Effect of ondansetron on moderate obstructive sleep apnoea, a single night, placebo‐controlled trial
Effect of ondansetron on moderate obstructive sleep apnoea, a single night, placebo‐controlled trial
复制标题
昂丹司琼对中度阻塞性睡眠呼吸暂停的影响，一晚安慰剂对照试验
DOI：
10.1046/j.1365-2869.2003.00342.x
复制
发表时间：
2003
影响因子：
4.4
通讯作者：
D. Carley
中科院分区：
文献类型：
--
作者：
J. Stradling;Debbie Smith;M. Radulovacki;D. Carley
关于5-羟色胺神经传递通路在阻塞性睡眠呼吸暂停(OSA)发病机制中的作用，一直存在相互矛盾的数据(Horner，2000)。5-羟色胺控制舌下神经元供应颧舌肌的某些方面，并可能刺激上呼吸道肌肉活动(Jelev等人。2001)。非快速眼动(NREM)睡眠时上呼吸道张力的降低被认为是由于5-羟色胺(5-HT)活动的停止所致，但快速眼动(REM)睡眠中上呼吸道肌肉活动进一步减少的机制尚不清楚。此外，刺激麻醉猫的迷走神经传入C纤维可以产生中枢性呼吸暂停(Vardhan等人。1993年)，这种效应可以通过用5-羟色胺刺激结状神经节来复制(Sutton 1981)。给予5-HT3受体拮抗剂可防止这些呼吸暂停(Yoshioka等人。(1992年)。在大鼠中，5-HT3拮抗剂恩丹西酮也阻止了REM睡眠中这种外周介导的中枢性呼吸暂停(Carley和Radulovacki 1999；Radulovacki等人)。1998年)。最近，在《睡眠》杂志上，一项单剂量研究表明恩丹西酮影响患有OSA的英国斗牛犬的呼吸暂停活动，主要是在REM睡眠期间，但也在一定程度上影响NREM睡眠(Veasey等人)。2001)。目前尚不清楚这是纯粹的外围效应，还是额外的中枢效应。这项研究旨在评估单剂量恩丹西酮在OSA斗牛犬模型中的这种效果是否可以在患有OSA的人类身上复制。我们研究了患有中度睡眠呼吸暂停的受试者，因为我们假设，与咽部塌陷力最大的重度疾病患者相比，这些受试者更容易影响他们的OSA。10名等待鼻腔持续正压(CPAP)治疗的有症状的OSA患者被要求参加这项研究，如果他们有中度疾病(>；4%SaO2 dips h)1在10到40之间，并且居住在牛津或附近(出于方便的原因)。诊断睡眠研究使用了五通道呼吸多导睡眠监测系统、血氧仪、心率、身体运动、鼾声和脉搏传递时间(提供唤醒和呼吸努力的指数；Visi实验室，斯托伍德科学系统，英国牛津)。实验睡眠研究是根据标准标准(匿名1999年)(Embla Medical，弗拉加，冰岛)用常规多导睡眠图进行的。对这些轨迹进行分析，以提供每小时呼吸暂停、每小时低呼吸暂停(HI)、合并(AHI)和每小时4%SaO2减饱和度(ODI)的单独数值。患者每周两次在20：00入院进行通宵研究。他们被要求在入院当天不喝酒、不吃镇静剂或喝咖啡。所有其他药物都被允许，他们被要求在研究间隙的一周内不要更换任何常规药物。在患者选择就寝时间前一小时，他们被随机分为恩丹西酮16毫克或与之匹配的安慰剂。两个晚上的结果用配对t检验进行了比较。这项研究由牛津郡临床研究伦理委员会通过(申请号：C01.005)，允许的最大口服剂量为16毫克。所有10名患者(一名女性)都完成了治疗方案，并且没有报告任何副作用。患者平均年龄(SD)53.1岁(11.4)，平均体重指数33.7(4.9)，AHI值17~62。结果如表1所示。正如预期的那样，睡眠结构是混乱的，在阶段1和2、阶段3和阶段4或阶段3和阶段4或REM的百分比或绝对量方面，两组之间没有显著差异。此外，根据个体睡眠阶段的不同，两组之间的呼吸指数没有显著差异。这项研究没有发现恩丹西酮对中度OSA的影响，与在患有轻度OSA的英国斗牛犬中发现的相反(Veasey等人。2001)。这可能有几个原因。首先，我们服用恩丹西酮的剂量明显低于斗牛犬的剂量，分别为20和40毫克，分别约为1和2毫克公斤。在我们的研究中，16毫克相当于0.15毫克公斤。根据其许可的适应症，我们使用了这个剂量，因为它是控制恶心的最高推荐单次口服剂量。其次，较长的剂量可能是合适的，因为有一些证据表明，神经性贪食症患者的治疗效果可能需要长达4周的时间才能出现(Faris等人。2000)。或者，可能是这项研究的力度太小，无法证明效果。然而，AI、AHI、HI和ODI在第一晚和第二晚的相关系数分别为0.64、0.88、0.27和通信：John Stradling，牛津大学呼吸医学中心，牛津大学丘吉尔医院，牛津，OX37LJ，UK。电话：44 1865225236；传真：44 1865225221；电子邮件：john.stradling@orh.nhs.uk J.睡眠研究(2003年)12,169-170
There has been conflicting data on the part played by 5-HT neurotransmission pathways in the pathogenesis of obstructive sleep apnoea (OSA) (Horner 2000). 5-HT controls some aspects of the hypoglossal neurones supplying genioglossus, and is likely to stimulate upper airway muscle activity (Jelev et al. 2001). Reduction in upper airway tone during non-rapid eye movement (NREM) sleep is thought to result from withdrawal of 5-HT activity, but the mechanism of further reduced upper airway muscle activity during rapid eye movement (REM) sleep is far from clear. In addition stimulation of vagal afferent C fibres in anaesthetized cats can produce central apnoeas (Vardhan et al. 1993) and this effect can be replicated by stimulation of the nodose ganglion with 5-HT (Sutton 1981). The administration of 5-HT3 receptor antagonists prevents these apnoeas (Yoshioka et al. 1992). In rats ondansetron, a 5-HT3 antagonist, also blocks this peripherally mediated central apnoea in REM sleep (Carley and Radulovacki 1999; Radulovacki et al. 1998). Recently in the journal Sleep, it has been shown in a single dose study that ondansetron influences apnoea activity in English bulldogs with OSA, mainly during REM sleep, but also to some extent in NREM sleep (Veasey et al. 2001). It is unclear whether this is due to purely peripheral effects, or additional central effects. This present study was designed to assess whether this effect of a single