Patients with previous stroke are at high-risk for myocardial infarction (MI). Concern regarding increased risk of bleeding or recurrent stroke in this patient population might alter therapeutic decisions. Data were collected from 281 hospitals in the United States in the NCDR ACTION Registry. Patients with ST-segment elevation MI (STEMI; n = 15,997) or non-STEMI (NSTEMI; n = 25,514) entered into the registry from January 1, 2007 through December 31, 2007 were included. We assessed use of evidence-based medications and procedures in patients with and without previous stroke. Risk-adjusted odds ratio of death, major bleeding not related to coronary artery bypass grafting, and a composite outcome (major adverse cardiac events [MACEs], i.e., death/MI/stroke/cardiogenic shock/congestive heart failure) were calculated using logistic regression. Previous stroke was reported in 5.1% of patients with STEMI and 9.3% of those with NSTEMI. Of patients with STEMI eligible for reperfusion therapy, those with previous stroke were less likely to receive reperfusion therapy compared to patients without previous stroke. Patients with previous stroke had longer door-to-needle and door-to-balloon times. Of patients with STEMI and NSTEMI, those with previous stroke were less likely to receive evidence-based therapies. Death, MACEs, and major bleeding were more common with previous stroke. When adjusted for baseline risk, patients with previous stroke were at increased risk of death (only those with STEMI) and MACEs but not bleeding. In conclusion, patients with STEMI and previous stroke are at increased risk for death and patients with STEMI and NSTEMI are at increased risk of MACE. Despite this, previous stroke patients are less likely to receive guideline-based MI therapies.
The purpose of this analysis was to characterize in-hospital treatment and outcomes related to previous stroke in a large community-based sample of patients with myocardial infarction (MI). We hypothesized that patients with a history of stroke would have worse outcomes, more bleeding, and would be less likely to receive procedures and medications known to be beneficial for patients with MI.
Methods
Data for this study were obtained from the National Cardiovascular Data Registry (NCDR) Acute Coronary Treatment and Intervention Outcomes Network (ACTION) Registry, a nationally representative, quality improvement registry of ST-segment elevation MI (STEMI) and non-STEMI (NSTEMI). Data were derived from records of patients presenting from January 1, 2007 through December 31, 2007 at 195 participating hospitals within 24 hours of onset of symptoms and a primary diagnosis of MI. Exclusion criteria included patients admitted into a nontertiary ACTION hospital, patients with cardiogenic shock on presentation, and patients for whom information on previous stroke was missing. Trained data collectors extracted data from medical records to a Web-based case record without direct patient contact.
Patients presenting with STEMI were analyzed separately from those with NSTEMI. Previous stroke was determined by a patient’s medical record and defined as any confirmed neurologic deficit of abrupt onset caused by a disturbance in cerebral blood supply that did not resolve within 24 hours. A listing of specific data fields and their definitions is available ( http://www.ncdr.com/WebNCDR/ACTION/Elements.aspx ).
Primary outcomes of interest were in-hospital death, death or stroke, death/MI/stroke, and major adverse cardiac events (MACEs; death/MI/stroke/cardiogenic shock/congestive heart failure [CHF]), and major bleeding unrelated to coronary artery bypass grafting (CABG). Major bleeding was defined as an absolute hematocrit decrease ≥12%, intracranial hemorrhage, retroperitoneal hemorrhage, or transfusion (with baseline hematocrit ≥28% or baseline hematocrit <28% and witnessed bleeding event). We analyzed use of in-hospital procedures (thrombolysis and primary percutaneous coronary intervention [PCI] for patients presenting with STEMI or PCI within 48 hours for patients with NSTEMI). Furthermore, we determined the rate of use for known cardioprotective medications acutely in hospital and at discharge. For all analyses, the denominator consisted of eligible patients without a contraindication documented. Contraindications for receiving fibrinolytic therapy included stroke within 3 months or previous hemorrhagic stroke.
