SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. References

Aim  In female children with drug-resistant seizures and developmental delay from birth, atypical Rett syndrome caused by mutations in the CDKL5 gene should be considered. Several clinical features resemble classic Rett syndrome. Respiratory and sleep abnormalities are frequently present in Rett syndrome, whereas little is known in patients with CDKL5 mutations.

Method  In four genetically confirmed female patients with CDKL5 mutations (age range 2–15y), the presence of breathing and sleep abnormalities was evaluated using the validated Sleep Disturbance Scale for Children and polysomnography (PSG).

Results  The Sleep Disturbance Scale for Children indicated disorders of initiating and maintaining sleep, daytime somnolence, and sleep breathing disorders. In one patient, PSG showed central apnoeas during sleep: her total apnoea–hypopnoea index (AHI) was 4.9, of which the central AHI was 3.4/h. When awake, central apnoeas were present in two of the four female children (central AHI 28/h and 41/h respectively), all preceded by hyperventilation. PSG showed low rapid eye movement (REM) sleep (9.7–18.3%), frequent awakenings, and low sleep efficiency (range 59–78%).

Interpretation  Episodic hyperventilation followed by central apnoeas was present while awake in two of four patients. This may indicate failure of brainstem respiratory centres. In addition, low REM sleep, frequent arousals (not caused by apnoeas/seizures), and low sleep efficiency were present. Similar to Rett syndrome, in patients with CDKL5 mutations PSG seems warranted to evaluate breathing and sleep disturbances.


Abbreviations
AHI

Apnoea-hypopnoea index

cAHI

Central apnoea index

CDKL5

Cyclin-dependent kinase-like 5

MECP2

Methyl-CpG binding protein 2

PSG

Polysomnography

What this paper adds

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. References
  •  Sleep and breathing disturbances are present in patients with CDKL5 mutations.
  •  Polysomnography revealed central apnoeas when awake, low rapid eye movement sleep, frequent awakenings, and low sleep efficiency.

Rett syndrome is a genetic neurodevelopmental disorder that mainly affects female children. The clinical diagnosis is based on the presence of regression in combination with the loss of purposeful hand skills, motor and communication function, cognitive impairment, the development of stereotypic hand function, and seizures.1 In addition, breathing disturbances such as hypoventilation, central apnoea, episodic hyperventilation, and breath holding may be present. They occur mainly when awake, and can be life-threatening.2 In 97% of patients with Rett syndrome a dominant mutation in the gene encoding methyl-CpG binding protein 2 (MECP2) on the X chromosome is found, which enhances or activates gene expression.3

In female children with drug-resistant seizures in the first year of life and severe psychomotor delay from birth, atypical Rett syndrome caused by a mutation in the cyclin-dependent kinase-like 5 (CDKL5) gene should be considered.4 Clinically the patients resemble those with classic Rett syndrome, and they have impairment of purposeful hand function, cannot sit without support or walk, lack expressive language, and demonstrate autistic-like features. Neurological examination reveals trunk hypotonia and deceleration of head growth.5 In contrast to classic Rett syndrome, female children with CDKL5 mutations experience no regression and the delay of psychomotor development is present from birth onwards.1 Furthermore, in female children with CDKL5 mutations, frequent daily epileptic seizures develop in the first 2 months of life, in contrast to classic Rett syndrome in which seizures develop at after 2 years of age. Seizures consist of generalized tonic seizures with flushing of the face and sometimes tonic seizures followed by a clonic phase. Epileptic encephalopathy with infantile spasms and hypsarrhythmia often develop. Seizures may diminish over time, but often daily tonic, myoclonic, and atypical absence seizures persist.5 Recently, a direct interaction was demonstrated between MECP2 and the CDKL5 gene, i.e. CDKL5 is a novel MECP2-repressed target gene.6 Therefore, it is likely that clinical symptoms overlap and that the well-known respiratory and sleep disturbances in patients with MECP2 are also present in those with CDKL5 mutations. However, because no data are available, this study examines the presence of breathing and sleep abnormalities in a small series of patients with CDKL5 mutations.

Method

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. References

We investigated four female children with CDKL5 mutations attending the Sleep Center of the Stichting Epilepsie Instellingen in the Netherlands because of sleep disturbances. As the protocol was similar to that used in all cases with clinical suspicion of a sleep disorder, it was not necessary to consult the medical ethical committee. Parents gave written informed consent for participation as well as permission to publish the results.

