[Value of oxygen saturation combined airflow in diagnosis of obstructive sleep apnea hypopnea syndrome].

OBJECTIVE: To assess the diagnostic value of oxygen saturation combined with airflow (OF) monitoring for diagnosis of obstructive sleep apnea hypopnea syndrome (OSAHS).

METHODS: Sixty-two subjects including suspected OSAHS and non-snorers underwent overnight polysomnography (PSG) and OF monitoring simultaneously in sleep laboratory. The apnea-hypopnea index (AHI), lowest oxygen saturation (LSaO(2)), and oxygen desaturation index (ODI) recorded by OF were compared with those recorded by PSG. The AHI and ODI data that showed skew distribution underwent square root transformation to approximate to normal distribution. Pair t test was used for difference hypothesis test. The agreement between the two measures was analyzed using Bland-Altman plot.

RESULTS: Forty-five of the 62 subjects were diagnosed as with OSAHS based on PSG with the mean AHI value of (40 +/- 27) events per hour. The mean AHI values derived from OF (OF-AHI) and PSG (PSG-AHI) according to total sleep time were (28 +/- 26) and (29 +/- 28) times per hour respectively, and those after square root transformation were (4.6 +/- 2.7) and (4.7 +/- 2.7) times per hour respectively (P = 0.08). The mean LSaO(2) derived from OF (OF-LSaO(2)) was (82 +/- 11)%, not significantly different from that derived from PSG (PSG-LSaO(2)) [(82 +/- 10)%, P = 0.65]. The ODI derived from OF (OF-ODI) after square root transformation was (3.0 +/- 2.4) times/h, significantly lower than that derived from PSG (PSG-ODI) after square root transformation [(4.0 +/- 2.9) times/h, P = 0.00]. The Bland-Altman plot revealed a good agreement between the OF-AHI and PSG-AHI in non-OSAHS people and patients with mild OSAHS (P = 0.28), however, the OF-AHI was lower than PSG-AHI (P = 0.00) in the patients with moderate to severe OSAHS. There was no significant difference between OF-LSaO(2) and PSG-LSaO2 (P = 0.65).

CONCLUSIONS: There is a good agreement between OF and PSG for AHI and LSaO(2). OF can be used to screen patients with suspected OSAHS in high risk population. However, OF tends to underestimate the AHI in moderate to severe OSAHS so additional manual analysis is necessary to confirm the diagnosis.