Distal Myocardial Protection During Percutaneous Coronary Intervention With an Intracoronary β-Blocker
Background— Experimental studies have demonstrated that intravenous β-blocker administration before coronary artery occlusion significantly reduces myocardial injury. Clinical studies have shown that intracoronary (IC) propranolol administration before percutaneous coronary intervention (PCI) delays myocardial ischemia. The present study tested the hypothesis that IC propranolol treatment protects ischemic myocardium from myocardial damage and reduces the incidence of myocardial infarction (MI) and short-term adverse outcomes after PCI.
Methods and Results— Patients undergoing PCI (n=150) were randomly assigned in a double-blind fashion to receive IC propranolol (n=75) or placebo (n=75). Study drug was delivered before first balloon inflation via an intracoronary catheter with the tip distal to the coronary lesion. Biochemical markers were evaluated through the first 24 hours and clinical outcomes to 30 days. Evidence of MI with creatine kinase–MB elevation after PCI was seen in 36% of placebo and 17% of propranolol patients (P=0.01). Troponin T elevation was seen in 33% of placebo and 13% of propranolol patients (P=0.005). At 30 days, the composite end point of death, postprocedural MI, non–Q-wave MI after PCI hospitalization, or urgent target-lesion revascularization occurred in 40% of placebo versus 18% of propranolol patients (hazard ratio 2.14, 95% CI 1.24 to 3.71, P=0.004).
Conclusions— IC administration of propranolol protects the myocardium during PCI, significantly reducing the incidence of MI and improving short-term clinical outcomes.
Received February 20, 2003; accepted February 26, 2003.
Myocardial necrosis (infarction) during percutaneous coronary intervention (PCI) may occur in almost half of otherwise successful PCI1–4 and is associated with an increased incidence of late adverse outcomes, particularly death, even with minor elevations in biochemical markers.2–5 The causes of myocardial enzyme elevation after otherwise successful PCI include the development of mechanical complications such as slow flow, side-branch occlusion, transient major vessel closure, or prolonged coronary spasm. In many cases, however, enzyme release occurs without any apparent mechanical complication. In this setting, the most likely cause is microembolization.6
Experimental studies have shown that intravenous administration of the β-blocker propranolol before coronary artery occlusion can significantly reduce the electrocardiographic, enzymatic, and histological indices of myocardial infarction (MI).7–10 Clinical PCI studies have shown that intracoronary (IC) propranolol delays the development of ischemia during balloon occlusion as determined by the time to and extent of ST-segment elevation on IC and surface ECGs.11,12 Currently, there are no data evaluating the impact of adjunctive IC β-blocker therapy before PCI on postprocedural MI, as reflected by biochemical marker release, and on clinical outcomes. The present study tested the hypothesis that IC propranolol protects ischemic myocardium during PCI and reduces MI and early adverse clinical events after PCI.
This study was a randomized, double-blind, placebo-controlled trial. Patients with coronary artery disease, including those receiving chronic β-blockade, who were undergoing PCI were considered for inclusion. Exclusion criteria were MI <24 hours previously or recent MI with creatine kinase (CK) or CK-MB elevation at time of randomization (a normal CK and CK-MB and elevated troponin T allowed inclusion); cardiogenic shock; systolic blood pressure <100 mm Hg; heart rate <50 bpm; New York Heart Association class III or IV heart failure; severe left ventricular dysfunction (ejection fraction <30%); severe renal insufficiency (creatinine >3.0 mg/dL); allergy to propranolol; or second planned intervention within 30 days. Patients with chronic lung disease were not specifically excluded. The protocol was approved by the University of Texas Medical Branch Biomedical Institutional Review Board, and informed consent was obtained from all patients.
Before PCI, patients were pretreated with aspirin and weight-adjusted heparin, with a target activated clotting time of >300 seconds in the absence of a glycoprotein (GP) IIb/IIIa inhibitor and 200 to 300 seconds with a GP IIb/IIIa inhibitor, which was allowed but not mandated.
Patients were randomized in a 1:1 ratio to receive either placebo or propranolol. The randomization allocation code was generated by complete block design. Randomization was performed after coronary angiography confirmed the need for PCI. In patients with a total occlusion, the lesion was crossed with a guidewire before randomization.
