A medullary tumor is a relatively rare disease that causes severe and complicated respiratory disorders, including sleep-related breathing disorders, due to dysregulation of respiratory control. A severely obese 12-year-old male was admitted to our hospital for worsening dyspnea and received a diagnosis of type II respiratory failure. Although obstructive sleep apnea (OSA) and/or obesity hypoventilation syndrome were suspected, a polysomnogram obtained during a nap (nap study) revealed central sleep apnea (CSA) and sleep-related hypoventilation disorder (SRHD) in addition to OSA. Brain magnetic resonance imaging showed a mass in the medulla oblongata. The patient received a diagnosis of CSA/SRHD caused by a medullary tumor, and with OSA. A partial brainstem tumor resection was performed. Noninvasive positive airway pressure therapy was initiated, and was continued after surgery. However, the patient died unexpectedly, 20 months after surgery. When children with sleep problems caused by OSA present with atypical symptoms of OSA, a neurological examination and polysomnography should be conducted to detect CSA and SRHD.
Fujimoto K, Kasai H, Kunii R, Terada J, Tatsumi K. Obstructive sleep apnea in a severely obese child with combined central sleep apnea and sleep-related hypoventilation disorder caused by a medullary tumor. J Clin Sleep Med. 2018;14(6):1071–1074.
A medullary tumor is a relatively rare disease that causes severe and complicated respiratory disorders such as central sleep apnea (CSA) and hypoventilation (especially during sleep) due to dysregulation of respiratory control.1 Brain tumors in children can cause obstructive sleep apnea (OSA) and CSA.2 However, few studies have reported an overlap between OSA and CSA. We present a case of OSA combined with CSA and sleep-related hypoventilation disorder (SRHD) caused by a medullary tumor in a severely obese child.
REPORT OF CASE
A 12-year-old male had presented with dyspnea that increased on exertion and general fatigue 3 years previously, and the symptoms gradually worsened. He was treated for asthma, but his symptoms did not improve. Because he was severely obese and suffered from snoring, OSA and/or obesity hypoventilation syndrome (OHS) were suspected. The patient was referred to our hospital for further evaluation.
On admission, the patient's symptoms worsened, and insomnia and frequent nighttime arousals developed, followed by headache and severe daytime sleepiness. A physical examination revealed an oxygen saturation of 94% in room air that decreased to 70% on exertion, a body mass index of 39 kg/m2, and tonsillar hypertrophy. A laboratory test revealed that the level of the brain natriuretic peptide was 134.7 pg/ mL. A chest radiograph showed cardiac enlargement, and an electrocardiogram showed sinus tachycardia and an inverted T wave in II, III, aVF, and V1−3. Transthoracic echocardiography showed enlargement of the right and left ventricles and a preserved left ventricle ejection fraction. The patient was considered to have congestive heart failure. An arterial blood gas analysis in room air showed respiratory failure (pH 7.33; PaO2, 53 mmHg; PaCO2, 65 mmHg). The patient's oxygen saturation decreased remarkably and continuously (SpO2 < 40%) during the first night of hospital admission, and noninvasive positive pressure ventilation (NPPV) was initiated. Standard overnight polysomnography (PSG) without NPPV support was deemed unsafe; thus, an abbreviated daytime baseline PSG was performed during a nap (nap study). The study lasted 3 hours, 12 minutes. The respiratory events recorded during sleep were scored according to The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications (AASM Scoring Manual) Version 2.1. An apnea-hypopnea index (AHI) of 136.5 events/h was revealed, with central hypopnea of 41.8 events/h and obstructive hypoventilation of 38.4 events/h (Table 1, Figure 1). The patient was almost 13 years old, his height was 157 cm, and his body weight was 97 kg. Consequently, the data were scored manually using both pediatric and adult criteria in the AASM Scoring Manual Version 2.3. There were no remarkable differences in the results of the two scoring systems. The results indicated CSA and central hypopnea in addition to OSA. After PSG, mild weakness was noted in the extremities during rehabilitation for weight reduction, although the patient was unaware of the symptom. A neurological examination showed normal cranial nerve function, increased deep tendon reflexes on the left side, and normal strength and proprioception in all extremities. The patient's parents stated that slight weakness in the extremities was present before admission. Based on the PSG and the aforementioned neurological symptoms, brain magnetic resonance imaging was performed; it revealed a cystic mass in the medulla oblongata that was isointense on T1-weighted magnetic resonance images and hyperintense on T2-weighted magnetic resonance images, and was spreading to the fourth ventricle and the upper cervical spinal cord (Figure 2). The patient received a diagnosis of CSA/SRHD caused by a medullary tumor and with OSA.
