis None identified Small sample size None identified This study has a small sample and only questionnaires were used. None identified Not specified Not applicable None identified Not specified Not applicable None identified 40 children with CP (aged 4 e11y) and their mothers 102 TD children (siblings or friends of children with CP (aged 4e11y) Mann-Whitney U; Kruskal Wallis test; Spearman's correlation coefficients; Pearson Chi-square test for categorical variables The lack of an age-matched control group of typically developing children, which would have allowed more detailed analysis for possible differences with CP children. Parent report increases the potential for reporter bias. Abbreviations: CP: cerebral palsy; EMG: electromyographic; HB: hyperbolide; EMG: electromyographic; NTPE: night-time postural equipment; OSA: obstructive sleep apnea; SDSC: sleep disturbance scale for children; TBS: tongue base suspension; TD: typically developing.

5 A.L.P.A. Lelis et al. / Sleep Medicine Reviews 30 (2016) 63e71 67 Table 2 Major publication results. Authors Major results Atmawidjaja et al., 2014 [17] Proportions of children with pathological scores, sleep hyperhidrosis, and disorder of arousal were significantly higher in the CP group than the comparison group. Caregiver sleep problems [sleep duration of less than 7 h (p ¼ 0.02) and sleep latency of more than 30 min (p ¼ 0.03)], rather than medical variables, were significantly associated with children's sleep disturbances. Elsayed et al., 2013 [20] Difficult morning awakening and visual impairment were positively correlated, as were psychiatric problems and SDB (p ¼ 0.000), bruxism (p ¼ 0.014) and EDS (p ¼ 0.000). PLMD/RLS showed a positive correlation with hearing problems (p ¼ 0.002). Insomnia and sleep disordered breathing (SDB) were more common with gross motor function classification system (GMFSC) grade V and IV, sleep bruxism in GMFSC grade III, and nightmares and sleepwalking in GMFSC grade II and I. Fitzgerald et al., 2009 [21] Factors increasing risk for children with CP for sleep disorders were: upper airway obstruction; severe visual impairment; brainstem dysfunctions; postural limitations and physical discomfort; epilepsy treatment with anticonvulsants, and pulmonary aspiration. Giannasi et al., 2014 [26] The child did not exhibit daytime bruxism and both dysphagia and salivary incontinence had been significantly reduced by using the HB mastication. There was a significant reduction in tooth grinding during sleep and salivary incontinence during waking hours and improvement in sucking-swallowing movements at meals. Hartzell et al., 2013 [22] The AHI decreased in the TBS group by an average of 16.5 units (p ¼ 0.03) while the non-tbs group also showed a significant improvement with an average decrease of 5.0 units (p ¼ 0.04). The AI of the TBS group showed significant improvement (change of 12.4; p ¼ 0.05), which was not significant in the non-tbs group (5.9; p ¼ 0.10). Hill et al., 2009 [33] There were no significant differences in sleep quality measures when children were in their NTPE compared with sleeping unsupported in bed. Mol et al., 2012 [27] Children using night orthoses only during daytime experienced significantly more disturbance in initiating and maintaining nighttime sleep (p < 0.05) in comparison with children using orthoses at night or using night orthoses during day and night (L ¼ 0.73; F (12,90) ¼ 1.27; p ¼ 0.25). No statistically significant differences in sleep disturbance were found between children using and not using night orthoses (L ¼ 0.91; F(6,73) ¼ 1.27; p ¼ 0.28) Newman et al., 2006 [5] A higher sleep score (poorer sleep quality) was significantly associated with active epilepsy as well as being the child of a singleparent family, and sleeping with parents (OR ¼ 17.1, 95% CI 2.5e115.3; p ¼ 0.004). Disorders of initiation and maintenance of sleep based on the SDSC were associated with diagnoses of spastic quadriplegia, dyskinetic CP (OR ¼ 12.9, 95% CI 1.9e88.0; p ¼ 0.009; OR ¼ 20.6, 95% CI 3.1e135.0; p ¼ 0.002), and severe visual impairment (OR ¼ 12.5, 95% CI 2.5e63.1; p ¼ 0.002), as well as with bed-sharing. Romeo et al., 2014 [19] An abnormal sleep score based on the SDSC was significantly associated with mental retardation, presence of epilepsy (controlled or active), CBCL scores [externalizing (r ¼ 0.28; p > 0.05), internalizing (r ¼ 0.36; p < 0.05), and total score (r ¼ 0.49; P < 0.01)] and level five on the GMFCS (p < 0.01). Dyskinetic CP children presented statistically significantly higher scores for sleepewake transition disorders (p < 0. 