All four patients had a severe developmental delay and experienced seizures. The validated Sleep Disturbance Scale for Children (Dutch version), a parental questionnaire, was used to evaluate sleep and night-time breathing disorders. Questions are answered on a Likert-type scale, with higher values reflecting greater clinical severity: 1, never; 2, occasionally; 3, sometimes; 4, often (three to five times a week); 5, always. The maximum score for all questions was 100, with a total score of 70 or more considered clinically relevant.7 Polysomnography (PSG) recordings were performed overnight, and during a few hours in the awake state during the day, using the guidelines of the American Association of Sleep Medicine for recording sleep/snoring in children.8 The following measurements were evaluated: snoring, air flow (thermistor and nasal pressure) and respiratory effort, oxygen saturation (SpO2), heart rate frequency, and transcutaneous partial pressure of carbon dioxide (pCO2; two patients). Video-electroencephalography, electro-oculography, and electromyography provided data for the assessment of sleep and epileptic discharges, and indicated the presence of seizures. Apnoeas were defined as a 90% reduction in the amplitude of airflow with a duration of at least two breaths. In central apnoea, there was no airflow and no abdominal–thoracic movement combined with at least 3% desaturation, or it was defined as an apnoea lasting at least 20 seconds. An obstructive apnoea was defined as no airflow, despite chest wall and abdominal movements. In mixed central–obstructive apnoea both phenomena were present. A hypopnoea was defined as a 50% reduction of breathing amplitude for at least two breaths combined with a decrease of 3% in SpO2. The apnoea–hypopnoea index (AHI) is the mean total number of apnoeas and hypopnoeas per hour of sleep. An AHI of 1 or more is regarded as pathological. Based on our normal values this threshold is 3/hour.

Breathing disorders in patients with classic Rett syndrome often occur in the awake state. A central apnoea index (cAHI) is calculated while awake (central apnoeas/total recording time minus total sleep time). SpO2 was continuously measured using pulse oximetry. Transcutaneous pCO2 was considered to be abnormal when >50mmHg and <35 mmHg. Sleep efficiency (total sleep time/time lights off ×100), percentage of sleep stages, sleep onset latency (duration from light off to sleep onset), and number of awakenings were evaluated.

Results

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. References

The four patients with CDKL5 mutations had a mean age of 6 years 6 months (range 2–15y). Their weight was average or below average for age (range −2SD to 0SD) whereas their height was far below average for age (range −3SD to −1SD). Body mass index (BMI) was normal in three of the children, whereas the eldest patient had a lower BMI of 13.3. Table I presents the patients’ clinical characteristics, genetic mutation, and PSG data.

Table I. Clinical characteristics and respiratory data
PatientAge (y)MutationAEDsSeizuresAHI (sleep)SpO2 (lowest %) (sleep)Central apnoea total (awake)Wake (h)cAHI (apnoea/awake)Central apnoea mean (s; shortest–longest)Snoring (min) (%)
  1. AED, antiepileptic drug; VGB, vigabatrine; LEV, levetiracetam; LTG, lamotrigine; RFN, rufinamide; OXC, oxcarbamazepine; CLB, clobazam; CBZ, carbamazepine; VPA, valproic acid; Ket, ketogenic diet; t, tonic seizures; m, myoclonic seizures; cp, complex partial seizures; AHI, apnoea–hypopnoea index during sleep; cAHI, central apnoea–hypopnoea index in the awake state (total number of apnoeas/time awake); SpO2, mean oxygen saturation; NA, not available.

12c.656A>CVGB, LEV, LTG, RFNYes, daily; t, m4.996 (87)3588.7417.1 (3.9–15.6)73.5 (16.6)
24c.660_664dupOXC, CLB, LEV, KetYes, daily; t, cp0.5NA36.70.48.3 (6.7–10.6)0.0
35c.283-3_290delVPA, CBZ, CLB, KetYes, daily; t0.496 (92)34.70.612.4 (7.6–17.7)0.0
415c.2635_2636delVPA, CBZYes, monthly; t0.196.9 (93)1485.32810.9 (5.2–38.1)38.4 (6.7)

Neurological investigation showed severe psychomotor developmental delay. All the younger female children had cerebral visual impairment to some extent, such as low vision and diminished eye contact, intermittent nystagmus, and squinting. In addition, they had profound trunk hypotonia, could not sit without support or walk, and had hand stereotypies only and no purposeful hand movements. The 15-year-old child had a milder phenotype; she could walk, had some voluntary hand movements, good eye contact, used a single word, and seizures were less frequent and severe. The younger female children had daily tonic seizures. All used multiple antiepileptic drugs. Two of the children were on a ketogenic diet, besides medication for gastroesophageal reflux, oral laxatives, and (patient 4) promethazine. Patient 2 used melatonin, but not during the PSG recordings.

