Background. Percutaneous coronary intervention (PCI) is increasingly used in patients with high-risk baseline characteristics. A prior stroke may identify patients who have a higher risk for post-PCI complications. However, no comparative data exist on post-PCI outcomes of patients with or without prior stroke. Methods. Review of a PCI database of 9,088 consecutive PCIs from 7/97 to 12/02 identified 812 PCIs in patients with a history of prior stroke and 8,044 PCIs without prior stroke. Results. Patients with prior stroke had high-risk baseline characteristics [diabetes, hypertension, hyperlipidemia, smoking, peripheral arterial disease, congestive heart failure, chronic renal failure, history of prior myocardial infarction and prior coronary artery bypass graft (CABG)] and high-risk coronary anatomy (p < 0.001 for each one). The triple composite (death, myocardial infarction and emergent CABG) and the triple composite plus post-PCI stroke were higher in patients with prior stroke (11.2% vs. 4.8%; p < 0.001; z = 7.617 and 12.1% vs. 5.0%; p < 0.001; z = 8.271, respectively. Conclusion. Patients with prior stroke constitute a high-risk PCI cohort with higher rates of in-hospital adverse events. A prior stroke history should be considered in evaluating potential candidates for PCI.

       Percutaneous coronary intervention (PCI) is increasingly used in patients with high-risk baseline characteristics.^1–4 A prior stroke may identify patients who are at higher risk for post-PCI complications. However, no comparative data exist on post-PCI outcomes of patients with or without prior stroke. The benefits of PCI among these sub-groups may be partially offset by increased complication rates. Recent studies have reported improvements in angiographic and clinical success rates of PCI among high-risk groups, largely due to the availability of stents and glycoprotein IIb/IIIa (GPIIb/IIIa) receptor inhibitors.^5,6 Nevertheless, GPIIb/IIIa inhibitors are cautiously used in patients with prior stroke due to the fear of intracranial bleeding (ICH). The objective of this study was to identify the frequency of prior stroke among consecutive PCIs and to determine the association of cardiovascular risk factors and PCI-related ischemic complications in patients with prior stroke.

Methods
       Data collection and study groups. A retrospective review of the database from July 1997 to December 2002 (9,088 consecutive PCIs performed on 7,264 patients) identified 879 PCIs in patients with a history of prior stroke and 8,209 PCIs in patients without prior stroke. Data for all PCIs performed at Wake Forest University School of Medicine are prospectively entered into a database during the index admission (CAOS, IBS, Winston-Salem, North Carolina). Individual review of the records done by a research fellow on patients with reported prior stroke confirmed prior stroke in 687 patients (812 PCIs), and asymptomatic carotid disease in the remaining 56 (67 PCIs) patients who were excluded from the study. Among the group of patients without a history of stroke, 144 patients (165 PCIs) were excluded due to incomplete history. The final study groups were: no prior stroke = 8,044 PCIs (6,377 patients) and prior stroke = 812 PCIs (687 patients) (9.2% of total PCIs). Out of these 812 PCIs, 108 PCIs were performed on 98 patients who had recent (< 2 years) non-ICH stroke and 20 PCIs on 14 patients who had prior ICH stroke. Patients were evaluated in the study under an institutionally approved review protocol.
       Definitions. Renal insufficiency was defined as a pre-procedural serum creatinine > 2 mg/dL, congestive heart failure (CHF) as NYHA class >= II symptoms. Peripheral arterial disease (PAD) was based on objective evidence such as peripheral vascular surgery or noninvasive lower arterial studies. Diabetes, hypertension and hyperlipidemia were defined as a previous medical diagnosis, or the use of medications or fasting blood sugar > 126 mg/dl, or admission blood pressure > 140/90 mmHg, or fasting total cholesterol > 200 mg/dl. Indications for PCI included ST-elevation myocardial infarction (MI) (primary or rescue), non-ST-elevation MI, unstable angina or stable coronary artery disease (CAD) according to standard ACC definitions.