05) and sleep hyperhidrosis (p < 0.01) than children with hemiplegia, quadriplegia, or diplegia. Simard-Tremblay et al., 2011 [23] Children with CP were at higher risk for obstructive sleep apnea syndromes because of abnormal upper airway muscle tone, primary central abnormalities affecting the central control of breathing, and concurrent epilepsy, intellectual disability, a primary sensory impairment (i.e., vision or hearing) and severity of neuromotor impairment. Singhi, Sankhyan 2012 [25] Contributing factors for sleep problems included airway obstruction, nocturnal seizures, spasticity, pain, inappropriate night orthoses, gastroesophageal reflux, and blindness. Children with spastic quadriparesis, dyskinetic CP, and severe visual impairment were significantly more likely to have sleep disorders. Wayte et al., 2012 [18] Children with CP had significantly higher CSHQ scores compared to TD children, higher subscale scores included: sleep anxiety, night waking, parasomnias and sleep-disordered breathing. Correlates of sleep disorders included maternal employment status (p < 0.05) and the extent of the child's visual and cognitive impairment, as well as epilepsy. Abbreviations: AHI: apnea hypopnea index; AI: arousal index; CBCL: child behavior check list; CSHQ: child sleep habits questionnaire; EDS: excessive daytime sleepiness; GMFCS: gross motor function classification system; HB: hyperbolide; NTPE: night-time postural equipment; PLMD: periodic limb movement disorder; RLS: restless leg syndrome; SDB: sleep disordered breathing; SDSC: sleep disturbance scale for children; TBS: tongue base suspension; TD: typical development. Table 3 Comorbidities and their relationship with factors related to sleep disorders in children with CP. Intrinsic Neurological Epilepsy/nocturnal seizures Brainstem dysfunction Abnormal melatonin secretion Intellectual disabilities/cognitive Impairment Mental health Psychiatric problems Behavioral problems Ophthalmological Blindness/visual impairment Hearing Hearing impairment Upper airway Abnormal airway muscle tone Pulmonary Pulmonary aspiration Gastrointestinal Gastro-esophageal reflux Musculoskeletal Type of cerebral palsy Postural limitations/muscle spasticity Physical discomfort/muscle pain Extrinsic Socio-familial variables Bed-sharing Single parents Caregiver employment status Caregiver sleep problems Clinical-surgical and pharmacological interventions Anticonvulsants Use of devices (night postural equipment, night orthoses, hyperbolide mastication) Surgery for upper airway patency (tongue base suspension)

6 68 A.L.P.A. Lelis et al. / Sleep Medicine Reviews 30 (2016) 63e71 identification of sleep disorders was by inference. For example, difficult morning awakening [20], insomnia [20], nightmares [20], difficulties in initiating and maintaining nighttime sleep (night waking) [5,18,26], and sleep anxiety [18] could be linked to behavioral sleep problems which are not specific to the pathophysiology associated with CP per se. Studies reported here largely relied on questionnaires designed for healthy children, e.g., the CSHQ [5,17e20,26,27]. Use of sleep questionnaires designed for children with neurodevelopmental disability (NDD) would have improved the studies. Unfortunately, sleep assessment questionnaires specifically designed for children with NDD have not been available. Reference is made in a paper by Tietze et al. (2012) to measures developed for patients with disabilities [2] but these were structured interviews [28] or a modified questionnaire for TD children [29]. Similarly to the outcomes of Singhi and Sankhyan [25], our findings demonstrated that comorbidities associated with CP (intrinsic factors) affected children's sleep, including airway obstruction, pulmonary aspiration, nocturnal seizures, spasticity, pain, and severe visual impairment. Extrinsic factors included environment, use of specific devices, socio-familial variables, and clinical-surgical and pharmacological interventions. Intrinsic