Obstructive sleep apnoea syndrome: is the "half-night polysomnography" an adequate method for evaluating sleep profile and respiratory events?

Recently, to reduce the costs of polysomnography, split-night studies have been introduced into routine practice: the first part of the night is used to make the diagnosis of obstructive sleep apnoea syndrome (OSAS) and the second part to achieve an appropriate level of continuous positive airway pressure. Since this split-night protocol has not yet been validated by the comparison of polysomnographic pictures obtained in the first and second parts of the night, the aim of this study was to evaluate sleep profile and respiratory disturbances in the first part (PSG1) and second (PSG2) portion of a standard full-night polysomnographic examination (PSGtot) in a group of OSAS patients. Twenty nine consecutive OSAS patients, aged 54+/-10 yrs; body mass index (BMI) 40+/-6 kg x m(-2) (mean+/-SD values), were studied by separate analyses of PSG1, PSG2 and PSGtot. PSG1 was found to have a low sensitivity value (66%). A significant difference was found between apnoea-hypopnoea indices (AHI) recorded in PSG1, PSG2 and PSGtot (mean+/-SD, AHI1 33+/-27, AHI2 45+/-28, AHItot 40+/-25 events x h(-1), respectively; p<0.01). A strong correlation was observed between AHItot and AHI1 (r=0.89) and between AHItot and AHI2 (r=0.92), but a weaker correlation between AHI1 and AHI2 (r=0.66). These correlations became weaker when patients were subdivided into two different classes on the basis of disease severity. PSG1 was representative of PSGtot and similar to PSG2 only in those patients with rapid eye movement (REM) phase sleep in the first part of the night. We conclude that split-night protocols are not appropriate for evaluating sleep-disordered breathing in obstructive sleep apnoea syndrome patients when rapid eye movement phase sleep does not occur in the first part of the night.