Objective: To construct an algorithm for the diagnosis of sleep apnea/hypopnea syndrome (SAHS) based on the optimized sequential use of nocturnal pulse oximetry and respiratory polygraphy (AutoSet, AS) in patients suspected of SAHS.

Methods: Simultaneous performance of pulse oximetry and AS in 145 patients highly suspected of SAHS (at least two of the following signs or symptoms: chronic snoring, observations of repeated apneic events, and abnormal daytime drowsiness). SAHS was diagnosed when the apnea/hypopnea index (AHI) was greater than or equal to 10, based on AS findings. The diagnostic algorithm was obtained by calculating the prevalence of SAHS, calculating the best diagnostic cutoff point for a desaturation index greater than or equal to 4% (ID4%) based on the AHI; analyzing the positive predictive value (PPV) and negative predictive value (NPV) for the cutoff and the correlation and agreement between ID4% and the AHI depending on the presence and severity of SAHS.

Results: The overall prevalence of SAHS was 83.4%. The best cutoff point was obtained using ID4% greater than or equal to 10 (sensitivity 82%; specificity 83%). The PPV and NPV for the prevalence calculated were 0.95 and 0.29, respectively. The correlation between ID4% and AHI was 0.84; overall agreement was 0.93. For patients who did not have SAHS, the PPV was 0.49, and the NPV was 0.35; both gradually increased in cases of mild, moderate and severe SAHS (to 0.90 and 0.95, respectively, for severe SAHS). Using these data we created a diagnostic algorithm according to which an ID4% greater than or equal to 30 for nocturnal pulse oximetry (sensitivity 72%, specificity 100%) would indicate that AS testing would be unnecessary. An ID4% less than 30, on the other hand, would lead to further testing (AS, or polysomnography in the event of a negative AS), based on the high prevalence of SAHS in our series.

Conclusions: The sequential use of nocturnal pulse oximetry and AS allows substantial savings of polysomnographic testing in groups at high risk of SAHS.