factors Upper airway obstruction, gastro-esophageal reflux, epilepsy/ nocturnal seizures, visual impairment, mental/cognitive impairment, type of CP, and postural limitations/muscle spasticity, with associated physical discomfort/muscle pain, are major factors associated with sleep disorders in children with CP. Behavioral problems, mental health problems, hearing impairment, and brainstem dysfunction/abnormal melatonin secretion have also been implicated in sleep problems, based on limited evidence from the standpoint of Melnyk's and FineouteOverholt's classification of evidence level [16]. Upper airway obstruction, pulmonary aspiration and gastroesophageal reflux have contributed to sleep-related breathing disorders in children with CP [21,23]. Airway obstruction is underappreciated, with potential for OSA only identified from a lengthy history of snoring. For children with stridor and obstructive symptoms when awake, symptomatology has been attributed to laryngeal dystonia, and reported in a small cohort of children with dystonic CP or severe laryngomalacia [21]. Other elements contributing to OSA in children with CP include abnormal central control of respiration and obesity [23], with potential sequelae including disturbed sleep and failure to thrive. Bulbar palsy in children with severe disabilities can have a similar effect; usually, rather than frank OSA, the hypotonic airway (with muscle weakness and scoliosis) causes hypoventilation and hypoxemia during sleep [24]. Cough and pulmonary aspiration are associated with lung disease. Because symptoms may be more apparent in the supine position during sleep a history should document cough and its etiology, gastro-esophageal reflux, and oral secretion management [21]. Neurological factors, such as epilepsy/nocturnal seizures and intellectual disabilities/cognitive impairment, affect children's sleep. Uncontrolled epilepsy/nocturnal seizures could trigger behavioral arousals and wakes. Discrepancies in comparisons of sleep disorders in cases of controlled or uncontrolled epilepsy [18,19] could be related to different kinds of epilepsy and seizures. Seizures occurring predominantly during sleep may have a subclinical presentation. Because questionnaires are non-specific, the co-existence of behavioral sleep disorders could increase sleep disorder scores. Differences in sleep outcomes could be explained by relationships between epilepsy and associated impairments like intellectual, visual, and hearing losses [30]. These losses are independently related to sleep disorders, and, in combination can increase children's risk of sleep disorders. Comorbidities associated with CP enhance extrinsic factors interfering with sleep; medications for treatment of comorbidities can increase intensity of sleep disorders. Nighttime seizure episodes in children with CP can contribute to sleep loss and disruption arousals, while sleep-disordered breathing can increase uncontrolled epilepsy. Links between epilepsy and sleep fragmentation with reduced sleep efficiency and more frequent arousals from sleep are commonly reported [18,19,23]. A child with frequent seizures may have post-ictal phases with somnolence as the initial symptom; however, using anticonvulsants to treat a difficult seizure disorder can contribute to somnolence or difficulties in initiating sleep [21]. Cognitive impairment has been associated with sleep disorders in children with CP. In two studies, an abnormal sleep score was linked with mental retardation [18,19]. Some studies included in a review reported that the degree of intellectual disability predicted sleep disorders [23,29,31,32] whereas others in the same review reported that intellectual disability was not associated with sleep disorders [23,33,34]. Unfortunately, Simard-Tremblay et al. [23] have not stated whether their samples of children had CP diagnoses associated with intellectual disabilities. Combining severely damaged brains and limited development of social relationships, i.e., failure to register others' daytime and nighttime cues, could explain abnormal sleepewake patterns in children [35]. Our review suggests primary visual impairment enhances risk for the development of sleep disorders. Between 20% and 50% of children with cerebral palsy have