       All PCIs were performed with standard techniques through the femoral approach. Hemostasis was achieved with a closure device whenever possible based on the access site angiogram. Coronary arteriography was performed in multiple right and left anterior oblique projections with cranial and caudal angulations for visualization of all segments of the coronary arteries. Biplane left ventriculography was performed in a 500 left anterior oblique and 300 right anterior oblique projections.
       Medications. All patients were treated with 325 mg of daily aspirin prior to the procedure and were advised to continue its use indefinitely. Most of the patients were also treated with thienopyridine antiplatelet drugs, either ticlopidine 250 mg b.i.d. or clopidogrel 300 mg load followed by 75 mg daily. Most of the patients received clopidogrel 3 to 24 hours prior to PCI. GP IIb/IIIa inhibitors were used in 95% of the PCIs (abciximab in 89% of the total PCIs). GPIIb/IIIa inhibitors were used during 7,671 PCIs (95.4%) in the no prior stroke group and 787 PCIs (96.9%) in the prior stroke group. During PCI, AB was administered as 0.25 mg/kg intravenous bolus followed by an infusion of 0.125 mcg/kg/min. Abciximab was administered from <= 1 hour before, until 12 hours after the PCI. All patients received unfractionated heparin (UFH). The UFH was given as standard bolus dose, weight-adjusted heparin regimen, with additional boluses as necessary to achieve and maintain activated clotting times around 225–275 seconds during the procedure. Heparin was discontinued after the PCI.
       Outcome. Creatine kinase (CK) and CK-MB measurements were performed every 8 hours post-PCI for a total of 3 assays or until discharge. Angiographic success was defined as the achievement of a minimum stenosis diameter < 50% in the presence of grade 3 TIMI flow. In-hospital outcomes included post-PCI stroke, all-cause mortality, myocardial infarction and emergent coronary artery bypass graft (CABG). ST-elevation MI was defined as new Q-waves in at least 2 contiguous leads. Non-ST-elevation MI was defined as elevation of the CK-MB isoenzyme to at least 3 times the upper limit of normal in at least 2 blood samples. In-hospital outcomes were analyzed for the occurrence of post-PCI stroke, a triple “ischemic” endpoint (death, MI and emergent CABG), and major adverse cardiac events (MACE) [the triple composite plus post-PCI stroke].
       Stroke Definitions. Patients with suspected strokes were identified from case report forms for each study. Clinical notes, discharge summary, neurological consultation notes and CT/MRI reports were collected for all patients. A fixed neurologic deficit was recorded as prior stroke. Using these criteria, prior stroke was classified as ICH, recent (< 2 years) non-ICH, and remote (> 2 years) non-ICH stroke.
       Post-PCI stroke was defined as abrupt onset of a new nonconvulsive, focal, neurologic deficit with residual symptoms lasting more than 24 hours that was confirmed by a neurologist, and CT/MRI findings occurred within 48 hours of the PCI which includes TIA (neurologic deficits < 24 hours). Strokes were classified as ischemic or hemorrhagic based on CT/MRI analysis performed by experienced neuroradiologists.
       Statistical analysis. Dichotomous variables were presented as percentages, comparisons made using the chi-square analysis or the Fisher’s exact test. Continuous variables were expressed as mean ± standard deviation, with comparisons using the Student’s t-test. The multivariate logistic regression model was used to compare the various independent predictors for MACE (the model included age, hypertension, hyperlipidemia, diabetes, prior MI, prior CABG, prior stroke and CHF). Since the primary aim of the study was to determine the influence of prior stroke on clinical outcomes after PCI, all comparisons were made in terms of PCIs, not patients. A p-value < 0.05 was considered significant. Statistical analysis was performed using SIGMASTAT version 2 statistical software (SPSS, Inc., Chicago, Illinois).

