Statin therapy is indicated after ST-segment elevation myocardial infarction (STEMI) to reduce recurrent ischemic events, but approximately 6% of patients with STEMI do not receive a statin prescription at discharge. This substudy aimed to define the clinical outcomes and patient characteristics associated with statin nonprescription after STEMI. We compared clinical, angiographic, and procedural characteristics and in-hospital, 30-day, 1-year, 2-year, and 3-year outcomes in 3,512 patients discharged after STEMI with and without (6%) statin prescriptions in the harmonizing outcomes with revascularization and stents in acute myocardial infarction trial ( www.clinicaltrials.gov , NCT00433966 ). Statin nonprescription was associated with female sex, nonwhite race, previous bypass surgery, heart failure, renal impairment, anemia, thrombocytopenia, care in the United States, lower prescription rates of antiplatelets and neurohormonal antagonists, less percutaneous coronary intervention and stents, and, in 26% of cases, angiographically normal or nonobstructed coronary arteries. At every time point of follow-up after discharge, patients with no discharge statin prescription had significantly higher rates of net adverse clinical events, major adverse cardiac events, major bleeding unrelated to bypass surgery, and death. After multivariable adjustment, absence of a discharge statin prescription independently predicted 3-year major adverse cardiac event (hazard ratio 1.54, 95% confidence interval 1.15 to 2.07, p = 0.0037) and death (hazard ratio 2.30, 95% confidence interval 1.41 to 3.77, p = 0.0009). In conclusion, within the framework of this randomized trial of patients presenting with STEMI, approximately 6% of patients were discharged without statin therapy. Absence of a discharge statin prescription after STEMI was an independent predictor of ischemic events including death.
A small but sizable proportion of patients with acute ST-segment elevation myocardial infarction (STEMI) depart the hospital without a statin prescription. In the National Cardiovascular Data Registry Acute Coronary Treatment and Intervention Outcomes Network-Get With the Guidelines registry, in 2009, 5.5% of patients with STEMI (n = 11,125) did not have a statin prescription at discharge. This figure likely underestimates the extent of statin nonprescription after STEMI, given the registry’s design as a quality improvement program with voluntary participation. Statin nonprescription may occur due to contraindications, including hypersensitivity, active liver disease, unexplained persistent elevations in serum transaminases, pregnancy and lactation, patient intolerance, including skeletal muscle toxicity, or oversight. Although there is concern that statin nonprescription deprives patients with STEMI of established benefits, the clinical outcomes and patient characteristics associated with statin nonprescription after STEMI are not well characterized. To evaluate this, we compared ischemic and bleeding end points in-hospital and at 1-, 2-, and 3-year follow-up in patients with and without discharge statin prescriptions in the Harmonizing Outcomes With Revascularization and Stents in Acute Myocardial Infarction (HORIZONS-AMI) trial.
Methods
HORIZONS-AMI was a prospective, open-label, multicenter, randomized controlled trial of patients with STEMI undergoing primary percutaneous coronary intervention (PCI; Clinical Trial Registration number, NCT00433966 ). In a 2 × 2 factorial design, subjects (n = 3,602) were first randomized 1:1 to receive unfractionated heparin plus routine use of a glycoprotein IIb/IIIa inhibitor versus the direct thrombin inhibitor bivalirudin plus provisional glycoprotein IIb/IIIa inhibitor use and then randomized (where eligible; n = 3,011) 3:1 to placement of paclitaxel-eluting or bare-metal stents. Eligible patients included men and women ≥18 years of age with STEMI undergoing randomization within 12 hours of symptom onset. Principal exclusion criteria included contraindications to study drugs; bleeding or bleeding diathesis, including anticoagulant use; recent stroke; extensive peripheral vascular disease; and inability to adhere to a thienopyridine antiplatelet drug for at least 6 months after PCI.
All patients underwent emergent coronary angiography after the first randomization with disposition to PCI, coronary artery bypass graft surgery (CABG), or medical therapy alone left to the discretion of the treating physician. Clinical follow-up was obtained at 30 days, 1 year, 2 years, and 3 years in all patients. Results of the primary analyses of HORIZONS-AMI have been reported previously. For the present study, analysis was restricted to patients surviving to hospital discharge (n = 3,512). All participants provided written informed consent. The study was approved by the institutional review board or ethics committee of every participating site.
All baseline angiograms were evaluated by an angiographic core laboratory blinded to randomization assignment and clinical outcomes. Quantitative coronary angiography was performed using standardized techniques, as described previously, for every lesion in the coronary tree with reference vessel diameter ≥1.5 mm.
For the present study, the primary end point was net adverse clinical events (NACEs), defined as the composite of major adverse cardiac events (MACEs)—death, reinfarction, stroke, or ischemia-driven target vessel revascularization—and non–CABG-related major bleeding, with individual end points as defined previously. All events were adjudicated by an independent clinical events committee blinded to treatment assignment using original source documents.
