What short‐term potentiation is and why it may be relevant to obstructive sleep apnoea

The fundamental problem in obstructive sleep apnoea (OSA) is that the upper airway (UA) collapses in the absence of sufficient activation of pharyngeal dilator muscles in subjects with an anatomical predisposition (White & Younes, 2012). These muscles are sufficiently strong that they can maintain the UA open despite negative pharyngeal pressures of −100 cmH2O during voluntary manoeuvres. Given the strength of these muscles, why do they fail to keep the airway open during sleep in these patients? During sleep (i.e. absent consciousness), UA dilators are activated by increases in respiratory drive and/or through reflexes responsive to negative pharyngeal pressure (White & Younes, 2012). Neither of these stimuli is in short supply during an obstructive event since respiratory drive increases as blood gas tensions deteriorate and pharyngeal pressure becomes progressively more negative in response to the increasing suction pressure produced by the diaphragm against a closed airway. So, what is the problem? Until recently, it was felt that UA dilators in these patients cannot be adequately activated by these subconscious stimuli and, accordingly, that conscious factors must be re-engaged, through arousal, to open the airway (Remmers et al. 1978). More recently, it became clear that UA dilators in most OSA patients do respond vigorously to these stimuli but only if they progress to a threshold level (White & Younes, 2012). While this threshold is very modest in most patients, it is not reached because the same stimuli result in arousal at an even lower threshold. Arousal restores patency at a time when chemical drive is elevated. Accordingly, ventilation increases, drive decreases and the stimulus to UA dilators disappears. The patient obstructs again, and the cycle repeats. Mechanical therapies for OSA (continuous positive airway pressure (CPAP), mandibular devices, surgery) address the disorder by reducing dependence on dilator activity. They are, however, either poorly tolerated or unreliable. Consequently, there is considerable interest in non-mechanical approaches to therapy. One approach is to use agents that increase arousal threshold so that chemical drive may reach the required threshold before arousal occurs. Unfortunately, in therapeutic doses, available sedatives do not increase arousal threshold enough in most patients. A second potential approach is to use or develop agents that target known excitatory receptors on the hypoglossal motor nucleus thereby causing direct activation and reducing the need for an increase in drive, with consequent arousal. Unfortunately, despite testing of numerous candidate drugs, an agent in this category that can sufficiently reduce OSA severity with acceptable side effects has not been found. And, here lies the relevance of short-term potentiation (STP) and its associated phenomenon, the after-discharge (AD). STP–AD is a phenomenon whereby motoneuron activity increases for a while even when the applied stimulus is constant (STP), and activity remains elevated for several seconds or minutes following stimulus withdrawal (AD). Such a response in UA dilators would be ideal in OSA patients. Given that at apnoea termination UA dilators are strongly activated (reflexly or by arousal), continued activity during the vulnerable period of low respiratory drive that follows the previous apnoea should protect against recurrence of the apnoea. STP and AD were first demonstrated in respiratory muscles (diaphragm) by Eldridge & Jill-Kumar (1978). AD was subsequently demonstrated in the genioglossus (GG) in healthy awake humans (Jordan et al. 2002). The decline in GG activity following stimulation in this study paralleled that of the diaphragm, suggesting that it provides no preferential excitation to UA muscles; preferential excitation is needed to overcome the obstruction (Remmers et al. 1978). The first study of STP–AD in OSA patients was in 2014 (Younes et al. 2014). GG activity was monitored on CPAP before, during and after induced apnoeas produced by transient reduction in CPAP pressure. GG activity remained elevated for a minute or more after ventilation had returned to, or below, the pre-event ventilation level, indicating that AD of the genioglossus outlasts the diaphragm's AD. The response was quite variable and was poor in many patients. Thus, enhancement of STP–AD offered a new approach to therapy that, unlike direct stimulation of GG nucleus, utilizes indirect stimulation of the GG motoneurons via natural pathways (those that mediate STP–AD). In parallel, it has long been observed that OSA tends to spontaneously disappear if an OSA patient entered slow-wave sleep (SWS). The exact mechanism for this was not known although it was suspected that higher arousal threshold in SWS contributed by delaying arousal and allowing the normal compensatory mechanisms to progress to the required level. In an article in this issue of The Journal of Physiology, Taranto-Montemurro et al. (2018) explore the possibility that the improved UA stability in SWS may result from a more potent AD in this stage. They indeed found this to be true both in healthy controls and in OSA patients; the duration of AD increased as the same subjects entered SWS. Apart from providing an additional mechanism by which SWS stabilizes breathing, this observation has important clinical implications. First, it showed that STP–AD can increase spontaneously in the same patient (i.e. it is not a fixed attribute) through perfectly natural pathways (in this case progression to deeper sleep). Second, there is a long list of commercial (i.e. safe) drugs that increase slow-wave activity (Walsh et al. 2009). Not all drugs that increase slow-wave activity actually increase sleep depth (Taranto-Montemurro et al. 2017), but some do. One or more of these drugs may be able to safely enhance STP–AD and offer relief for at least some OSA patients. None. Sole author. None received for this work.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.