Study drug kits were packaged by the hospital pharmacy to be indistinguishable from each other with a concentration of 1 mg/mL; the injectant identity (ie, propranolol or placebo) was not known to the investigators or patients, and equal volumes were injected. Propranolol at a dose of 15 μg/kg12 or 0.9% NaCl (placebo) was injected into the coronary artery through the dilatation catheter, the distal tip of which was positioned across the stenosis or stenoses (if more than 1 site underwent PCI). The study drug was thus delivered directly into the myocardial region supplied by the stenotic coronary artery undergoing PCI. Balloon inflation was performed after the IC study drug injection. Blood samples were taken before and immediately after PCI, every 8 hours for 24 hours to assay CK and CK-MB, and at 16 hours for troponin T. Additional blood levels were obtained when any patient experienced possible ischemic symptoms. The 30-day contact was made by clinic visit or telephone interview.
Study End Points
The primary end point was the incidence of postprocedural MI as evidenced by CK-MB elevation. The threshold for an excess in adverse events relative to CK-MB level above normal appears to be a continuous function beginning with any elevation of CK-MB or troponin T, with the frequency of long-term adverse events, particularly death, increasing with the level of enzyme rise.2 Secondary end points included (1) incidence of postprocedural total CK and troponin T elevation; (2) median peak values of CK, CK-MB, and troponin T; (3) rescue therapy with GP IIb/IIIa inhibitors; and (4) combination of death of any cause, post-PCI MI, non–Q-wave MI, and Q-wave MI after PCI hospitalization or severe myocardial ischemia requiring urgent coronary artery surgery or target-lesion revascularization (TLR) within 30 days of intervention.
Evidence of post-PCI MI was defined as an increase above the normal limit in total CK, CK-MB, or troponin T. Non–Q-wave MI after PCI hospitalization was defined as an acute coronary syndrome with or without ECG changes and elevation of a cardiac marker.13 Urgent TLR included any coronary artery bypass surgery or a second PCI on the original target lesion performed for recurrent myocardial ischemia. Successful PCI was defined as residual stenosis of the target lesion <30% of reference vessel diameter without major adverse outcome (death, Q-wave MI, or emergent CABG). Major bleeding was defined as bleeding that necessitated transfusion, surgery, prolongation of hospital stay, or >8% drop in hematocrit after PCI. Rescue GP IIb/IIIa use was defined as drug given after the guidewire passage during PCI. Standardized definitions were used for risk factors.14 Mechanical complications included dissection, abrupt or threatened closure of PCI vessel, transient or permanent side-branch closure, angiographic embolization, development of new intracoronary angiographic thrombus, decrease in flow (“slow flow”), or stoppage of flow (“no flow”) in the epicardial PCI vessel in the setting of an adequately opened lesion.
Data Management and Statistical Analysis
Data were collected prospectively on case-report forms with investigators and study coordinators blinded to treatment assignment until after the database was sealed after entry of all data. All analyses were performed with the intention-to-treat principle. Continuous variables were compared between groups by the 2-tailed t test and categorical variables by the χ2 statistic or Fisher exact test wherever appropriate. Treatment effects by subgroups are reported as relative risks (RRs) with 95% CIs. Multivariate logistic regression analysis was used to determine the independent predictors of the primary end point, and variables shown in Figure 4 were included in the model. Methods for survival analysis were used for the 30-day combined end point. Time to first occurrence of the composite end point is shown by Kaplan-Meier survival curves.
Sample-Size Determination and Interim Analyses
A sample size of 150 patients, with 75 patients in each group, was required to have 80% power based on the assumption of an incidence of postprocedural CK-MB increase of 8% in the propranolol group and 24% in the control group.2 Because no published reports are available for the expected incidence of the primary end point in the propranolol group, 2 blinded interim analyses at 50 and 100 samples were planned to determine whether there was a trend worthy of trial continuation. The O’Brien-Fleming multiple testing method was used to determine that a significant difference at the final analysis of 150 patients was considered to be reflected by a probability value <0.045.