Subjects were classified into 2 groups based on the presence or absence of a statin prescription on the discharge medication list. Baseline, clinical, and procedural characteristics were compared. Categorical variables are presented as proportions and compared using the chi-square or Fisher’s exact test as appropriate. Continuous variables are presented as median with interquartile range except where noted and compared using the Wilcoxon rank-sum test. In-hospital adverse events are presented as rates. Event rates at follow-up were derived using Kaplan-Meier estimates and compared using the log-rank test. All statistical tests were 2-tailed. p Values <0.05 were deemed statistically significant.
Acknowledging the presence of both measured and unmeasured selection biases influencing discharge statin use, we next sought to identify independent predictors of 3-year NACE, MACE, death, and non-CABG-related major bleeding using a multivariable stepwise Cox regression model (entry and exit, p <0.1). Candidate variables included common predictive variables from the HORIZONS-AMI trial and additional variables identified a posteriori after univariate comparison of baseline, clinical, and procedural characteristics in the statin and no-statin groups. These included age, sex, race, baseline creatinine clearance, hyperlipidemia, diabetes mellitus, location of enrollment (United States or outside the United States), primary management strategy (primary PCI, deferred PCI, CABG without PCI, or medical management only), absence of obstructive coronary artery disease, final (post-PCI) thrombolysis in myocardial infarction (MI) flow grade, nadir hemoglobin >25% below baseline, platelet count <150,000, and other discharge medications (including angiotensin-converting enzyme inhibitors [ACEIs], angiotensin receptor blockers [ARBs], β blockers, aspirin, and thienopyridines).
Results
Among 3,512 patients analyzed, 219 (6%) had no statin prescription at discharge. In comparison with patients discharged on statin therapy, patients with no statin prescription had a higher frequency of female sex, nonwhite race, previous CABG, heart failure, impaired kidney function (defined by creatinine clearance <60 ml/min), and anemia (defined by hemoglobin <10 mg/dl) and more often received care in the United States. Patients without a statin prescription had a lower frequency of hyperlipidemia and a lower median body mass index. Baseline characteristics were otherwise similar ( Table 1 ).
| Variable | Discharge Statin Prescription | p Value | |
|---|---|---|---|
| Yes (n = 3,293) | No (n = 219) | ||
| Age (yrs) | 59.9 (52.4, 69.4) | 60.7 (50.3, 73.5) | 0.52 |
| Women | 23 | 29 | 0.03 |
| White | 94 | 87 | <0.001 |
| Body mass index (kg/m 2 ) | 27.1 (24.6, 30.4) | 26.0 (23.5, 29.0) | 0.0003 |
| Previous MI | 11 | 12 | 0.58 |
| Previous PCI | 11 | 11 | 0.95 |
| Previous coronary artery bypass grafting | 2.7 | 5.0 | 0.0496 |
| Family history of premature CAD | 30 | 28 | 0.66 |
| Previous angina | 22 | 16 | 0.06 |
| Hypertension | 53 | 59 | 0.08 |
| Hyperlipidemia | 44 | 36 | 0.02 |
| Smoker | 64 | 59 | 0.10 |
| Diabetes mellitus | 16 | 15 | 0.49 |
| Peripheral vascular disease | 4.1 | 5.9 | 0.2 |
| Heart failure | 2.6 | 6.4 | 0.0008 |
| Creatinine clearance ∗ <60 ml/min at baseline | 15 | 22 | 0.009 |
| Hemoglobin <10 mg/dl at baseline | 0.6 | 2.0 | 0.049 |
| Location of study enrollment | <0.0001 | ||
| United States | 21 | 40 | |
| Outside United States | 79 | 60 | |
∗ Creatinine clearance was estimated using the Cockcroft-Gault equation. Hypertension and hyperlipidemia were defined by patient history.
With respect to other medications, at baseline, a minority of patients in both groups were treated with aspirin, thienopyridines, β blockers, ACEI, or ARB. At discharge, patients without statin prescriptions were also significantly less likely to be prescribed ACEI/ARB, β blockers, aspirin, or thienopyridines ( Table 2 ).
| Variable | Discharge Statin Prescription | p Value | |
|---|---|---|---|
| Yes (n = 3,293) | No (n = 219) | ||
| At baseline | |||
| ACEI/ARB | 24 | 27 | 0.26 |
| Amiodarone | 0.5 | 0.5 | 1 |
| Aspirin | 23 | 26 | 0.43 |
| β blocker | 21 | 27 | 0.04 |
| Calcium antagonist | 10 | 11 | 0.53 |
| Cyclooxygenase inhibitor | 0.7 | 1.8 | 0.09 |
| Digoxin | 0.8 | 1.4 | 0.25 |
| Insulin | 4.3 | 2.7 | 0.28 |
| Oral hypoglycemic | 9.6 | 11 | 0.5 |
| Thienopyridine | 2.7 | 3.7 | 0.42 |
| At discharge | |||
| ACEI/ARB | 82 | 57 | <0.0001 |
| Amiodarone | 2.8 | 4.6 | 0.12 |
| Aspirin | 99 | 84 | <0.0001 |
| β blocker | 92 | 71 | <0.0001 |
| Coumadin | 3.9 | 6.4 | 0.06 |
| Digoxin | 1.3 | 2.3 | 0.21 |
| Diuretics | 22 | 23 | 0.56 |
| Insulin | 5.7 | 5.1 | 0.68 |
| Other lipid-lowering drug | 4.0 | 11 | <0.0001 |
| Thienopyridine | 95 | 63 | <0.0001 |
Management strategies differed according to the subsequent statin prescription at discharge ( Table 3 ). Patients with no discharge statin prescription were less likely to undergo primary PCI and more likely to undergo CABG or medical management only. In the statin nonprescription group, among those assigned to medical management only (n = 62), the most common reason for this strategy was normal coronary arteries or nonobstructive coronary artery disease. Among patients assigned to primary PCI, patients with no discharge statin prescription were less likely to undergo stent implantation and more likely to have thrombolysis in MI grade 0 or 1 flow after PCI. Among patients receiving stents, usage of drug-eluting and bare-metal stents was similar in both groups.