2-minute segment of PSG recording.
PSG was performed during a daytime nap and showed the presence of central hypopneas (A) and obstructive hypopneas (B). PSG = polysomnography.
2-minute segment of PSG recording.
Brain magnetic resonance imaging of the patient.
Note the cystic mass (red arrow), which is isointense on T1-weighted (T1WI) (A) and hyperintense on T2-weighted (T2WI) (B, C) magnetic resonance images of the medulla oblongata that is spreading to the fourth ventricle and the upper cervical spinal cord.
Brain magnetic resonance imaging of the patient.
A complete tumor resection from the brainstem was not possible; therefore, a small piece of tumor tissue was left in place to preserve brainstem function. The pathologic diagnosis was pilocytic astrocytoma. Following surgery, the patient was continuously treated with positive airway pressure (PAP) ventilation using various devices and settings. The respiratory symptoms (such as dyspnea) and symptoms related to sleep (such as daytime sleepiness) partially improved, although weakness of the extremities persisted. An arterial blood gas analysis in room air revealed improvement (on admission: pH 7.33; PaO2 53 mmHg; PaCO2 65 mmHg versus after surgery and the initiation of PAP therapy: pH 7.41; PaO2 96 mmHg; PaCO2 46 mmHg). The patient's adherence to PAP therapy was good in the year following surgery (% days used ≥ 4 hours: > 90%; median daily usage: 6 to 7 hours; days used: 6 to 7 d/wk). However, his adherence gradually decreased (% days used ≥ 4 hours: 30% to 53%; median daily usage: 6 hours; days used: 2 to 3 d/wk), and he showed slight daytime sleepiness without worsened respiratory symptoms. Although the brainstem tumor did not recur, the patient died suddenly during a nap, at home, 20 months after surgery, at the age of 14 years.
The current case includes notable clinical findings. CSA and SRHD caused by a medullary tumor can be the first manifestations without obvious neurological symptoms. In addition, as illustrated in this case, despite the probable OSA diagnosis supported by clinical findings, PSG should be performed in patients presenting with atypical symptoms of OSA, such as respiratory failure during the day and neurological symptoms, to avoid overlooking the potential involvement of CSA and SRHD.
Cancers of the central nervous system (CNS) can lead to sleep and/or respiratory disorders due to direct effects on brain structures including the brainstem and the hypothalamus, and due to impaired sleep function. Brain tumors constitute 26% of all cancers in children.3 Medullary tumors might be etiologies of CSA; however, the prevalence of sleep disorders in children with neoplasms involving the CNS remains unclear.4 The patient had several symptoms of sleep respiratory disorders, including snoring, insomnia, nighttime arousals, and daytime sleepiness, that were compatible with OSA. Other symptoms, including hypercapnic respiratory failure during the day (alveolar hypoventilation) and weakness of the extremities, were not compatible with OSA. The PSG indicated that CSA and SRHD overlapped with OSA. Finally, medullary tumor was diagnosed in this patient.
The patient was classified as having complicated combined SRHD according to the International Classification of Sleep Disorders, 3rd edition, which were the diagnostic criteria at the time of diagnosis. In addition to OSA and CSA detected by PSG, SRHD in this patient included combined OHS and sleep-related hypoventilation due to a physical disability (medullary tumor). He presented with sustained hypoxia and hypercapnia (worsening during sleep) with congestive heart failure and received NPPV. Additionally, because neurological symptoms except sleep-associated complaints were overlooked on admission, the patient was initially considered to have OSA or OHS due to severe obesity and type II respiratory failure. Rosen et al. studied sleep problems in 14 children with CNS neoplasms and reported that two children with tumors of the medulla oblongata presented with central apnea leading to respiratory insufficiency and required mechanical ventilation.4 Specifically, sleep disorders in children may be caused by CNS lesions. Thus, even in patients with suspected OSA based on clinical findings, a neurological examination and PSG should be performed to obtain an accurate clinical picture of patients presenting with atypical symptoms of OSA.
We herein present the case of a severely obese child with OSA in combination with CSA and SRHD caused by a medullary tumor, which was diagnosed using PSG. OSA, CSA, and SRHD can overlap in children with brainstem tumors. Therefore, a neurological examination and PSG are essential in children with sleep problems considered to be caused by OSA, because these tests can determine the potential involvement of CSA and SRHD. In addition, if a PSG shows atypical OSA with mixed CSA and SRHD, brain imaging should be performed to rule out a brain tumor.
Work for this study was performed at the Department of Respirology, Chiba University Hospital, Chiba, Japan. All authors have seen and approved this manuscript. The authors report no conflicts of interest.