The history of the patients revealed sleeping disturbances in all the children, consisting mainly of insomnia and frequent awakenings. The Sleep Disturbance Scale for Children questionnaire confirmed the presence of different sleep disorders (patients 1–3; Table II). Most parents reported disorders of initiating and maintaining sleep, sleep–wake transition disorders, excessive daytime somnolence, and (in patient 3 only) sleep breathing disorders. In patient 4 the Sleep Disturbance Scale for Children scores were not (significantly) abnormal. PSG was started in the afternoon (while awake) and continued until the following morning. The recording time in the sleep state ranged from 916 to 972 minutes and in the awake state from 447 to 471 minutes. PSG showed severely delayed sleep onset in one female child only (patient 3; sleep onset 133min). Remarkably, the percentage of rapid eye movement (REM) sleep was low in all our patients with CDKL5 mutations (9.7–18.3%) and absent in one child. In all patients the sleep efficiency was low (median range 59–78%), caused by frequent and long-lasting awakenings (median range 7–52 times during the night). Although interictal epileptiform activity was abundant, these awakenings were not caused by seizures.

Table II. Sleep characteristics: Sleep Disturbance Scale for Children and polysomnographic data
PatientAge (y)Sleep onset (min)Stage 2 (min) (% TST)SWS (min) (% TST)REM latency (min)REM sleep (min) (% TST)Number of awakeningsSleep efficiency (%)Sleep Disturbance Scale for Children
  1. Sleep onset, time before any sleep stage was reached after lights out. TST, total sleep time; SWS, slow-wave sleep; REM, rapid eye movement sleep; Sleep Disturbance Scale for Children, total score ≥70 significant (bold type); DIMS, disorders of initiating and maintaining sleep (maximum score 26, significant >17); SWTD, sleep–wake transition disorders (maximum score >21, significant >14); DOES, disorders of excessive somnolence (maximum score 20, significant >13); SBD, sleep breathing disorders (maximum score 15, significant >7).

125.8180.5 (40.4)214.5 (48)39.543.5 (9.7)759.356 (DIMS 23, SWTD 12, DOES 8)
241.3479.5 (86.7)68.0 (12.3)No REMNo REM2678 71 (DIMS 17, SWTD 22, DOES 17)
35133.3209.0 (44)167.0 (35.5)150.581.0 (17.2)106260 (DIMS 18, SWTD 15, SBD 10)
41514.8315.5 (52.9)117.5 (19.7)136.5109.0 (18.3)528338 (DIMS 14, SWTD 6, DOES 10)

During sleep, only the youngest child (patient 1, AHI 4.9/h) has some obstructive apnoeas (oAHI 0.5/h) not clinically relevant, and some central apnoeas (cAHI 3.4/h). Some of these central apnoeas occurred during REM sleep and were considered physiological. In the awake state, central apnoeas were more prominently present in two of the four patients (cAHI 28/h and 41/h respectively), and most frequent in the youngest child (patient 1). The mean duration of these central apnoeas was longer in the oldest child (patient 4), with a mean of 10.9 seconds (compared with patient 2 with a mean of 7.1s); the longest central apnoea in patient 4 lasted for 38 seconds. The PSGs showed that a period of agitation, increase of hand stereotypies, and hyperventilation preceding the apnoeas (Fig. 1). Unfortunately transcutaneous pCO2 measurements were successfully performed in only one child (patient 2).