cortical visual impairment [23]. Children with CP and blindness have had free running circadian rhythms or even inverted sleepewake cycles [5,21]. Blindness or greatly reduced vision may reduce effects of natural light on wakefulness during the daytime by disrupting the secretion of melatonin [21]. Findings are inconsistent because visual disturbance was not statistically significantly related to the level of sleep impairment [18] in one study, or associated with difficult morning awakening in another [20]. Our results suggested that musculoskeletal features linked to CP contribute to sleep pattern changes. Level of motor function, postural restrictions, degree of spasticity and accompanying physical and muscle discomfort have been associated with sleep disorders [5,20,23]. Disorders of initiation and maintenance of sleep have been associated with total body involvement (spastic quadriplegia and dystonic/dyskinetic CP); those children were significantly more likely to have caregiver-reported sleep disorders [5]. Parental reports indicated that dyskinetic CP children presented significantly higher scores for sleepewake transition disorders and sleep hyperhidrosis than children with hemiplegia, quadriplegia, or diplegia, and an abnormal sleep score was significantly associated with level five on the gross motor function classification system (GMFCS) [19]. Another study positively associated sleep disorders with level of motor functioning by using GMFSC. Insomnia and sleep disordered breathing (SDB) were associated with GMFSC grade V and IV while sleep bruxism was more common in grade III. Nightmares and sleep walking were more common in GMFSC grade I and II [20]. Postural limitations and physical discomfort are increased by muscle spasms resulting from dislocated hips, scoliosis, and joint contractures, together with an inability to alter posture in response to discomfort [21]. Physical discomfort and the impaired ability to adjust sleep positions would adversely affect sleep quality. Children may present with increased tiredness, restless sleep, night-time crying or agitation. Antihistamines or benzodiazepines may be used but they do not manage underlying problems, such as

7 A.L.P.A. Lelis et al. / Sleep Medicine Reviews 30 (2016) 63e71 69 dislocated hips or pressure sores. Sedative agents reduce upper airway muscle tone, increasing snoring and problems with the coordinated swallowing of oral secretions [21]. Benzodiazepines used to treat other comorbidities (e.g., behavioral and psychiatric problems) for children with CP contribute to a higher incidence of OSA, making a pharmacological history important. Decreased pain or discomfort can improve sleep quality. A recent study of 35 children with bilateral cerebral palsy (i.e., spastic quadriplegia or diplegia) and severe spasticity demonstrated a reduction in nighttime awakening within six months of implantation of an intrathecal baclofen pump [23]. Children's behavioral sleep disorders are generally linked to bedtime resistance and lack of consistent sleep routines [37e39]. Relationships between sleep scores and internalizing/externalizing behaviors are common in typically developing children [36] so it is unsurprising to encounter those relationships in children diagnosed with CP. The incidence, however, is much higher. About 50% of children with abnormal scores on the child behavior check list (CBCL) had abnormal SDSC scores. Externalizing disorders are more common in children with hemiplegia and dyskinetic CP than in those with quadriplegia and diplegia; more children with dyskinetic CP demonstrate internalizing disorders [19]. Treatment of CP comorbidities with medications like benzodiazepines could increase SDB and excessive daytime sleepiness (EDS), by decreasing upper airway muscle tone and increasing somnolence. Elsayed et al. [20] have reported positive correlations between psychiatric problems and SDB, and bruxism and EDS. Our integrative review identified hearing impairment as a comorbidity associated with CP and children's sleep disorders. Elsayed et al. [20] suggested a correlation between periodic limb movement disorder (PLMD)/restless leg syndrome (RLS) and hearing problems. Inadequate evidence supports that direct link. Other authors have suggested that PLMD/RLS may