Interim analysis after 50 patients revealed a difference between the 2 groups in CK-MB elevation after PCI (19% versus 45%; P=0.04). The interim analysis after 100 patients showed a greater statistical difference (P=0.009), 18% and 42%, respectively. The incidence of serious end points other than myocardial necrosis, including death, non–Q-wave MI after PCI hospitalization, and urgent TLR, over the first 30 days was very low, and there was no significant difference between the 2 groups in these end points at the first and second interim analyses. We therefore chose to continue the double-blind study until the original sample size was reached in order to increase the robustness of the final results.
Baseline Characteristics of the Patients
From July 18, 2000, to May 30, 2002, 150 patients were enrolled, with the last follow-up on July 1, 2002. Baseline demographic and angiographic characteristics between groups were similar (Tables 1 and 2⇓). There was no change in the heart rate or blood pressure after IC injection of propranolol or placebo. There were no adverse systemic effects after IC injection in either group, including respiratory difficulties in 5 patients with a history of chronic obstructive pulmonary disease or asthma.
Post-PCI MI occurred significantly more frequently in the placebo than the propranolol group. CK-MB elevation was seen in 36% of placebo patients and 17% of propranolol patients (P=0.01). Troponin T elevation occurred in 33% of placebo patients and 13% of propranolol patients (P=0.005; Table 3; Figure 1). The median peak values after PCI of CK-MB, CK, and troponin T in the placebo and propranolol groups, respectively, were as follows: CK-MB, 2.6 and 2.1 U/L; total CK, 76 and 67 U/L; and troponin T, 0.03 and 0.03 ng/mL.
Patients with mechanical complications had a higher incidence of elevated CK-MB, total CK, and troponin T than patients without mechanical complications (Table 4). Among patients without mechanical complications, there was a highly significant difference in the incidence of CK-MB elevation with placebo compared with propranolol treatment (28% for placebo versus 10% for propranolol; P=0.015). Among patients with mechanical complications (n=35), there was a trend toward a lower incidence of post-PCI MI in the propranolol group (61% for placebo versus 41% for propranolol; P=0.24).
CK-MB elevation ≥3 times the upper limit of normal, often used as a criterion for post-PCI MI, developed in 7 propranolol patients and 6 placebo patients. Of these 13 patients, 11 (84.6%) had an associated mechanical complication during PCI. The incidence of mechanical complications was significantly greater in patients with CK-MB ≥3 times the upper limit of normal than in all other patients (84.6% versus 17.5%; P<0.0001).
The 30-day composite end point (Figure 2) occurred in 40% of placebo and 18% of propranolol patients (P=0.004). This difference was due almost exclusively to the difference in post-PCI MI. There were no deaths. Non–Q-wave MI after PCI hospitalization occurred in 3 patients (placebo 2, propranolol 1). Urgent TLR was required in 2 placebo patients. “Rescue” GP IIb/IIIa inhibitor use was required in 15% of placebo and 11% of propranolol patients (P=0.46). Significant bleeding occurred in 3% each of the placebo and propranolol patients. Subgroup analysis demonstrated that the favorable effect of propranolol on preventing MI was broadly applicable (Figure 3).
Factors associated with increased risk of CK-MB elevation after PCI are shown in Figure 4. IC β-blocker therapy was associated with a decreased risk of mechanical complications during PCI, and age ≥60 years was associated with an increased risk of CK-MB release.
