Sleep-disordered breathing in spinal cord-injured patients: a short-term longitudinal study

Spinal cord-injured subjects were assessed during the acute admission for sleep-disordered breathing. Polysomnography demonstrated a high incidence of sleep apnoea that persisted during the acute phase. There was no correlation with respiratory function tests or measures of ventilatory control. Screening of this population is worthwhile although the clinical significance is uncertain.

Previous studies have demonstrated an increased incidence of sleep apnoea in spinal cord-injured patients. Many of these studies were performed in long-term, stable spinal cord injury (SCI). The aims of this study were: (i) to determine the prevalence of sleep-disordered breathing (SDB) in acute SCI; (ii) to document the change in SDB over time during the rehabilitation period; and (iii) to correlate the degree of SDB with ventilatory parameters.

Sixteen subjects with an acute SCI level T12 and above with complete motor impairment (American Spinal Injury Association impairment scale A or B) were recruited. Assessment, including polysomnography, respiratory function testing, and hypoxic and hypercapnic ventilatory responses, were performed 6-8 weeks post SCI, and repeated 6 months post SCI.

Eleven of 16 subjects (73%) had evidence of sleep apnoea, five of whom were moderate to severe. This high incidence persisted during the acute admission, with 9 of 12 subjects (75%) having sleep apnoea on polysomnography 20 weeks following injury. There was no correlation between the severity of SDB and other measures, such as level or completeness of injury, respiratory function tests or measures of ventilatory responses.

We have demonstrated a high incidence of sleep apnoea in the acute phase of SCI that persisted during the acute admission. Despite the high incidence of sleep apnoea, patients were relatively asymptomatic. Screening of this population would appear worthwhile given the high prevalence, although the significance of the sleep apnoea and clinical impact is not known.