Baseline characteristics, treatment profiles, procedure use, and clinical outcomes were compared by the presence or absence of a history of stroke. Continuous variables are presented as medians with interquartile ranges and categorical variables are expressed as percentages. Univariate analysis was done using Wilcoxon rank-sum test for continuous variables and Pearson chi square test for categorical variables.
In examining the association between previous stroke and outcome, multivariable logistic regression was used to estimate effects of previous stroke. The generalized estimating equation method with exchangeable working correlation structure was used to account for within-hospital clustering because patients at the same hospital are more likely to have similar responses compared to patients in other hospitals (i.e., within-center correlation for response). The method produces estimates similar to those from ordinary logistic regression, but estimated variances of estimates are adjusted for the correlation of outcomes within each hospital. Variables adjusted in the model are gender, race, body mass index, age, history of hypertension, diabetes mellitus, peripheral artery disease, current or recent smoking, dyslipidemia, previous MI, previous PCI, previous CABG, previous CHF, current need for dialysis, heart rate, systolic blood pressure on presentation, and sign of CHF at presentation.
Subgroup analysis was done using patients with STEMI who underwent primary PCI. Odds ratios were calculated using the same method described earlier. Variables adjusted in these models are gender, race, and age.
Adjusted associations are displayed as odds ratios (95% confidence intervals). A p value <0.05 was considered statistically significant for all tests. No adjustments were made for multiple comparisons because all analyses were exploratory in nature. All statistical analyses were performed by the Duke Clinical Research Institute using SAS 9.2 (SAS Institute, Cary, North Carolina).
Results
Of a total population of 50,517, 7,522 were excluded because they were transferred into an ACTION hospital after initial care was received, 1,043 were excluded because of cardiogenic shock on admission, and 441 were excluded because of absence of information on previous stroke. Of the 41,511 patients included in the study, 25,514 presented with NSTEMI and 15,997 presented with STEMI. Of patients with NSTEMI, 9.3% (n = 2,365) had previous stroke compared to 5.1% of patients (n = 814) presenting with STEMI. Patients with previous stroke were a decade older, more likely to be non-Caucasian, and more likely to be women ( Table 1 ). As expected, patients with previous stroke also had higher rates of hypertension, diabetes mellitus, and peripheral vascular disease ( Table 1 ). Such patients were more likely to have previously manifested coronary artery disease as evidenced by previous MIs or previous revascularization procedures. On arrival patients with a history of stroke were more likely to be tachycardic, hypotensive, or to have CHF ( Table 1 ).
| Demographics and Medical History | STEMI | NSTEMI | ||||