Results
       Patient characteristics. Baseline demographics and clinical characteristics of patients with and without prior stroke are shown in Table 1. Patients with prior stroke were older and had a significantly higher prevalence of diabetes, hypertension, hyperlipidemia, smoking, PAD, prior history of MI, CABG, CHF, renal insufficiency, left main disease, proximal left anterior descending artery disease, and multi-vessel disease (p < 0.001 for each one).
In-hospital outcomes. Rates for all ischemic complications following PCI were higher in patients with prior stroke; however, only non-ST MI reached statistical significance (6.2% vs. 1.9%; p < 0.001; z = 7.677). The triple composite and the triple composite plus post-PCI stroke were higher in patients with prior stroke (11.2% vs. 4.8%; p < 0.001; z = 7.617 and 12.1% vs. 5.0%; p < 0.001; z = 8.271, respectively) (Table 2).
       In a multivariate analysis, along with other risk factors, MACE was significantly higher in patients with prior stroke (Table 3). Age and diabetes were the only clinical variables that were associated with elevated risk for MACE (Table 3).
       Post-PCI stroke. Out of 8,856 PCIs, 37 (0.42%) strokes were reported. Stroke occurred in 24 (0.30%) of the 8,044 PCIs performed on 6,377 patients without a prior stroke and 13 (1.6%) of the 812 PCIs performed on 687 patients (p < 0.001; z = 5.179) with a prior stroke. Both non-ICH and ICH stroke were higher in patients with prior stroke (1.36% vs. 0.27%; p < 0.001; z = 4.573 and 0.25% vs. 0.025%; p = 0.046; z = 1.999, respectively) (Table 2). Out of 20 patients who had prior history of ICH, none of them had a post-PCI stroke. Among 108 PCIs performed on 98 patients with recent non-ICH stroke history, none of them had post-PCI-ICH stroke and 2 of them had post-PCI, non-ICH stroke.
       CT and MRI were performed in all patients who had post-PCI stroke. ICH stroke was documented in 4 patients (10.8% of total stroke), most often presenting as intra-parenchymal bleeding (75%). Out of 33 non-ICH strokes (89.2% of total strokes), 11 were TIA (CT/MRI = normal) and 22 involved the territory of major vessels [middle cerebral artery, 12 (54.5%); posterior cerebral artery, 9 (40.9%) and multiple territories, 1 (4.5%)].
       Out of 37 post-PCI strokes, 26 strokes were associated with severe hypotension requiring treatment following PCI; 15 due to procedural complications (4 emergent CABG; 7 ventricular fibrillation; 4 left ventricular failure), 7 due to acute-ST MI on presentation with primary PCI requiring emergency IABP therapy, and 4 due to hematoma at the catheter site with severe hypotension. Of the 4 patients with ICH stroke, all were elderly (76, 78, 80 and 82 years of age) and hypertensive, and 2 had received an oral anticoagulant for atrial fibrillation.
       Sub-group analysis. As shown in Table 4, post-PCI stroke occurred in distinct sub-groups of patients (age > 75 years, female gender, hypetension, PAD, CHF, LVEF < 40%, patients with prior stroke, patients who had in-hospital non-ST MI and emergent CABG). Patients who had emergent CABG had the highest incidence of post-PCI stroke [3 post-PCI strokes out of 26 post-PCI emergent CABG vs. 34 post-PCI strokes out of 8,830 PCIs without post-PCI emergent CABG (11.5% vs. 0.39%, p < 0.001)]. Finally, patients with post-PCI stroke had higher adverse in-hospital outcomes (Table 5).

Discussion
       Our analysis is the first to report significant differences in baseline characteristics and post-PCI complications between patients with or without prior stroke. Patients with a prior stroke clearly represented a high-risk PCI cohort, with a higher prevalence of cardiovascular risk factors and higher-risk coronary anatomy. The rates of all ischemic complications following PCI were higher in patients with prior stroke.
