Validation of British Thoracic Society guidelines for the diagnosis of the sleep apnoea/hypopnoea syndrome: can polysomnography be avoided?

BACKGROUND The British Thoracic Society report on the diagnosis and treatment of the sleep apnoea/hypopnoea syndrome (SAHS) suggests that, if the pulse oximetry baseline oxygen saturation is above 90%, then 15 4% oxygen desaturation/hour in bed will diagnose SAHS requiring treatment. The diagnostic outcome of applying these guidelines has been studied. METHODS One hundred patients referred to a district general hospital sleep clinic were recruited. After initial clinical assessment, overnight pulse oximetry measurements were performed, followed by full polysomnography at the regional laboratory. RESULTS Sixty nine patients underwent both pulse oximetry and polysomnography. All 10 patients with more than 15 4% desaturations/hour on pulse oximetry had SAHS confirmed on polysomnography (specificity = 100%). Twenty two patients with SAHS were misdiagnosed using pulse oximetry alone (sensitivity = 31%). These patients had low apnoea scores but high hypopnoea scores. CONCLUSIONS The BTS pulse oximetry criteria are highly specific when positive (specificity = 100%), but may miss patients with the SAHS who have hypopnoeic episodes which cause arousal but not significant oxygen desaturation (sensitivity = 31%). It should be emphasised that pulse oximetry alone does not always give sufficient information to discriminate between those patients with or without SAHS. Patients with "negative" pulse oximetry and symptoms of SAHS should undergo polysomnography or multi-channel recording.