Intensity of and Adherence to Lipid-Lowering Therapy as Predictors of Major Adverse Cardiovascular Outcomes in Patients With Coronary Heart Disease

Abstract

Background The effectiveness of lipid-lowering therapy (LLT) is affected by both intensity and adherence. This study evaluated the associations of LLT intensity, adherence, and the combination of these 2 aspects of LLT management with the risk of major adverse cardiovascular events (MACE) in people with coronary heart disease. Methods and Results This is an observational study of all adults who suffered a myocardial infarction or had coronary revascularization during 2012 to 2018 and initiated LLT in Stockholm, Sweden. Study exposures were LLT adherence (proportion of days covered), LLT intensity (expected reduction of low-density lipoprotein cholesterol), and the combined measure of adherence and intensity. At each LLT fill, adherence and intensity during the previous 12 months were calculated. The primary outcomes were MACE (nonfatal myocardial infarction or stroke and death); secondary outcomes were low-density lipoprotein cholesterol goal attainment and individual components of MACE. We studied 20 490 patients aged 68±11 years, 75% men, mean follow-up 2.6±1.1 years. Every 10% increase in 1-year adherence, intensity, or adherence-adjusted intensity was associated with a lower risk of MACE (hazard ratio [HR], 0.94 [95% CI, 0.93-0.96]; HR, 0.92 [95% CI, 0.88-0.96]; and HR, 0.91 [95% CI, 0.89-0.94], respectively) and higher odds of attaining low-density lipoprotein cholesterol goals (odds ratio [OR],1.12 [95% CI, 1.10-1.15]; OR, 1.42 [95% CI, 1.34-1.51], and OR, 1.16 [95% CI, 1.19-1.24], respectively). Among patients with good adherence (≥80%), the risk of MACE was similar with low-moderate and high-intensity LLT despite differences in the low-density lipoprotein cholesterol goal attainment with the treatment intensities. Discontinuation ≥1 year increased the risk markedly (HR,1.66 [95% CI, 1.23-2.22]). Conclusions In routine care, good adherence to LLT was associated with the greatest benefit for patients with coronary heart disease. Strategies that improve adherence and use of intensive therapies could substantially reduce cardiovascular risk.

Keywords: MACE; adherence; intensity; lipid‐lowering treatment.

Figure 1. Study design schema (A) and flowchart of study inclusion according to Consolidated Standards of Reporting Trials (B).

CHD indicates coronary heart disease; LLT, lipid‐lowering therapy; and PDC, proportion of days covered.

Figure 2. Graphical representation of the associations between continuous measures of statin (A) adherence, (B) intensity, or (C) adherence‐adjusted intensity, and the risk of MACE.

†The exposure statin intensity ranges from 0% to 66%, given that the highest doses of the most potent lipid‐lowering therapy are estimated to lower low‐density lipoprotein cholesterol (LDL‐C) by 66% in clinical trials (see Table 1). The exposure of adherence ranges from 0% (nonuse) to 100%. The combined exposure of adherence‐adjusted intensity is the product between the 2, thus ranging from 0% to 66%. $Models adjusted for demographics (ie, age, sex), estimated glomerular filtration rate, LDL‐C, number of LDL‐C measurements, all comorbidities (history of previous myocardial infarction, previous revascularization, diabetes, hypertension, heart failure, peripheral artery disease, valve disorder, stroke, transient ischemic attack, atrial fibrillation, other arrhythmias, chronic respiratory disease, other lung diseases, venous thromboembolism, liver disease, cancer, fracture in previous year), medications (β‐blockers, calcium channel blockers, diuretics, renin‐angiotensin‐system inhibitors, digoxin, nitrates, antiplatelet, anticoagulants, β‐2 agonist, anticholinergic inhalants, glucocorticoids, inhalants, oral glucocorticoids, NSAIDs, opioids), health care use (cardiovascular hospitalizations in the previous year, noncardiovascular hospitalizations during the past year, outpatient contacts for cardiovascular causes during the past year, outpatient contacts for noncardiovascular reasons in previous year, number of unique dispensed drugs during the past year), calendar year, and education level.

Figure 3. Cardiovascular risk reduction associated with each 10% increment in adherence‐adjusted intensity across subgroups.

CHD indicates coronary heart disease; eGFR, estimated glomerular filtration rate; LDL‐C, low‐density lipoprotein cholesterol; and LLT, lipid‐lowering therapy. †The P value is from the test statistic for testing interaction between the exposure and any subgroup variable. Models adjusted for demographics (ie, age, sex), eGFR, LDL‐C, average number of LDL‐C measurements, all comorbidities (history of previous myocardial infarction, previous revascularization, diabetes, hypertension, heart failure, peripheral artery disease, valve disorder, stroke, transient ischemic attack, atrial fibrillation, other arrhythmias, chronic respiratory disease, other lung diseases, venous thromboembolism, liver disease, cancer, fracture in previous year), medications (β‐blockers, calcium channel blockers, diuretics, renin‐angiotensin‐system inhibitors, digoxin, nitrates, antiplatelet, anticoagulants, β‐2 agonist, anticholinergic inhalants, glucocorticoids, inhalants, oral glucocorticoids, NSAIDs, opioids), health care use (cardiovascular hospitalizations in the previous year, noncardiovascular hospitalizations during the past year, outpatient contacts for cardiovascular causes during the past year, outpatient contacts for noncardiovascular reasons in previous year, number of unique dispensed drugs during the past year), calendar year, and education level.

Figure 4. The effect of adherence (y axis, %) on the predicted probability of low‐density lipoprotein cholesterol (LDL‐C) goal attainment across levels of intensity (x axis, %) and vice versa.

The exposure treatment intensity ranges from 0% to 66%, given that the highest doses of the most potent lipid‐lowering therapy are estimated to lower LDL‐C by 66% in clinical trials (see Table 1). The exposure of treatment adherence ranges from 0% (nonuse) to 100%. 1.8 mmol/L LDL‐C=70 mg/dL. Results of repeated‐measures logistic regression with an interaction between 2 continuous covariates. The model includes subjects as random effects and fixed effects for patient demographics (ie, age, sex), estimated glomerular filtration rate, average number of LDL‐C measurements, all comorbidities (history of previous myocardial infarction, previous revascularization, diabetes, hypertension, heart failure, peripheral artery disease, valve disorder, stroke, transient ischemic attack, atrial fibrillation, other arrhythmias, chronic respiratory disease, other lung diseases, venous thromboembolism, liver disease, cancer, fracture in previous year), medications (β‐blockers, calcium channel blockers, diuretics, renin‐angiotensin‐system inhibitors, digoxin, nitrates, antiplatelet, anticoagulants, β‐2 agonist, anticholinergic inhalants, glucocorticoids, inhalants, oral glucocorticoids, NSAIDs, opioids), health care use (cardiovascular hospitalizations in the previous year, noncardiovascular hospitalizations during the past year, outpatient contacts for cardiovascular causes during the past year, outpatient contacts for noncardiovascular reasons in previous year, number of unique dispensed drugs during the past year), calendar year, and education level.