INTRODUCTION
Ticagrelor directly and reversibly inhibits the platelet P2Y12 receptor and exhibits a more potent and more predictable antiplatelet effect compared to clopidogrel.1 The use of ticagrelor for the treatment of acute coronary syndrome patients has been shown to reduce death, myocardial infarction, and stroke, which has resulted in its approval for clinical use by the European Medicines Agency (EMA) in 2010.2 However, although the safety and efficacy of ticagrelor is well established, the tolerability of ticagrelor appears to be impaired.3 In Phase III trials, premature therapy discontinuation has been consistently reported to be 2–5% more frequent among ticagrelor-treated individuals compared to clopidogrel-treated individuals.4 As non-adherence to cardiovascular medication is a recognised and common problem that has been shown to be associated with impaired long-term outcomes,5 we aimed to elucidate the frequency, reasons, and clinical implications of ticagrelor discontinuation in routine clinical practice.
METHOD
As of January 2009, all patients undergoing percutaneous coronary intervention (PCI) at Bern University Hospital, Bern, Switzerland, have been prospectively entered into the Bern PCI Registry.6,7 The present analysis included all consecutive patients treated with ticagrelor who were enrolled between November 2011, when ticagrelor was introduced in Switzerland, and June 2014. Exclusion criteria were balloon treatment only and enrolment in a new clinical trial before 1-year follow-up. Demographic and clinical characteristics, procedural information, in-hospital outcomes, and 1-year follow-up data were systematically collected. A clinical event committee consisting of two cardiologists (and a third referee for cases of disagreement) adjudicated all events using original source documents.
RESULTS
Out of 4,837 consecutive patients prospectively enrolled into the Bern PCI Registry between November 2011 and June 2014, 1,278 patients were treated with ticagrelor. Within the recommended 1-year therapy duration, 152 patients (11.9%) prematurely changed to clopidogrel or prasugrel, and 60 patients (4.7%) prematurely discontinued ticagrelor therapy without replacement. Ticagrelor discontinuation occurred more frequently at a higher age, in female patients, in patients with renal failure, and following a history of previous PCI, and was less frequent among smokers.
The underlying reasons for premature ticagrelor therapy discontinuation were adverse effects (49%), oral anticoagulation (19%), preference by the general practitioner (10%), medical intervention (5%), and financial reasons (2%). Bleeding, dyspnoea, and gastrointestinal symptoms were responsible for 88% of adverse effects leading to premature ticagrelor discontinuation. Female sex and bleeding events were independent predictors for premature ticagrelor discontinuation, while smoking emerged as a predictive factor for adherence.
A total of 31 patients (2.4%) prematurely stopped ticagrelor therapy because of dyspnoea, of which 24 patients (77%) reported an immediate amelioration, while in 7 patients (23%) no improvement was observed. Patients with definite ticagrelor related dyspnoea more frequently had a previous PCI, a previous myocardial infarction, and were more often treated for stable angina.
With respect to clinical outcomes, there were no differences in all-cause death, cardiovascular death, myocardial infarction, target-vessel revascularisation, and the occurrence of stent thrombosis between patients who remained on ticagrelor and patients who prematurely discontinued ticagrelor. Patients who prematurely discontinued ticagrelor therapy suffered more frequently from Bleeding Academic Research Consortium (BARC) Type 2–5 and TIMI minor or major bleedings.
CONCLUSION
In this cohort study of patients undergoing PCI in routine clinical practice and treated with ticagrelor, premature discontinuation affected 16% of patients and did not associate with adverse cardiovascular outcomes. Adverse effects, onset of oral anticoagulation, and general practitioners’ preference were the leading reasons for discontinuation.