       Overall, our stroke rate (0.42%) in the combined study population was slightly higher than the reports from the Cleveland Clinic (0.3%), national Canadian report (0.25%), and the 30-day stroke rate of 0.36% observed with abciximab.^4,7,8 Fuchs et al. showed that the rate of peri-procedural stroke in an unselected group of patients undergoing PCI was 0.38%.⁹ Patients with prior recent stroke (ischemic stroke within 2 years) were excluded from enrollment in these studies and, furthermore, patients with prior stroke constituted only 2% of the total study populations.⁷ On the other hand, the incidence of ICH stroke (10.8% of total stroke) in our study is lower when compared with other trials.^7–9
       We found an increased incidence of post-PCI stroke (1.6%) (both ICH and non-ICH) in patients with prior stroke that has not been previously reported.^8,9 A history of prior stroke has been identified as a major predictor for stroke after CABG.^10–12 Post-PCI stroke occurred in a distinct group of patients: those with age > 75 years, females, history of prior stroke, hypertension, CHF, low LVEF and PAD. These findings are consistent with the established risk factors for both ICH and non-ICH stroke among diverse populations undergoing PCI.^8,9,13,14 Risk factors that were not found in our study but have been reported by others include diabetes and unstable angina for post-PCI stroke.⁹ Poor left ventricular function has been identified as a risk factor for stroke after CABG.¹⁵ We confirmed this association in stroke occurring after PCI. Consistent with earlier reports, we found that patients who had post-PCI emergent CABG had a higher incidence of post-PCI stroke.¹⁶ The non-ST MI following PCI could contribute to hemodynamic instability and aggravate cerebral hypoperfusion.⁷ This can explain the higher incidence of post-PCI stroke in patients who had in-hospital non-ST MI in our study. We also observed that 70% of the post-PCI stroke patients had post-procedural severe hypotension. Cerebral vascular hemodynamic impairment in these patients would increase the risk of ischemic stroke due to decreased cerebral perfusion.¹⁷
       Stroke was accompanied by significant mortality and morbidity. Patients with in-hospital stroke had a higher rate of death, non-ST MI, emergent CABG, and the triple composite endpoint.
       Limitations. The present study carries the inherent limitations of a retrospective analysis. Our major limitation is that we do not have data using alternative therapies (medical therapy or CABG) on post-PCI stroke or PCI-related complications in patients with prior stroke. Despite the improvement in surgical techniques and cardioplegic agents, the stroke rate is significantly higher in patients undergoing CABG, especially those with prior cerebrovascular disease.^18–20 These historic data support the concept that PCI is associated with less risk in patients with prior stroke compared to CABG. We do not have data on statin therapy. Since hyperlipidemia was greater in the patients with prior stroke, statin data would be interesting. We found that there was no differences in timing of ADP inhibitors between the two groups; however, we did not evaluate the difference in clinical outcomes with respect to timing of ADP inhibitor administration. We do not have data on what percentage of patients in either group were on chronic treatment with ADP inhibitors. Accordingly, we did not evaluate the effect of chronic treatment with ADP inhibitors on post-PCI stroke. Our database does not have ACT values for all patients. Finally, our database does not include the number of patients who had hypotension following PCI. The research fellow reviewed the medical records of patients who had post-PCI stroke. Due to this, we were not able to provide the data on the number of individuals who had hypotension without developing post-PCI stroke. Due to the overall low incidence of post-PCI stroke, the power of the study to detect small differences between groups as well as the identification of the other predictors for stroke is limited.

Conclusion
       Patients with prior stroke constitute a high-risk PCI cohort with a higher prevalence of cardiovascular risk factors, high-grade CAD, and experience a higher rate of adverse in-hospital clinical outcomes. Prior stroke history was associated with an increased incidence of post-PCI stroke. A prior stroke history should be considered in evaluating potential candidates for PCI.

1. Smith SC Jr, Dove JT, Jacobs AK, et al. ACC/AHA guidelines for percutaneous coronary intervention (revision of the 1993 PTCA guidelines). Circulation 2001;103:3019–3041.
