[Sensitivity and specificity of pulse oximetry in diagnosis of sleep-related respiratory disorders].
As there is a great demand for polysomnography, screening tests had to be established in the diagnosis of sleep-related disordered breathing. Oxygen saturation is often used as a main diagnostic criterion of these screening tests. In 207 patients (42 women, 165 men) suspected to suffer from a sleep apnea-syndrome we compared oxygen saturation (pulse oxymetry) with apnea-hypopnea-index (AHI, polysomnography). An AHI > or = 5/h and a minimal oxygen saturation < 90% and/or more than 5 oxygen desaturations/h > or = 4% were regarded as pathological. We found a good correlation of significant oxygen desaturations measured with pulse oxymetry and AHI (r = 0.85). Compared to AHI, pulse oxymetry was seen as a sensitive method (sensitivity 0.97) to recognize sleep apnea-patients. Coincidence of COPD decreased specificity of pulse oxymetry from 0.48 to 0.23 but had no influence on its sensitivity. 13 percent of the patients without further cardiorespiratory disease had a normal AHI (< 5/h) but a pathological pulse oxymetry. We conclude, that pulse oxymetry in combination with a positive case history is a sensitive method to recognize sleep apnea-patients. Not all sleep apnea-patients' can be diagnosed only by regarding AHI without taking oxygen saturation into account.