image

Figure 1.  A 3-minute polysomnographic recording at night, during the awake state. Eye movements (electro-oculograpjy [EOG]; channel 1), muscle activity on the chin (electromyography [EMG]; channel 2), and video-electroencephalography (EEG; channel 3) confirm that this female child is awake. Simultaneous with stereotypic hand movements and agitation, central apnoeas (duration 7–16.23s) occur. 1, EOG (eye movements); 2, EMG at the chin; 3, EEG (six channels); owing to muscle artefacts, for better visualization three left-side channels are shown; 4, EMG for detecting movement of the legs; 5, nasal pressure (nasal air flow [pressure measurements between inspiration and expiration]); 6, thermistor (air flow [temperature measurements between inspiration and expiration]); 7, Flow_Cu (computer calculation of signals 5, 6, 8 to detect flattening of airflow); 8 (top), thorax movements; 8 (bottom), abdominal movements; 9, pulse (heart frequency/min); 10, SpO2 (oxygen saturation).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. References

In two of our four patients with CDKL5 mutations, respiratory disturbances (all of the central type) were present during the awake state and were preceded by frequent episodic hyperventilation (cAHI 28 and cAHI 41 respectively). Although formal pCO2 measurements failed in three of the four patients, we postulate that these apnoeas were probably of the hypocapnic type and indicate failure of the respiratory feedback loop and central regulatory nuclei in the brainstem.

Since 2003, about 85 female children with atypical Rett syndrome and a CDKL5 gene mutation have been described, characterized by severe intellectual disability and developmental delay, intractable seizures, pronounced hypotonia, and hand stereotypies.4,5 The presence of hyperventilation was mentioned in only one of the first clinical surveys.5 A recent study in 10 patients with CDKL5 mutations demonstrated the presence of cardiorespiratory dysrhythmias such as tachypnoea (11.7%), deep breathing (4%), and apnoeas (1.4%) and breath holding (3.25%).9

Figure 1 shows that in a time window of 3 minutes, seven periods of central apnoeas occurred, all preceded by a period of hyperventilation. This is not compatible with breath holding as an explanation of the phenomena in our patients.

In classic Rett syndrome, PSG studies revealed the presence of hyperventilation in the awake state, often followed by prolonged apnoeas, which result in hypoxaemia.10 Three groups of daytime respiratory disturbances were identified: the forceful type (breath-holding and air expulsion), feeble (superficial breathing), and apneustic breathers.11 Differentiating between these breathing abnormalities is important, especially because the feeble breathers do not tolerate opiates.11 Central apnoeas can be treated with acetazolamide,12 nasal oxygen therapy,13 and positive air pressure therapy.14

Periods of agitation and increase in hand stereotypies, and (at the same time) episodic hyperventilation, preceded the occurrence of central apnoeas. The breathing disorder diminishes when the hands are fixated. Although behavioural factors might influence the occurrence of the breathing disorder, clinical studies in classic Rett syndrome have shown dysphagia and cardiorespiratory irregularities during the day and night, also suggesting brainstem dysfunction.2 Thus, there is evidence that brainstem dysfunction is also the pathogenic mechanism behind the respiratory disturbances. In our four patients with CDKL5 mutations, the parents reported sleep disturbances, mainly disorders of initiating and maintaining sleep and excessive daytime napping, but no report of night-time emotional behaviour, which is typical for classic Rett syndrome.

In patients with classic Rett syndrome, sleep questionnaires revealed sleep problems in at least 80%.15 They consist of irregular sleep/wake patterns, excessive daytime sleep, and night-time emotional behaviour such as screaming, crying, and laughing.15 Sleep diaries (7 × 24h), showed increased total night-time sleep and excessive daytime sleep, but no decrease with age.16 PSG revealed a decreased REM sleep percentage.17 This was also found in all our patients with CDKL5 mutations. We cannot explain this low percentage of REM sleep, because this is not typical of children with a developmental delay. In addition, frequent arousals (not caused by apnoeas or seizures) and low sleep efficiency were present. This is similar to classic Rett syndrome.17

Epilepsy in patients with CDKL5 mutations is often more severe than in those with classic Rett syndrome. Electroencephalography showed frequent epileptic discharges in all and sometimes tonic seizures. Having seizures is known to increase total sleep time and daytime sleep.16

Some limitations of our study need discussing. Because the pCO2 measurement failed, we were unable to prove our hypothesis formally. Also, owing to the rarity of this disorder, we could evaluate only four patients with CDKL5 mutations. Finally, the follow-up evaluation of these children is still in progress.