operate as a prevalent symptom in children with OSA [40]; OSA seems to be linked to hearing impairment [41e43]. Extrinsic factors In addition to family context as a link to behavioral sleep problems, authors have investigated effects of treatments, such as anticonvulsants, use of devices and surgery for upper airway patency, as exacerbating or relieving children's sleep disorders. Effects of anticonvulsants on sleep disorders are controversial. While some authors argued that anticonvulsants decrease sleep latency and improve sleep continuity, phenobarbitone and the benzodiazepines have suppressed rapid eye movement (REM) sleep [21]. Newer anticonvulsants, like lamotrigine, gabapentin and topiramate, do not impact the sleep architecture, but febalmate may trigger insomnia [44]. On the other hand, anticonvulsant agents decrease the disruptive effects of seizures on sleep physiology [45], which may improve sleepewake cycling and reduce somnolence. Those changes frequently reduce sleep-related hypoventilation and concurrent OSA with a resulting improvement in gas exchange [21]. Antiepileptic medications can increase the risk of concurrent sleep problems or demonstrate no association [23]. For example, in a prospective cohort study with 31 children with abnormal sleep scores, 14 presented with active epilepsy, four had controlled epilepsy, and 13 showed no signs of epilepsy. No specific relationship was observed between the type of antiepileptic agent and sleep scores [19]. Uncomfortable devices for improving postural problems commonly used for children with CP [46] have potential to disrupt children's sleep from parents' viewpoints, but our review suggested night positioning equipment did not affect the quality of children's sleep. Nighttime positioning is a frequent caregiver preoccupation, and reports of settling or sleeping difficulties related to the use of nighttime splints are encountered in clinical practice [5]. No significant differences on polysomnographic sleep quality measures were demonstrated when children were in their nighttime postural equipment (NTPE) compared with sleeping unsupported in bed. Use of NTPE was associated with improved mean overnight SpO 2 in three children, but the converse was the case for six participants [33]. Thus, effects of NTPE on children's sleep disorders are unclear. A study investigating night orthoses, including children's ages and GMFCS-levels in analyses, resulted in no statistically significant differences in sleep disturbance between children using night orthoses and those not using them. On the other hand, children using night orthoses only during daytime experienced significantly more difficulty initiating and maintaining sleep compared with children using them at night or during days and nights [27]. A case study evaluated the effect of the HB mastication on electromyographic (EMG) activity in the jaw-closing muscles and reduction in sleep bruxism in a child with CP [26]. Following treatment, the child demonstrated a marked balance in the activity of the jaw-closing muscles, no daytime bruxism, and reduced dysphagia and salivary incontinence. Furthermore, the parents reported a significant reduction in tooth grinding during sleep as well as salivary incontinence during waking hours and a considerable improvement in sucking-swallowing movements at meals. These differences were interpreted as improvements in the child's quality of life [26]. In terms of surgical strategies to treat OSA, a retrospective chart review study examined the efficacy of combined surgical techniques for OSA including tongue base suspension (TBS), using perioperative polysomnograms (PSG). Children diagnosed with CP who received TBS were compared with those who underwent surgical management of OSA without TBS. Both groups showed significant improvements in their PSG data. The apnea hypopnea index (AHI) improved in the TBS group by an average of 16.5 units while the non-tbs group improved significantly with an average decrease of 5.0 units. Only in the TBS group did the arousal index (AI) improve significantly. Moderate to severe OSA in this population may safely benefit from using tongue base suspension [22]. In addition to treatment effects, familial factors were associated with an increase in some sleep