2. Maier W, Windecker S, Boersma E, et al. Evolution of percutaneous transluminal coronary angioplasty in Europe from 1992–1996. Eur Heart J 2001;22:1733–1740.
3. Weintraub WS, Mahoney EM, Ghazzal ZM, et al. Trends in outcome and costs of coronary intervention in the 1990s. Am J Cardiol 2001;88:497–503.
4. Batchelor WB, Anstrom KJ, Muhlbaier LH, et al. Contemporary outcome trends in the elderly undergoing percutaneous coronary interventions: Results in 7,472 octogenarians. National Cardiovascular Network Collaboration. J Am Coll Cardiol 2000;36:723–730.
5. Singh M, Mathew V, Garratt KN, et al. Effect of age on the outcome of angioplasty for acute myocardial infarction among patients treated at the Mayo Clinic. Am J Med 2000;108:187–192.
6. Singh M, Rihal CS, Berger PB, et al. Improving outcome over time of percutaneous coronary interventions in unstable angina. J Am Coll Cardiol 2000;36:674–678.
7. Brown DL, Topol EJ. Stroke complicating percutaneous coronary revascularization. Am J Cardiol 1993;72:1207–1209.
8. Akkerhuis KM, Deckers JW, Lincoff AM, et al. Risk of stroke associated with abciximab among patients undergoing percutaneous coronary intervention. JAMA 2001;286:78–82.
9. Fuchs S, Stabile E, Kinnaird TD, et al. Stroke complicating percutaneous coronary interventions: Incidence, predictors, and prognostic implications. Circulation 2002;106:86–91.
10. Stamou SC, Hill PC, Dangas G, et al. Stroke after coronary artery bypass: incidence, predictors, and clinical outcome. Stroke 2001;32:1508–1513.
11. Herlitz J, Wognsen GB, Haglid M, et al. Risk indicators for cerebrovascular complications after coronary artery bypass grafting. Thorac Cardiovasc Surg 1998;46:20–24.
12. Rao V, Christakis GT, Weisel RD, et al. Risk factors for stroke following coronary bypass surgery. J Card Surg 1995;10(Suppl 4):468–474.
13. Mahaffey KW, Harrington RA, Simoons ML, et al. Stroke in patients with acute coronary syndromes: incidence and outcomes in the platelet glycoprotein IIb/IIIa in unstable angina. Receptor suppression using integrilin therapy (PURSUIT) trial. Circulation 1999;99:2371–237.
14. De Jaegere PP, Arnold AA, Balk AH, et al. Intracranial hemorrhage in association with thrombolytic therapy: Incidence and clinical predictive factors. J Am Coll Cardiol 1992;19:289–294.
15. Mickleborough LL, Walker PM, Takagi Y, et al. Risk factors for stroke in patients undergoing coronary artery bypass grafting. J Thorac Cardiovasc Surg 1996;112:1250–1258.
16. Cronin L, Mehta SR, Zhao F, et al. Stroke in relation to cardiac procedures in patients with non-ST-elevation acute coronary syndrome: A study involving > 18,000 patients. Circulation 2001;104:269–274.
17. Weiller C, Ringelstein EB, Reiche W, et al. Buell U: Clinical and hemodynamic aspects of low-flow infarcts. Stroke 1991;22:1117–1123.
18. Furlan AJ, Sila CA, Chimowitz MI, Jones SC. Neurologic complications related to cardiac surgery. Neurol Clin 1992;10:145–166.
19. Hogue CW, Murphy SF, Schechtman KB, Davila-Roman VG. Risk factors for early or delayed stroke after cardiac surgery. Circulation 1999;100:642–647.
20. Roach GW, Kanchuger M, Mangano CM, et al. Adverse cerebral outcomes after coronary bypass surgery. N Engl J Med 1996;335:1857–1863.