Ticagrelor and central sleep apnoea: Impact of withdrawal and reintroduction

Ticagrelor, a reversible P2Y12 receptor antagonist used mostly after acute coronary syndrome, is known to be responsible for dyspnoea unrelated to underlying cardiac or pulmonary dysfunction.1, 2 Recently a few cases of Cheyne-Stokes respiration with ticagrelor were reported.1, 3, 4 Here we describe two additional cases providing further evidence of central sleep apnoeas induced by ticagrelor. A 64-year-old man treated with Continuous Positive Airway Pressure (CPAP) for obstructive sleep apnoea syndrome for 6 years was hospitalised for an acute coronary syndrome. Before myocardial infarction, adherence to CPAP was excellent and the residual apnoea-hypopnea index (AHI) was 0.9 events/hour. A treatment with ticagrelor, aspirin, β blocker, angiotensin converting enzyme inhibitor, and statin was started. He then complained of dyspnoea, unusual fatigue, and poor sleep quality. CPAP adherence tracking system showed unusual central AHI up to 60/h (Table 1). Autopiloted pressure was switched with constant pressure improving residual AHI below 10/h. A first polysomnography without CPAP, 3 months after myocardial infarction, confirmed tracking system data with many central apnoeas (central AHI = 14/h). Finally, 5 months after myocardial infarction, ticagrelor was replaced by prasugrel due to persistent dyspnoea. Patient reported immediate improvement in sleep, supported by the disappearance of residual central AHI from the tracking system. A second polysomnography was performed 1 week after ticagrelor was replaced by prasugrel with rare residual central apnoeas (central AHI = 7, Figure 1). After informed consent was obtained, ticagrelor was reintroduced 1 week later to perform a new polysomnography. Fatigue, dyspnoea, and poor quality of sleep immediately reappeared. Central sleep apnoea index dramatically increased up to 42/h with patterns of Cheyne-Stokes respiration (Figures 2 and 3). A 68-year-old man, with a history of myocardial infarction 6 months before, complained of dyspnoea, significant fatigue, and frequent awakenings during night, which began after the coronary event. His treatment included ticagrelor, aspirin, beta blocker, statin, and eplerenone. Ticagrelor was discontinued and replaced by prasugrel for 1 week and restarted for another week. On polysomnographies, central apnoea indexes were three to four times higher on ticagrelor as compared to prasugrel and related to fatigue intensity (Table 2). Moreover, dyspnoea that had disappeared during treatment with prasugrel dramatically recurred after ticagrelor reintroduction. To our knowledge, these are the first case reports of central sleep apnoeas that recurred after ticagrelor reintroduction. In the previous case reports published, ticagrelor was discontinued in three of six patients with a gradual resolution of Cheyne-Stokes respiration.1, 3, 4 However, the prevalence of central sleep apnoea in cardiovascular patients is high,5-7 making it difficult to attribute apnoeas to ticagrelor use, only on the basis of case reports. A safety signal for sleep apnoea syndrome and ticagrelor has recently been raised by a disproportionality analysis, with a ROR of 4.16 (95% CI, 2.87-6.03).8 However, disproportionality does not provide information on causal association between ticagrelor use and sleep apnoea. In the current case reports, experiences of positive dechallenge and recurrence of the event after ticagrelor reintroduction in two patients establish the causal relationship between ticagrelor and central apnoeas. In July 2018, The Pharmacovigilance Risk Assessment Committee (PRAC) asked the Marketing Authorisation Holder to provide a cumulative review of cases of sleep apnoea syndrome, including cases with positive dechallenge and/or positive rechallenge together with a proposal to update the product information as applicable.9 Besides, central sleep apnoea appears to be related to the severity and short-term prognosis of acute coronary syndrome,6 even if the directionality of that relation deserves to be explored by supplementary studies. These are both convincing case reports that emphasise the need for clinicians to be aware of this side effect of ticagrelor and search for central apnoeas in patients complaining from dyspnoea and fatigue during treatment. Key protein targets and ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org,10 the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY. We thank Michael Charles Bryan for his help reviewing the manuscript. There are no competing interests to declare.