dose of ondansetron in the bulldog model of OSA could be replicated in humans with OSA. We studied subjects with moderate sleep apnoea, as we hypothesized that in these subjects it would be easier to influence their OSA, in comparison with patients with severe disease where pharyngeal collapsing forces are greatest. Ten patients with symptomatic OSA awaiting nasal continuous positive air pressure (CPAP) were asked to take part in this study if they had moderate disease (>4% SaO2 dips h )1 of between 10 and 40), and lived in or near Oxford (for reasons of convenience). The diagnostic sleep studies were performed using a five-channel respiratory polysomnography system, oximetry, heart rate, body movement, snoring and pulse transit time (providing indices of arousal and respiratory effort; Visi-Laboratory, Stowood Scientific Systems, Oxford, UK). The experimental sleep studies were carried out with conventional polysomnography according to standard criteria (Anonymous 1999) (Embla Medical, Flaga, Iceland). The tracings were analysed to provided separate values for apnoeas per hour, hypopnoeas per hour (HI), combined (AHI), and >4% SaO2 desaturations per hour (ODI). Patients were admitted at 20:00 h for overnight studies on two occasions a week apart. They were asked to refrain from alcohol, sedatives or coffee on the day of admission. All other drugs were allowed and they were asked not to change any regular medications during the week between studies. One hour before the patients chosen bedtimes they were randomized to receive either ondansetron 16 mg or matching placebo. Results from the two nights were compared using paired t-tests. This study was passed by the Oxfordshire Clinical Research Ethics Committee (application no. C01.005) who allowed a maximum oral dose of 16 mg. All 10 patients (one woman) completed the protocol and reported no side-effects. The patients had a mean (SD) age of 53.1 (11.4), mean body mass index of 33.7 (4.9) and their AHI values ranged from 17 to 62. The results are shown in the Table 1. The sleep structure was chaotic, as would be expected, and there were no significant differences between the two groups for percentage or absolute amounts of stages 1 and 2, or 3 and 4, or REM. In addition, there were no significant differences in respiratory indices between the two groups when expressed according to the individual sleep stages. This study has found no effect of ondansetron on moderate OSA, contrary to that found in the English bulldog with mild OSA (Veasey et al. 2001). This could be for several reasons. First, the dose of ondansetron we administered was significantly lower than that used in the bulldogs, 20 and 40 mg, which approximated to 1 and 2 mg kg, respectively. In our study, a dose of 16 mg approximated to 0.15 mg kg. We used this dose as it is the highest recommended single oral dose for nausea control, according to its licensed indication. Secondly, it could be that a longer period of dosing would be appropriate, as there is some evidence in patients with bulimia nervosa that the therapeutic effect may take up to 4 weeks to appear (Faris et al. 2000). Alternatively, it may be that the study was too under-powered to demonstrate an effect. However, the correlation coefficients between night 1 and night 2 for AI, AHI, HI and ODI were 0.64, 0.88, 0.27 and Correspondence: John Stradling, Oxford Centre for Respiratory Medicine, Oxford University, Churchill Hospital, Oxford, OX3 7LJ, UK. Tel.: 44 1865225236; fax: 44 1865225221; e-mail: john.stradling@orh.nhs.uk J. Sleep Res. (2003) 12, 169–170