Ahmadi Negar, Shapiro Gilla K, Chung Sharon A, Shapiro Colin M
Sleep Research Unit, Toronto Western Research Institute, University Health Network, ON, Canada.
Sleep Breath. 2009 Aug;13(3):221-6. doi: 10.1007/s11325-008-0234-2. Epub 2008 Dec 9.
The purpose of this study was to investigate apnea-hypopnea index (AHI) across two polysomnographies (PSGs) to examine AHI variability and impact on clinical diagnosis.
Two-night PSGs of 193 sleep clinic patients were reviewed, and the AHI variability was analyzed. Anonymized records from five patients with significant night-to-night AHI variability were used in this study: the two-night PSGs from two patients were represented as four individual PSGs; the two-night PSG for two others were represented as being obtained from two different sleep clinics; the last patient's PSG was shown as a two-night study. Twenty-two sleep experts attending the Associated Professional Sleep Societies meeting were recruited to make diagnoses based on the PSGs. They were told that the PSGs were from seven patients: four with single-night PSG; two with two PSGs, each one from a different clinic; and one patient with a two-night PSG.
Twenty-one percent of the 193 sleep clinic patients had a nightly PSG AHI variability of greater than 5. Forty-eight percent of all patients had a significantly higher AHI on the first night, and 41% had a significantly higher AHI on the second night. Using an AHI > 15 diagnostic criteria, sleep apnea would have been undetected in 20% (n = 39) of patients due to low AHI on one night. Furthermore, 13% of all patients had a more severe sleep apnea classification based on the second night of PSG. For the seven cases, 27-36% of sleep experts failed to identify sleep apnea especially when presented with the PSG containing the lower AHI. Incidences of missed sleep apnea diagnoses were reduced to 15-18% when information from two PSGs was presented to the sleep experts.
Utilizing a large patient population, this study supports the significant night-to-night variability in PSG respiratory variables. Identification of sleep apnea in some patients is reduced when sleep experts are provided with only one PSG recording. The clinical implication is that about 13% of sleep clinic patients might benefit from a second night of PSG.
本研究旨在通过两次多导睡眠图（PSG）检查来调查呼吸暂停低通气指数（AHI），以检验AHI的变异性及其对临床诊断的影响。
回顾了193例睡眠诊所患者的两晚PSG检查结果，并分析了AHI的变异性。本研究使用了五例夜间AHI变异性显著的患者的匿名记录：两名患者的两晚PSG被表示为四个独立的PSG；另外两名患者的两晚PSG被表示为来自两个不同的睡眠诊所；最后一名患者的PSG显示为两晚的研究。招募了22名参加联合专业睡眠协会会议的睡眠专家，根据PSG进行诊断。他们被告知这些PSG来自七名患者：四名患者有一晚的PSG记录；两名患者有两晚的PSG记录，且每晚的记录来自不同的诊所；一名患者有两晚的PSG记录。
193例睡眠诊所患者中，21%的患者每晚PSG的AHI变异性大于5。所有患者中，48%的患者第一晚的AHI显著更高，41%的患者第二晚的AHI显著更高。采用AHI>15的诊断标准，由于某一晚AHI较低，20%（n = 39）的患者的睡眠呼吸暂停可能未被检测到。此外，所有患者中，13%的患者根据第二晚的PSG检查结果被归类为更严重的睡眠呼吸暂停。对于这七例患者，27% - 36%的睡眠专家未能识别出睡眠呼吸暂停，尤其是当呈现AHI较低的PSG时。当向睡眠专家提供两晚PSG的信息时，漏诊睡眠呼吸暂停的发生率降至15% - 18%。
本研究纳入了大量患者，支持了PSG呼吸变量存在显著的夜间变异性这一观点。当睡眠专家仅获得一份PSG记录时，部分患者的睡眠呼吸暂停识别率会降低。临床意义在于，约13%的睡眠诊所患者可能会从第二晚的PSG检查中获益。
