Suppr超能文献

[通过呼吸音记录和多导睡眠图检测呼吸暂停 - 一项比较研究]

[Apnea Detection by Means of Respiratory Sound Recordings and Polysomnography - A Comparative Study].

作者信息

von Heydebrand M, Hildebrandt O, Cassel W, Schäfer A, Kesper K, Weissflog A, Taher N, Sohrabi K, Gross V, Koehler U

机构信息

Klinik für Innere Medizin, SP Pneumologie, Intensiv- und Schlafmedizin, Philipps-Universität, Marburg.

Thora Tech GmbH, Gießen.

出版信息

Pneumologie. 2017 Sep;71(9):594-599. doi: 10.1055/s-0043-112888. Epub 2017 Jul 31.

Abstract

In this study, recognition of apnoeas by means of polysomnography (PSG) and nocturnal respiratory sound recordings in patients with obstructive sleep apnoea (OSA) was analyzed and compared.  In 45 patients with OSA requiring treatment (AHI > 15/h), concomitant polysomnographic recordings and long term respiratory sound recordings by means of LEOSound were performed. Patients' average age was 58 ± 12 years (mean ± standard deviation), average BMI was 33 ± 7 kg/m. Audio-visual apnoea detection by LEOSound was compared to polysomnographic apnoea detection. Increased artifact rate due to dislocation of microphones led to rejection of 11 out of 45 recordings for detailed analysis.  Comparison of apnea detection by audio-visual analysis and polysomnography yielded a median of 164 apneas for LEOSound recordings and 158 apneas for PSG. Median apnoea index (AI) was calculated to be 20/h for respiratory sounds recording and 21/h for PSG. The correlation of apnea indices from acoustic long term registration and PSG was 0.939 (p < 0.001).  Acoustic long term registration of primary and secondary respiratory sounds is also capable to recognize apnoeas. Exact differentiation between apnoeas and hypopnoeas is only possible in a limited fashion.

摘要

在本研究中,对通过多导睡眠图(PSG)和夜间呼吸音记录来识别阻塞性睡眠呼吸暂停(OSA)患者的呼吸暂停情况进行了分析和比较。在45例需要治疗的OSA患者(呼吸暂停低通气指数>15次/小时)中,同时进行了多导睡眠图记录以及通过LEOSound进行的长期呼吸音记录。患者的平均年龄为58±12岁(均值±标准差),平均体重指数为33±7kg/m²。将LEOSound的视听呼吸暂停检测结果与多导睡眠图的呼吸暂停检测结果进行了比较。由于麦克风移位导致伪影率增加,45份记录中有11份被排除用于详细分析。通过视听分析和多导睡眠图进行呼吸暂停检测的比较结果显示,LEOSound记录的呼吸暂停中位数为164次,PSG记录的为158次。呼吸音记录的呼吸暂停指数(AI)中位数计算为20次/小时,PSG的为21次/小时。声学长期记录与PSG的呼吸暂停指数相关性为0.939(p<0.001)。对一级和二级呼吸音进行声学长期记录也能够识别呼吸暂停。但呼吸暂停和呼吸浅慢之间的精确区分仅在有限程度上可行。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验