disorders. Caregivers' variables, such as being single parents, bed-sharing, maternal employment, low socio-economic status, and sleep problems increased the frequency of children's sleep disorders and disorders of arousal. A child's sleep score was significantly positively associated with living in a singleparent family, and sleeping with parents [5]. A case-control study indicated children's sleep disorders were correlated with maternal employment status (unemployed status was associated with higher CSHQ scores) [18]. Mothers who are unemployed may have more time to devote to children's surveillance at night. Bed-sharing was associated with an increase in total sleep disorders, disorders of initiation, and maintenance of sleep, and sleepewake transition disorders. Children may be unable to selfsoothe, with parents using bed-sharing to calm them, or bedsharing parents' may have increased awareness of their children's night awakenings [5]. Studies have supported the latter explanation because single parents of TD children who bed-share are more aware of children's sleep. As a consequence, they report more sleep problems in their children than non-bed-sharing parents. A case control study demonstrated that caregiver sleep duration of less than 7 h and sleep latency of more than 30 min were significantly associated with higher sleep scores (poorer sleep quality) for children but co-sleeping was not a significant factor [17]. The review has a number of limitations. We limited our review to papers published in the last 10 y because diagnosis of CP has

8 70 A.L.P.A. Lelis et al. / Sleep Medicine Reviews 30 (2016) 63e71 changed with the introduction of the GMFCS and more sophisticated imaging, and more children with extreme prematurity accounting for the prevalence of CP [47]. It is important to provide a review of the nature of sleep problems and factors associated with sleep problems for a variety of forms of CP not only to sensitize health professionals to potential sleep problems but also because there is some controversy about the traditional classifications of types of CP [47]. The paucity of relevant papers for the study purpose and low level of available evidence further emphasizes the importance of a broad integrative review. Our review highlights the ubiquitous presence of sleep disorders in children with CP, illuminates their nature, and describes relationships between children's sleep problems and comorbidities that commonly affect these children. The review also exposed extrinsic factors affecting children's sleep that cross types of CP, including socio-familial variables, and clinical-surgical and pharmacological interventions. Interactions among extrinsic and intrinsic factors might be viewed as cyclical. The combination of factors influencing children's sleep could exacerbate sleep disorders in children with CP due to multiple related factors in a child's health history. Practice points 1) The main sleep disorders in children with CP are difficult morning awakening, nightmares, difficulties in initiating and maintaining nighttime sleep, and sleep anxiety, which could be linked to behavioral sleep problems. 2) Sleep disorders in children with CP are associated with intrinsic comorbidities and extrinsic environmental factors. 3) The interaction among intrinsic and extrinsic factors in children's health histories can contribute to development of sleep disorders and increase their gravity. Research agenda 1) It is necessary to study associations between Gross Motor Function Classification System levels and sleep in large samples of children. 2) Measures specifically designed for children with neurodevelopmental disabilities need to be incorporated in sleep studies of children with CP. 3) Both effects of intrinsic pathophysiological factors associated with CP and extrinsic factors, such as family environments, should be taken into account when studying children's sleep disorders. Conflicts of interest The authors do not have any conflicts of interest to disclose. Support No financial support. Acknowledgments We acknowledge funding of a PhD-Scholarship for Ana Luíza Paula de Aguiar Lelis from CAPES-Program, Brazilian Ministry of Education. References [1] Owens J. Classification and epidemiology of childhood sleep disorders. Prim Care