The principal study finding is that IC administration of propranolol decreases the incidence of myocardial infarction after coronary intervention compared with placebo. Experimental data have strongly suggested that intravenous β-blocker administration can decrease infarct size if an adequate amount of β-blocker can be delivered to the ischemic zone before irreversible damage.7,9,10 The largest amount of myocardial salvage occurs if the β-blocker is given before coronary occlusion, although animal data have demonstrated some degree of myocardial salvage when intravenous β-blocker is given as long as 3 to 4 hours after coronary occlusion.10,15 Furthermore, studies using transient occlusion followed by reperfusion tend to show more myocardial salvage than preparations with permanent occlusion.15,16 In experimental studies, the use of intravenous β-blocker has produced significant decreases in heart rate, blood pressure, and cardiac output.7 In clinical trials, intravenous β-blockers during MI (as described most extensively in the prethrombolytic era, when an infarct-related artery was likely to be patent in <50% of patients) have been given, on average, at least several hours after presentation (frequently exceeding 4 hours) in a dose usually one tenth or less of that given in animal studies. Thus, it is not surprising that myocardial salvage after MI with intravenous β-blocker has shown only modest clinical benefit.17–20
Human studies have demonstrated that IC propranolol before transient coronary artery occlusion can delay the onset and severity of myocardial ischemia as detected by surface electrocardiography without decreasing heart rate or blood pressure.11,12 The present study demonstrates that IC β-blocker can also protect against myocardial injury in this setting. We propose that the effects of microembolization during PCI are often transient and that reperfusion of small myocardial areas occurs after dissolution of these microemboli. This scenario is ideal for β-adrenergic blockade by the method used in the present study. IC administration of propranolol before PCI allows for a high local concentration and minimal, if any, systemic effects, changes in heart rate, or myocardial depression in the non-PCI territory. With the expectation of ultimate revascularization, this method provides an enlarged window of safety during the performance of PCI.
Patients with mechanical complications during PCI demonstrated an increased risk of CK-MB and troponin T elevations, as has been reported previously.21 Mechanical complications have been shown to almost totally negate the cardioprotective effects of GP IIb/IIIa inhibition during PCI.22 In the present study, there was a trend toward a cardioprotective effect of propranolol in patients with mechanical complications. The present study, however, was not powered to specifically evaluate this subgroup.
One may ask whether intravenous or oral propranolol also may be cardioprotective during PCI. Previous studies have shown only a modest benefit of intravenous β-blocker given before PCI in delaying the onset and severity of regional ischemia as detected by surface ECG.23,24 The possible benefit from long-term oral β-blocker use was examined in an observational retrospective study that showed a reduction in the incidence of MI in patients who were taking long-term oral β-blocker therapy before PCI compared with those who were not.25 However, a second large retrospective analysis did not confirm this finding,26 nor did the present study (see Figure 3). In the present study, patients taking long-term oral β-blockers had a similar benefit to those who were not, demonstrating the additional benefit of IC propranolol (Figure 4).
The sample size in the present study is relatively small; the possibility that the results occurred by chance must be considered. The increasing level of statistical significance in the interim analyses strongly suggests that the results reflect a real phenomenon. The present results were collected at a single center, which raises the issue of whether the results are generalizable and reproducible. The fact that the patient population and results are typical of PCI study cohorts and that the percentage increase in myocardial necrosis in the placebo group is similar to that reported in previous studies22 suggests that these results are generalizable.
A controversial issue is the threshold at which postprocedural cardiac biomarker elevation is indicative of increased long-term risk of adverse events, particularly death. A recent meta-analysis of 8838 patients undergoing PCI in 5 large prospective trials has demonstrated that long-term mortality risk is a continuous function, increasing at any level above normal postprocedural CK-MB.4,5 Thus, a primary end point, MI as defined in the present study, has direct clinical relevance. It remains to be demonstrated that preventing post-PCI MIs can in fact decrease late mortality. However, should the ability to decrease myocardial necrosis translate into a survival benefit, data from the above-cited study4 and the present study suggest that for every 1000 patients undergoing PCI, 20 more patients will be alive at 6 months if IC propranolol is used. If one assumes that IC β-blocker is used routinely and is applicable to 80% of patients (on the basis of inclusion criteria from the present study), there will be 10 000 more PCI patients alive at 6 months in the United States alone if this therapy is used.27 It is important to emphasize that IC β-blocker in the dose used in the present study is extremely safe and without any adverse effects, even in patients with moderate left ventricular dysfunction and chronic lung disease. Although not specifically analyzed, cost is not likely to be significantly increased by this approach because the β-blocker used, propranolol, is relatively inexpensive. Thus, the use of IC β-blocker during PCI may significantly improve clinical outcomes and potentially decrease mortality and may do so cost-effectively. Further studies will be helpful in confirming the efficacy of this approach. The combined use of IC propranolol with the only presently available proven cardioprotective agents during PCI, the GP IIb/IIIa platelet receptor inhibitors, awaits further investigation.