|---|---|---|---|---|---|---|
| Previous Stroke (n = 818) | No Previous Stroke (n = 15,179) | p Value | Previous Stroke (n = 2,365) | No Previous Stroke (n = 23,149) | p Value | |
| Age (years) | 72.0 (61.0–81.0) | 60.0 (51.0–71.0) | <0.0001 | 75.0 (66.0–83.0) | 66.0 (55.0–77.0) | <0.0001 |
| Women | 44.5% | 29.2% | <0.0001 | 47.1% | 37.7% | <0.0001 |
| Body mass index (kg/m 2 ) | 27.4 (23.9–31.3) | 28.2 (25.1–32.0) | <0.0001 | 27.4 (23.8–31.6) | 28.3 (24.9–32.7) | <0.0001 |
| Ethnicity | ||||||
| Caucasian | 81.5% | 85.2% | <0.0001 | 82.2% | 84.2% | <0.0001 |
| Black | 11.4% | 7.0% | 11.3% | 8.8% | ||
| Asian | 2.1% | 1.3% | 1.7% | 1.4% | ||
| Hispanic | 2.9% | 3.5% | 2.5% | 3.3% | ||
| Other | 1.8% | 2.7% | 1.8% | 1.9% | ||
| Hypertension | 84.0% | 59.1% | <0.0001 | 86.8% | 71.5% | <0.0001 |
| Diabetes mellitus | 38.1% | 21.3% | <0.0001 | 47.7% | 32.1% | <0.0001 |
| Peripheral arterial disease | 15.4% | 5.2% | <0.0001 | 24.2% | 10.1% | <0.0001 |
| Current/recent smoker | 29.6% | 43.5% | <0.0001 | 19.4% | 30.5% | <0.0001 |
| Previous myocardial infarction | 33.0% | 17.8% | <0.0001 | 40.0% | 25.9% | <0.0001 |
| Previous coronary intervention | 24.3% | 18.6% | <0.0001 | 25.8% | 23.2% | 0.005 |
| Previous coronary artery bypass grafting | 12.7% | 7.0% | <0.0001 | 26.9% | 17.7% | <0.0001 |
| Previous congestive heart failure | 16.3% | 4.5% | <0.0001 | 30.1% | 13.8% | <0.0001 |
| Home medications | ||||||
| Aspirin | 51.0% | 33.8% | <0.0001 | 59.9% | 47.2% | <0.0001 |
| Clopidogrel | 22.3% | 8.7% | <0.0001 | 29.1% | 14.9% | <0.0001 |
| Warfarin | 12.5% | 2.7% | <0.0001 | 15.3% | 5.7% | <0.0001 |
| Presentation features | ||||||
| Onset to arrival (hours) | 3.1 (1.5–6.3) | 2.9 (1.4–5.5) | 0.0207 | 4.1 (1.7–10.0) | 4.9 (1.9–11.5) | <0.0001 |
| Heart failure on presentation | 20.8% | 8.6% | <0.0001 | 33.3% | 18.8% | <0.0001 |
| Heart rate (beats/min) | 81.0 (66.0–88.0) | 77.0 (65.0–92.0) | <0.0001 | 86.0 (72.0–103.0) | 82.0 (70.0–97.0) | <0.0001 |
| Systolic blood pressure (mm Hg) | 136.0 (115.0–158.0) | 138.0 (119.0–158.0) | 0.1489 | 142.0 (121.0–166.0) | 144.0 (124.0–164.0) | 0.0283 |
| Baseline laboratory values | ||||||
| Creatinine clearance, Modification of Diet in Renal Disease (ml/min) | 59.2 | 72.5 | <0.0001 | 54.78 | 67.42 | <0.0001 |
| Hematocrit (%) | 40.0 | 42.7 | <0.0001 | 38.2 | 41.0 | <0.0001 |
Patients with previous stroke were less likely to receive cardioprotective medications within 24 hours of presentation ( Table 2 ). Of patients presenting with NSTEMI, those with stroke were less likely to receive aspirin, clopidogrel, and anticoagulants ( Table 2 ). Similarly, aspirin, clopidogrel, and statins were used less frequently in patients with previous stroke presenting with STEMI ( Table 2 ). Most patients received cardioprotective medications on discharge ( Table 3 ) regardless of previous stroke status. In patients with NSTEMI and STEMI, aspirin and β blockers were prescribed at similar rates but patients with previous stroke were less likely to receive clopidogrel and statins but more likely to be prescribed warfarin ( Table 3 ).