Clin Off Pract 2008;35:533e46. [2] Tietze AL, Blankenburg M, Hechler T, Michel E, Koh M, Schlüter B, et al. Sleep Med Rev 2012;16:117e27. [3] Jan JE, Freeman RD. Melatonin therapy for circadian rhythm sleep disorders in children with multiple disabilities: what have we learned in the last decade? Dev Med Child Neurol 2004;46:776e82. [4] Jan JE, Owens JA, Weiss MD, Johnson KP, Wasdell MB, Freeman RD, et al. Sleep hygiene for children with neurodevelopmental disabilities. Pediatrics 2008;122:1343e50. *[5] Newman C, O'Regan M, Hensey O. Sleep disorders in children with cerebral palsy. Dev Med Child Neurol 2006;48:564e8. [6] Bax M, Goldstein M, Rosenbaum P, Leviton A, Paneth N, Dan B, et al. Proposed definition and classification of cerebral palsy. Dev Med Child Neurol 2005;47: 571e6. [7] Odding E, Roebroeck ME, Stam HJ. The epidemiology of cerebral palsy: incidence, impairments and risk factors. Disabil Rehabil 2006;28:183e91. [8] Astill RG, Van der Heijden KB, Van IJzendoorn MH, Van Someren JW. Sleep, cognition, and behavioral problems in school-age children: a century of research meta-analyzed. Psychol Bull 2012;138:1109e38. [9] Dewald JF, Meijer AM, Oort FJ, Kerkhof GA, B ogels SM. The influence of sleep quality, sleep duration and sleepiness on school performance in children and adolescents: a meta-analytic review. Sleep Med Rev 2010;3:179e89. [10] Sadeh A. Consequences of sleep loss or sleep disruption in children. Sleep Med Clin 2007;2:513e20. [11] Sandella DE, O'Brien LM, Shank LK, Warschausky SA. Sleep and quality of life in children with cerebral palsy. Sleep Med 2011;12:252e6. [12] Whittemore R, Knafl K. The integrative review: updated methodology. J Adv Nurs 2005;52:546e53. [13] Whittemore R. Combining evidence in nursing research: methods and implications. Nurs Res 2005;54:56e62. [14] Souza MT, Silva MD, Carvalho R. Integrative review: what is it? how to do it? Einstein 2010;8:102e6. [15] Higgins JPT, Green S, editors. Cochrane handbook for systematic reviews of interventions. Version [updated March 2011]. Available from: Cochrane Collaboration. [16] Melnyk BM, Fineout-Overholt E. Evidence-based practice in nursing & healthcare: a guide to best practice. Philadelphia: Lippincot Williams & Wilkins; p. 3e24. *[17] Atmawidjaja RW, Wong SW, Yang WW, Ong LC. Sleep disturbances in Malaysian children with cerebral palsy. Dev Med Child Neurol 2014;56: 681e5. *[18] Wayte S, McCaughey E, Holley S, Annaz D, Hill CM. Sleep problems in children with cerebral palsy and their relationship with maternal sleep and depression. Acta Paediatr 2012;101:618e23. *[19] Romeo DM, Brogna C, Quintiliani M, Baranello G, Pagliano E, Casalino T, et al. Sleep disorders in children with cerebral palsy: neurodevelopmental and behavioral correlates. Sleep Med 2014;15:213e8. *[20] Elsayed RM, Hasanein BM, Sayyah HE, El-Auoty ME, Tharwat N, Belal TM. Sleep assessment of children with cerebral palsy: using validated sleep questionnaire. Ann Indian Acad Neurol 2013;16:62e5. *[21] Fitzgerald DA, Follett J, Van Asperen PP. Assessing and managing lung disease and sleep disordered breathing in children with cerebral palsy. Paediatr Respir Rev 2009;10:18e24. *[22] Hartzell LD, Guillory RM, Munson PD, Dunham AK, Bower CM, Richter GT. Tongue base suspension in children with cerebral palsy and obstructive sleep apnea. Int J Pediatr Otorhinolaryngol 2013;77:534e7. *[23] Simard-Tremblay E, Constantin E, Gruber R, Brouillette RT, Shevell M. Sleep in children with cerebral palsy: a review. J Child Neurol 2011;26:1303e10. [24] Seddon PC, Khan Y. Respiratory problems in children with neurological impairment. Arch Dis Child 2003;88:75e8. *[25] Singhi P, Sankhyan N. Co-morbidities of cerebral palsy and developmentally challenged children. Indian J Pract Pediatr 2012;14:420e6. *[26] Giannasi LC, Batista SRF, Matsui MY, Hardt CT, Gomes CP, Amorim JBO, et al. Effect of a hyperbolide mastication apparatus for the treatment of severe sleep bruxism in a child with cerebral palsy: long-term follow-up. J Bodyw Mov Ther 2014;18:62e7. [27] Mol EM, Monbaliu E, Ven M, Vergote M, Prinzie P. The use of night orthoses in cerebral palsy treatment: sleep disturbance in children and parental burden or not? Res Dev Disabil 2012;33:341e9. * The most important references are denoted by an asterisk.