The authors acknowledge the technical assistance of Nisha Wadhwa, RN, and Alecia Parks, RN. We thank Daniel H. Freeman, PhD; the fellows, nurses, and technicians in the UTMB Cardiac Catheterization Laboratory; and M’Linda Lasswell for their assistance in performing this investigation.
Simoons ML, Brand M, Lincoff M, et al. Minimal myocardial damage during coronary intervention is associated with impaired outcome. Eur Heart J. 1999; 20: 1112–1119.
Abdelmeguid AE, Topol EJ, Whitlow PL, et al. Significance of mild transient release of creatine kinase-MB fraction after percutaneous coronary interventions. Circulation. 1996; 94: 1528–1536.
Akkerhuis KM, Alexander JH, Tardiff BE, et al. Minor myocardial damage and prognosis: are spontaneous and percutaneous coronary intervention–related events different? Circulation. 2002; 105: 554–556.
Ellis SG, Chew D, Chan A, et al. Death following creatine kinase-MB elevation after coronary intervention. Circulation. 2002; 106: 1205–1210.
Topol EJ, Yadav JS. Recognition of importance of embolization in atherosclerotic vascular disease. Circulation. 2000; 101: 570–580.
Maroko PR, Kjekshus JK, Sobel BE, et al. Factors influencing infarct size following experimental coronary artery occlusion. Circulation. 1971; 43: 67–82.
Reimer KA, Rasmussen MM, Jennings RB. Reduction by propranolol of myocardial necrosis following temporary coronary occlusion in dogs. Circ Res. 1973; 33: 353–363.
Miura M, Thomas R, Ganz W, et al. The effect of delay in propranolol administration on reduction of myocardial infarct size after experimental coronary artery occlusion in dogs. Circulation. 1979; 59: 1148–1157.
Gold HK, Leinbach RC, Maroko PR. Regional blockade with intracoronary propranolol. Circulation. 1976; 54 (suppl II): II-173.Abstract.
Zalewski A, Goldberg S, Dervan JP, et al. Myocardial protection during transient coronary artery occlusion in man: beneficial effects of regional beta-adrenergic blockade. Circulation. 1986; 73: 734–739.
The Joint European Society of Cardiology/American College of Cardiology Committee. Myocardial infarction redefined: a consensus document of The Joint European Society of Cardiology/American College of Cardiology Committee for the Redefinition of Myocardial Infarction. Eur Heart J. 2000; 21: 1502–1513.
Wilson PW, D’Agostino RB, Levy D, et al. Prediction of coronary heart disease using risk factor categories. Circulation. 1998; 97: 1837–1847.
Fox K, Welman E, Selwyn A. Myocardial infarction in the dog: effects of intravenous propranolol. Am J Cardiol. 1983; 45: 769–774.
Peter T, Norris RM, Clarke ED, et al. Reduction of enzyme levels by propranolol after acute myocardial infarction. Circulation. 1978; 57: 1091–1095.
Boyle DM, Barber JM, McIlmoyle EL, et al. Effect of very early intervention with metoprolol on myocardial infarct size. Br Heart J. 1983; 49: 229–233.
Feldman RL, Macdonald RG, Hill JA, et al. Effect of propranolol on myocardial ischemia occurring during acute coronary occlusion. Circulation. 1986; 73: 727–733.
Sharma SK, Kini A, Marmur JD, et al. Cardioprotective effect of prior β-blocker therapy in reducing creatine kinase-MB elevation after coronary intervention. Circulation. 2000; 102: 166–172.
Ellis SG, Brener SJ, Lincoff M, et al. β-Blockers before percutaneous coronary intervention do not attenuate postprocedural creatine kinase isoenzyme rise. Circulation. 2001; 104: 2685–2688.
2002 Heart and Stroke Statistical Update. Biostatistical Fact Sheet; Miscellaneous: Cardiovascular Procedure. American Heart Association Web site. Available at: http://www.americanheart.org/downloadable/heart/104575420599FS18CVP3.pdf. Accessed May 15, 2003.