Patients with Rett syndrome and those with CDKL5 mutations show similar sleep and breathing features. All our patients with CDKL5 mutations showed disturbed sleep characterized by frequent awakenings, low sleep efficiency, and decrease in REM sleep. During the night in our youngest patient with CDKL5 mutations, obstructive sleep apnoeas (not clinically relevant) and some central apnoeas were present. Two of our patients had central apnoeas during the awake state. In classic Rett syndrome, respiratory evaluation is common practice. We suggest that assessment of sleep and breathing during the day and night is also warranted in patients with CDKL5 mutations.

References

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. References
  • 1
    Neul JL, Kaufmann WE, Glaze DG, et al.Rett syndrome: revised diagnostic criteria and nomenclature. Ann Neurol2010; 68: 94450.
  • 2
    Rohdin M, Fernell E, Eriksonn M, Albage M, Langercrantz H, Katz-Salamon M. Disturbances in cardiorespiratory function during day and night in Rett syndrome. Pediatr Neurol2007; 37: 33844.
  • 3
    Neul JL, Fang P, Barrish J, et al.Specific mutations in Methyl-CpG-Binding Protein 2 confer different severity in Rett syndrome. Neurology2008; 70: 131321.
  • 4
    Weaving LS, Christodoulou J, Williamson SL, et al.Mutations in CDKL5 cause a severe neurodevelopmental disorder with infantile spasms and mental retardation. Am J Hum Genet2004; 75: 107993.
  • 5
    Bahi-Buisson N, Nectoux J, Rosas-Vargas H, et al.Key clinical features to identify girls with CDKL5 mutations. Brain2008; 131: 264761.
  • 6
    Carouge D, Host L, Aunis D, Zwiller J, Anglard P. CDKL5 is a brain MeCP2 target gene regulated by DNA methylation. Neurobiol Dis2010; 38: 41424.
  • 7
    Bruni O, Ottaviano S, Guidetti V, et al.The Sleep Disturbance Scale for Children (SDCS). Construction and validation of an instrument to evaluate sleep disturbances in childhood and adolescence. J Sleep Res1996; 5: 25161.
  • 8
    Iber C, Ancoli-Israel S, Chesson AL Jr, Quan SF. The American Academy of Sleep Medicine (AASM) Manual for the Scoring of Sleep and Associated Events, 1st edn. Westchester, IL: American Academy of Sleep Medicine, 2007: 489.
  • 9
    Pini G, Bigoni S, Engerström IW, et al.Variant of Rett syndrome and CDKL5 gene: clinical and autonomic description of 10 cases. Neuropediatrics2012; 43: 3743.
  • 10
    Southall DP, Kerr AM, Tirosh E, Amos P, Lang MH, Stephenson JB. Hyperventilation in the awake state: potentially treatable component of Rett syndrome. Arch Dis Child1988; 63: 103948.
  • 11
    Julu PO, Kerr AM, Hansen S, Apartopoulus F, Jamal GA. Immaturity of medullary cardiorespiratory neurones leading to inappropriate autonomic reactions as a likely cause of sudden death in Rett’s syndrome. Arch Dis Child1997; 77: 4645.
  • 12
    Peña F, Garcia O. Breathing generation and potential pharmacotherapeutic approaches to central respiratory disorders. Curr Med Chem2006; 13: 268193.
  • 13
    Javaheri S, Ahmed M, Parker TJ, Brown CR. Effects of nasal O2 on sleep-related disordered breathing in ambulatory patients with stable heart failure. Sleep1999; 22: 11016.
  • 14
    McNamara F, Harris MA, Sullivan CE. Effects of nasal continuous positive airway pressure on apnoea index and sleep in infants. J Paediatr Child Health1995; 31: 8894.
  • 15
    Young D, Nagarajan L, de Klerk N, Jacoby P, Ellaway C, Leonard H. Sleep problems in Rett syndrome. Brain Dev2007; 29: 60916.
  • 16
    Ellaway C, Peat J, Loenard H, Christodoulou J. Sleep dysfunction in Rett syndrome: lack of age related decrease in sleep duration. Brain Dev2001; 23: S1013.
  • 17
    Caratenuto M, Esposito M, D’Aniello A, et al.Polysomnographic findings in Rett syndrome; a case-control study. Sleep Breath2012; Mar 7. (epub ahead of print).