Clinical Impact of Dyspnea after Ticagrelor Treatment and the Effect of Switching to Clopidogrel in Patients with Myocardial Infarction

The incidence of dyspnea and causal relationship to treatment with ticagrelor has been a matter of debate since phase 2 and 3 trials and the launch of ticagrelor. In a previous review, Cattaneo and Faioni[1] elucidated on potential mechanisms and suggested that inhibition of P2Y12 on C fibers of sensory neurons mainly accounts for the observed incidence of dyspnea which is not related to the antiplatelet effects but likely explained by differences in pharmacokinetic properties of P2Y12 inhibitors. In addition, reduced cellular uptake and metabolism of adenosine by inhibiting the sodium-independent equilibrate nucleoside transporter (ENT1) has been discussed as other mechanism.[2]

In the landmark phase-3 PLATO trial, most dyspnea episodes were mild or moderate and occurred at a rate of 14.5% in ticagrelor-treated patients compared with 8.7% in clopidogrel patients. Severe dyspnea episodes were reported very rarely and at similar rates in both treatment groups, although these episodes did not alter the efficacy or safety of ticagrelor treatment compared with clopidogrel.[3]

The incidence of ticagrelor discontinuation related to dyspnea varies suggesting that it depends on patient characteristics and probably education. Ticagrelor discontinuation because of dyspnea was only 0.9% in the PLATO trial, whereas it was reported in 6 to 10% of cases and tended to occur earlier rather than late after drug initiation and percutaneous coronary intervention (PCI) in the TWILIGHT trial investigating ticagrelor monotherapy.[4] Independent predictors include ethnic background with lower incidence in Asian population, smoking, prior PCI, hypercholesterolemia, prior coronary artery bypass, peripheral artery disease, obesity, and older age in the TWILIGHT study.[4]

In this issue of the journal, Kim et al[5] investigated the clinical impact of dyspnea in a larger dataset of Asian patients enrolled in the TALOS-AMI study, a randomized trial investigating the effects of unguided de-escalation from ticagrelor to clopidogrel 1 month after acute myocardial infarction (AMI) and initial dual antiplatelet therapy (DAPT) with ticagrelor. Dyspnea occurred in 20.6% of patients during 1 month of DAPT with ticagrelor after PCI for AMI. Again, distinct clinical characteristics were independently associated with dyspnea, e.g., older age, male gender, previous history of asthma, stroke, or previous heart failure. Adherence to ticagrelor in the continuation arm was more affected by dyspnea compared with the clopidogrel de-escalation arm. Early de-escalation to clopidogrel was associated with a greater reduction in the rate of subsequent dyspnea among patients who experienced dyspnea during the initial month of DAPT with ticagrelor. In consistence with previous analyses, the presence of dyspnea at randomization was not associated with increased risk of major adverse cardiac events, bleeding events (Bleeding Academic Research Consortium type 2, 3, or 5), or worse net clinical outcome (p = 0.84). The effect of DAPT de-escalation on clinical outcomes was consistent regardless of the presence or absence of dyspnea, without significant interaction.

The current analysis has limitation due to its post-hoc character, due to significant inter-group differences between patients with and without dyspnea (e.g., poorer left ventricular function, higher prevalence of asthmatic disease) and due to self-reporting of dyspnea without objective confirmation by pulmonary function testing. Therefore, no conclusion can be drawn with regard to the true causal relationship of dyspnea to ticagrelor treatment. Dyspnea was only assessed after 1 month of treatment and pre-existing dyspnea was not evaluated making it hard to assign effects to ticagrelor treatment only.

Some findings are surprising and contradictory to previous observations. First, the prevalence of reported dyspnea was very high for an Asian cohort, approximately 10 times higher compared with previous and contemporary studies that reported much lower incidence rates in Asian populations.[4] [6] [7] Second, the prevalence of nonsmokers was higher among patients with dyspnea; however, smoking status did increase likelihood of dyspnea in particular leading to discontinuation of ticagrelor.[3] Of note, dyspnea declined over time in both the ticagrelor continuation and the clopidogrel de-escalation arm, although the effect was more pronounced in the latter group. Thus, a certain habituation effect and time-dependent effects influencing dyspnea independent of drug therapy play an additional and possibly underestimated role.

Nevertheless, it confirms previous observations regarding the clinical impact of ticagrelor-associated dyspnea. The safe and potentially beneficial aspect of de-escalation therapy 1 to 3 months after AMI/PCI found in several recent trials[8] seem to be consistent in patients who experienced dyspnea under ticagrelor. Thus, de-escalation might represent a reasonable strategy in patients' groups who are likely to develop dyspnea and consequently are at risk for drug discontinuation, e.g., elderly patients who may also benefit by a reduction of bleeding events by early switching to clopidogrel.[9]

However, more evidence is needed to elucidate causal relationships and the true impact of de-escalation or swopping strategies in these patients, with different antithrombotic regimes and patient populations, including different ethnic groups.[10] [11] One systematic review showed that in East Asians, the reduction of DAPT intensity or duration can minimize bleeding, without safety concerns; however, in non-Asians, reduction of DAPT intensity may increase ischemic events, while DAPT abbreviation had no overall benefit.[10]

A dedicated trial to evaluate different concepts including measures to improve adherence by patient education and alternative concepts of de-escalation including ticagrelor dose reduction is highly warranted.

Received: 19 September 2024

Accepted: 20 September 2024

Accepted Manuscript online:
23 September 2024

Article published online:
26 November 2024

© 2024. Thieme. All rights reserved.

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany