Sleep-disordered breathing in patients with Chiari malformation: Improvement after surgery

Arnold-Chiari malformations (ACMs) are congenital or acquired herniations of the cerebellum through the foramen magnum that may result in brainstem compression. Forty to 80% of ACMs are associated with syringomyelia.¹ The anatomic site of respiratory centers and their possible injury may explain the presence of sleep-disordered breathing in patients with ACMs. Improvement of central sleep apneas has been reported after surgical decompression in isolated cases.^2,3 In this study, we examined the occurrence of sleep apneas in patients with ACMs and syringomyelia and the impact of decompression surgery on sleep-disordered breathing.

Methods.

The subjects were recruited consecutively from the population of patients referred to the Neurosurgery Unit for symptoms of craniocervical malformation. All patients completed an individual sleep questionnaire about the usual length of sleep, apneas observed by a sleeping partner, difficulties waking up in the morning, chronic fatigue, morning headaches, waking during the night, snoring, and symptoms of restless legs. Subjects were asked whether they experienced sleepiness during the day and the frequency of this problem. The Epworth Sleepiness Scale (EES) was also completed.⁴ All patients underwent a full-night polysomnography (PSG). Standard nocturnal PSG was performed with recording of sleep stages (EEG, chin muscles EMG, electrooculography), measurements of oronasal airflow, snoring, respiratory movements, and oxygen saturation (SaO₂) with a finger pulse oximeter (Biox 3740, Ohmeda). Sleep stages were scored using standard criteria.⁵ Microarousals were scored according to American Sleep Disorders Association criteria.⁶ Apnea was defined as cessation of airflow for at least 10 seconds. Apnea was classified as obstructive in the presence of respiratory efforts throughout the apnea, central in the absence of respiratory efforts, and mixed when it started with central apnea and ended with obstructive apnea. Hypopnea was defined as a 30% reduction of airflow or respiratory movements accompanied by a 3% decrease in oxygen saturation or followed by an arousal. Sleep apnea syndrome was defined as an apnea-hypopnea index (AHI) >10. A second PSG was proposed after decompression surgery in patients in whom sleep apnea syndrome was diagnosed on the preoperative recording.

Results.

Sixteen subjects (including seven women) were enrolled in the study (table); their mean age was 38.1 ± 3.9 years and their mean body mass index was 25.9 ± 1.1 kg/m². All patients presented symptoms of craniocervical disorders and had ACM type I with syringomyelia diagnosed on MRI.¹ Forty-three percent of subjects reported snoring and 81.3% had excessive daytime sleepiness. The mean ESS score was 9.1 ± 1.3 and was ≥10 in 50% of patients. For the whole population, the mean AHI was 36.6 ± 7.7 and AHI was ≥10 in 12 of 16 patients (75%). There was no significant correlation between scores on the ESS and AHI. The mean apnea index was 13.1 ± 4.1. On average, 48% of apneas were classified as central apneas. The mean microarousal index was 33.0 ± 4.4.

Eight of the 12 patients with sleep apnea syndrome underwent surgical decompression. Postoperative PSG was performed in six of eight patients an average of 203 (127 to 313) days after surgery. One patient was lost to follow-up. Postoperative PSG was not performed in Patient 14 whose AHI was only 11 before surgery. Postoperative PSG showed a decrease in AHI from 56.5 ± 11.5 to 37.2 ± 15 (p = 0.1), a decrease in apnea index from 23.5 ± 7.9 to 9.8 ± 6.6 (p = 0.1), and a decrease in central apnea index from 14.9 ± 5.5 to 1.3 ± 0.6 (p = 0.03) (figure). The microarousal index also improved after surgery from 37.2 ± 13.7 to 26.2 ± 17.0 (p = 0.03). On average, there was no significant change in ESS score after surgery (9.7 ± 2.2 before and 10.7 ± 2.4 after).

• Download figure
• Open in new tab
• Download powerpoint

Figure. Polysomnographic data before and after decompression surgery in six patients with Arnold-Chiari malformation and sleep apnea syndrome.

Discussion.

Our study showed a high frequency of sleep apnea syndrome in a group of 16 unselected patients with ACM type I and syringomyelia. About 50% of the sleep apneas observed in our patients were classified as central apneas. Surgical decompression was associated with a significant reduction of the number of central apneas and microarousal.

In a previous study of nine patients with ACM type I, sleep apnea syndrome, defined by an AHI >10, was diagnosed in 55% of cases.⁷ In a second series, the same authors detected sleep apnea syndrome (AHI >5) in 60% of patients with isolated ACM type I, 44% of patients with ACM type I associated with syringomyelia and 12% of control subjects.⁸ The higher frequency of sleep apnea syndrome (75% for AHI >10 and 87% for AHI >5) observed in our study can be explained by the fact that we only included patients with ACMs associated with syringomyelia. The transducer used to detect sleep-disordered breathing in our study, i.e., the nasal cannula, was also more sensitive than the thermistances used in previous studies.^7,8

About 50% of sleep apneas in our series were classified as central apneas on the basis of two signals, thoracoabdominal movements and suprasternal pressure, which have been shown to be an accurate method for apnea classification.⁹ We decided not to classify hypopneas in the absence of measurement of respiratory effort by esophageal pressure. There are several hypotheses addressing the possible mechanisms of central sleep apnea in patients with ACMs. Physical compression of the brainstem might be responsible for the depression of the respiratory center or reticular activating system. An abnormal chemosensitivity may be due to damaged peripheral chemoreceptors or stretching of lower cranial nerves that carry afferent traffic from the carotid bodies to the medulla.¹⁰ Medullar compression may also compromise blood flow to the medulla therefore delaying feedback to the respiratory chemoreceptors.

Previous publications based on isolated cases^2,3 reported postoperative improvement of sleep apnea syndrome with a risk of secondary deterioration.³ In our study, the postoperative follow-up was performed at least 4 months after surgery in order to take this risk into account. Improvement of sleep apnea syndrome mainly concerned the central apnea index and was accompanied by a reduction of the microarousal index. Despite this reduction in the mean sleep fragmentation index, there was no improvement of sleepiness as assessed by the ESS. The lack of improvement in sleepiness was explained by the persistence of a severe obstructive sleep apnea syndrome in two patients (Patients 7 and 9) who were subsequently treated by continuous positive airway pressure. In one patient (Patient 2) whose AHI was normalized after surgery, sleepiness worsening was related to a depressive syndrome. Patients with mild postoperative sleep apnea syndrome (Patients 10 and 12) had no subsequent treatment.

Our results suggest that sleep-disordered breathing should be investigated in patients with ACMs associated with a complaint of daytime sleepiness or an ESS score >10.