P2Y12 inhibitor adherence trajectories in ACS

Prognostic implications of P2Y12 inhibitor adherence trajectories in patients with acute coronary syndrome undergoing percutaneous coronary intervention

Key Takeaway

This population-based cohort study evaluating the P2Y12 inhibitor adherence trajectories among patients with ACS after PCI using the APPROACH^* registry demonstrated:

• Patients follow one of five distinct P2Y12 inhibitor adherence trajectories that are associated with several socioeconomic and clinical risk factors
• Rapid decline and delayed initiation of P2Y12 inhibitors were associated with a higher risk of MACE
• Early consistent nonadherence trajectory was significantly associated with "MACE", only in patients with a drug-eluting stent
• Future studies evaluating decision support tools are needed, which can identify patients at risk of a non-adherent trajectory, plus individualized intervention to modify these trajectories.

Why This Matters

• P2Y12 inhibitor nonadherence is common among patients with ACS, which can lead to an increased risk of recurrent MACE.
• Nonadherence can follow several different patterns or trajectories, which may have distinct causes that may allow for tailored interventions.

Study Design

Adult patients ≥18 years of age who underwent PCI for ACS between 1 April 2012 and 31 March 2016 in the province of Alberta, Canada were included.

Exclusion criteria 

• Prevalent P2Y12 inhibitor users
• First fill for prasugrel
• Patients who died during or post index ACS hospitalization

Outcomes evaluated

Patients were followed for 12 months after discharge to assess:

• P2Y12 inhibitor adherence trajectories (primary outcome)
• MACE occurrence (secondary outcome)

Key Results

Overall, 12,844 patients were included in this analysis (mean age = 62.4 years; female = 24.6%)

Five distinct P2Y12 inhibitor adherence trajectories were seen

Group	Adherence trajectory	Observation	Prevalence (%) within the cohort
1	Early consistent non-adherence	No initiation or prompt discontinuation of P2Y12 inhibitor use within the first month	11.0%
2	Rapid decline	Discontinuation of P2Y12 inhibitor use after persisting for 3 months	7.7%
3	Delayed initiation	Poor initial P2Y12 inhibitor adherence with improvement over study period	6.0%
4	Gradual decline	Initial high P2Y12 inhibitor adherence with steady decline	20.5%
5	Persistent adherence	High P2Y12 inhibitor adherence throughout study period	54.8%

Risk factors for P2Y12 inhibitor use trajectories

Group	Risk factors significantly associated with P2Y12 inhibitor use as compared to group 5 (Persistent adherence)
1	Older age and lower likelihoods of PVD, ticagrelor or PPI use
1 and 2	Prior PCI, current smoking, heart failure, malignancy, bare-metal stent, PCI without stent, lower eGFR, and lower concomitant use of other preventive medications
2	High Pampalon material deprivation
3	Younger age, lower socioeconomic status, higher likelihood of smoking, bare-metal stent or no stent, EF <20%, and lower ACEI/ARB, beta-blocker, statin, or PPI use
4	Younger age, lower socioeconomic status, higher likelihood of prior PCI or CABG, smoking, bare-metal stent placement, EF 20–34%, ticagrelor use, and lower ACEI/ARB, or PPI use ≤120 days after index date

Association of P2Y12 inhibitor use with mace*

Risk factors significantly associated with P2Y12 inhibitor use as compared to group 5 (Persistent adherence)
Group	HR†	95% CI	Status
1	1.10	0.86–1.40	Not significant risk
2	1.23	1.01–1.49	Significantly higher risk
3	1.41	1.12–1.78	Significantly higher risk
4	1.11	0.96–1.29	Not significant risk

Association of P2Y12 inhibitor use with mace in group 1^*

Risk of mace in group 1
Group	HR†	95% CI	Status
≥1 drug-eluting stent	2.44	1.60–3.71	Significantly higher risk
Bare-metal stents	0.85	0.62–1.17	Not significant risk

^* Statistically significant associations are displayed in bold
^† Adjusted for significant variables

Association of P2Y12 inhibitor use with major bleeding*

Risk of major bleeding compared to group 5‡
Group	HR†	95% CI	Status
1	1.50	1.12–2.02	Higher risk§
2	1.40	1.09–1.79	Higher risk§

^* Statistically significant associations are displayed in bold
^†Adjusted for significant variables
^‡No association was seen for Groups 3 and 4
^§A higher risk was seen early after the index event

Limitations

• Data from a central prescription drug dispensation registry was used to measure P2Y12 inhibitor exposure and adherence, which assumed that patients took them as filled and may overestimate true adherence.
• Despite adjustment and measurement for known angiographic, clinical, and laboratory variables, residual unmeasured confounding may persist while evaluating associations between adherence trajectories and MACE.

^*APPROACH: Alberta Provincial Project for Outcome Assessment in Coronary Heart Disease

MAT-BH-2300018/V1/JAN2023