| STEMI | NSTEMI | |||||
|---|---|---|---|---|---|---|
| Previous Stroke | No Previous Stroke | p Value | Previous Stroke | No Previous Stroke | p Value | |
| Aspirin | 97.3% | 98.5% | 0.0420 | 95.0% | 97.0% | <0.0001 |
| Clopidogrel | 81.2% | 88.0% | <0.0001 | 57.1% | 61.6% | 0.0002 |
| β Blocker | 95.2% | 96.6% | 0.0561 | 92.5% | 93.6% | 0.0741 |
| Angiotensin-converting enzyme inhibitor or angiotensin II receptor blocker | 54.7% | 58.4% | 0.0744 | 55.1% | 51.0% | 0.0002 |
| Statin | 62.4% | 68.9% | 0.0001 | 57.2% | 59.2% | 0.0994 |
| Glycoprotein Iib/IIIa inhibitor | 60.9% | 76.2% | <0.0001 | 32.0% | 46.8% | <0.0001 |
| Unfractionated heparin | 72.6% | 79.2% | <0.0001 | 54.2% | 56.9% | 0.0139 |
| Low-molecular-weight heparin | 19.9% | 16.0% | 0.0035 | 40.7% | 40.6% | 0.8937 |
| Direct thrombin inhibitor | 12.2% | 10.9% | 0.1444 | 9.7% | 11.9% | 0.0051 |
| Medication overdose | ||||||
| Unfractionated heparin | — | — | — | 30.3% | 27.2 | 0.0231 |
| Low-molecular-weight heparin | — | — | — | 13.8% | 13.5% | 0.8271 |
| Glycoprotein IIb/IIIa inhibitor | 24.1% | 8.2% | <0.0001 | 20.6% | 9.7% | <0.0001 |
| STEMI | NSTEMI | |||||
|---|---|---|---|---|---|---|
| Previous Stroke | No Previous Stroke | p Value | Previous Stroke | No Previous Stroke | p Value | |
| Aspirin | 98.3% | 98.7% | 0.5749 | 96.5% | 97.3% | 0.0258 |
| Clopidogrel | 85.9% | 90.7% | 0.0001 | 70.1% | 74.5% | <0.0001 |
| Warfarin | 15.2% | 7.2% | <0.0001 | 15.0% | 8.3% | <0.0001 |
| β Blocker | 96.2% | 97.1% | 0.1729 | 95.3% | 95.5% | 0.4334 |
| Angiotensin-converting enzyme inhibitor or angiotensin II receptor blocker | 81.3% | 86.7% | 0.0645 | 74.3% | 74.3% | 0.8410 |
| Statin | 87.1% | 91.7% | <0.0001 | 82.8% | 86.6% | <0.0001 |
Patients with previous stroke presenting with STEMI were less likely to receive thrombolytics or primary PCI ( Table 4 ). The odds ratio for receiving any reperfusion therapy was significantly lower for patients with stroke after adjusting for baseline and presenting characteristics (odds ratio 0.76, 95% confidence interval 0.59 to 0.97). Furthermore, patients with a history of stroke had significantly longer door-to-needle times ( Table 4 ) and door-to-balloon times ( Table 4 ). After adjusting for differences in the 2 patient populations, those with history of stroke were less likely to have a door-to-balloon time <90 minutes (odds ratio 0.76, 0.58 to 0.98, p = 0.037). Of patients presenting with NSTEMI, those with previous stroke were less likely to undergo diagnostic catheterization ( Table 4 ) and were less likely to have PCI overall ( Table 4 ) even after adjusting for baseline and presenting characteristics (odds ratio 0.77, 0.70 to 0.86, p <0.001).
| Previous Stroke | No Previous Stroke | p Value | |
|---|---|---|---|
| ST-segment elevation myocardial infarction | |||
| Reperfusion (any method) | 87.3% | 94.9% | <0.0001 |
| Thrombolytic therapy | 10.0% | 16.3% | 0.0023 |
| Door to drug ≤30 minutes | 22.8% | 37.3% | 0.0179 |
| Arrival to thrombolytic (minutes) | 47.5 (30.0–72.0) | 35.00 (23.0–55.0) | 0.0042 |
| Primary percutaneous coronary intervention | 82.0% | 85.7% | 0.0480 |
| Door to balloon ≤90 minutes | 57.9% | 68.6% | <0.0001 |
| Arrival to primary percutaneous coronary intervention (minutes) | 81.0 (64.0–110.0) | 75.0 (58.0–97.0) | 0.0001 |
| Coronary artery bypass grafting | 9.6% | 8.2% | 0.1846 |
| Non–ST-segment elevation myocardial infarction | |||
| Catheterization | 83.3% | 92.8% | <0.0001 |
| Percutaneous coronary intervention | 42.2% | 56.9% | <0.0001 |
| Coronary artery bypass grafting | 13.3% | 14.1% | 0.3289 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