| Variable | Discharge Statin Prescription | p Value | |
|---|---|---|---|
| Yes (n = 3,293) | No (n = 219) | ||
| Primary management strategy | |||
| Primary PCI | 95 | 66 | <0.0001 |
| CABG without PCI | 1.6 | 4.6 | 0.004 |
| Medical only | 3.5 | 28 | <0.0001 |
| Reason for use of only medical management | |||
| Normal coronary arteries | 1.2 | 21.9 | <0.0001 |
| Nonobstructive CAD | 1.0 | 3.7 | 0.003 |
| Other | 0.9 | 2.7 | 0.02 |
| Modality of PCI | |||
| PTCA only | 3.7 | 7.7 | 0.02 |
| ≥1 stent implanted ∗ | 95 | 83 | <0.0001 |
| Door to balloon time (min) | 98 [73, 135] | 96 [75, 124] | 0.49 |
| Final TIMI flow after PCI | |||
| 0/1 | 1.8 | 4.6 | 0.03 |
| 2 | 5.9 | 4.6 | 0.51 |
| 3 | 92 | 91 | 0.49 |
| Post-PCI laboratory data | |||
| Nadir hemoglobin | |||
| <10 mg/dl | 7.0 | 19 | <0.0001 |
| >25% below baseline | 8.2 | 18 | <0.0001 |
| Peak creatinine | |||
| >25%/0.5 mg/dl above baseline | 15 | 13 | 0.49 |
| Platelets | |||
| <150,000 (any time) | 15 | 22 | 0.002 |
| <100,000 (any time) | 2.5 | 5.6 | 0.008 |
∗ In cases involving stent implantation, usage of drug-eluting and bare-metal stents was similar (overall, 72% and 29%, respectively).
After cardiac catheterization, hematologic abnormalities were more common in the statin nonprescription group. Patients with no discharge statin prescription were more likely to have a nadir hemoglobin substantially below baseline and more likely to show thrombocytopenia. Rates of acute kidney injury were similar. Liver function tests were not obtained.
During the index hospitalization for STEMI and at every time point of follow-up after discharge, including 30 days, 1 year, 2 years, and 3 years, patients with no discharge statin prescription had higher rates of NACE, MACE, non-CABG major bleeding, and death ( Table 4 and Figure 1 ). In-hospital stroke was rare but more common among patients with no statin.
| Outcome | Discharge Statin Prescription | p Value | |
|---|---|---|---|
| Yes (n = 3,293) | No (n = 219) | ||
| In-hospital | |||
| NACE (primary end point) | 7.3 | 14.6 | <0.001 |
| MACE | 2.0 | 4.6 | 0.03 |
| Reinfarction | 1.0 | 1.4 | 0.48 |
| Stroke | 0.4 | 1.8 | 0.02 |
| Any revascularization | 1.8 | 2.7 | 0.3 |
| Ischemia-driven target vessel revascularization | 1.5 | 2.3 | 0.4 |
| Ischemia-driven non-target vessel revascularization | 0.6 | 0.9 | 0.65 |
| Non-CABG major bleeding | 5.8 | 11.4 | 0.0008 |
| Blood transfusion | 2.2 | 4.1 | 0.07 |
| 3-year | |||
| NACE (primary end point) | 24.3 | 34.6 | 0.0002 |
| MACE | 19.6 | 29.0 | 0.0008 |
| Death | 4.3 | 11.4 | <0.0001 |
| Cardiac death | 1.9 | 4.8 | 0.002 |
| Noncardiac death | 2.5 | 6.9 | 0.0007 |
| Bleeding-related death | 0.0 | 0.7 | 0.002 |
| Stroke-related death | 0.1 | 1.3 | 0.001 |
| Reinfarction | 6.8 | 9.7 | 0.1 |
| Stroke | 1.7 | 3.3 | 0.11 |
| Any revascularization | 18.7 | 22.3 | 0.2 |
| Ischemia-driven target vessel revascularization | 12.9 | 14.5 | 0.64 |
| Ischemia-driven nontarget vessel revascularization | 9.4 | 12.2 | 0.18 |
| Target lesion thrombosis | 5.1 | 9.4 | 0.04 |
| Non-CABG major bleeding | 7.7 | 14.5 | 0.0003 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
