Pretreatment with Nitroprusside for Microcirculatory Protection in Saphenous Vein Graft Interventions

From the ^*Division of Cardiovascular Disease, University of Alabama at Birmingham, Birmingham, Alabama, ^†Birmingham VA Medical Center, Birmingham, Alabama, and ^‡Baptist Medical Center, Nashville, Tennessee. The authors report no conflicts of interest regarding the content herein. Presented in part at the Society for Cardiovascular Angiography and Interventions, 29th Annual Scientific Sessions, Chicago, Illinois on May 12, 2006, and at the American Heart Association Scientific Sessions, Chicago, Illinois on November 15, 2006. Address for correspondence: William B. Hillegass, MD, MPH, University of Alabama at Birmingham, 383 Boshell Diabetes Building, 1808 7th Avenue South, Birmingham, AL 35294-0012. E-mail: hillegas@uab.edu

________________________________________________
ABSTRACT: Objectives. We hypothesized that the prophylactic administration of sodium nitroprusside (NTP) during saphenous vein graft (SVG) PCI would ameliorate the detrimental effects of distal embolization and reduce the frequency and magnitude of post-procedural myonecrosis. Methods. Sixty-four consecutive patients with normal preprocedural cardiac enzymes underwent SVG PCI without embolic protection devices and received prophylactic intragraft NTP before initial device activation. For each case, 2 control patients were selected in reverse chronologic order and were matched for stent use, thromboatherectomy device use, clinical presentation, presence of thrombus and pre-PCI thrombolysis in myocardial infarction (TIMI) flow. Results. Mean patient age was 66 ± 10 years, 78% of whom were males. Stent and thromboatherectomy use was 95.3% and 3.1%, respectively in both groups (p = ns). Prior to intervention, TIMI 3 x the upper limit of normal (ULN) occurred in 6.3% of cases vs. 16.4% of controls (p = 0.049) and > 5 x ULN in 1.6% of cases vs.10.9% of controls (p = 0.022). In a multivariate regression model that included stent use, in-stent restenosis, thrombus presence, preprocedural TIMI 3 flow, MI as procedural indication, NTP and glycoprotein IIb/IIIa use, NTP was the only independent and significant predictor of reduced post-procedural CK-MB elevation > 5 x ULN. Conclusion. Prophylactic administration of intragraft NTP during SVG PCIs results in a lower frequency and magnitude of post-procedural cardiac enzyme elevation.

J INVASIVE CARDIOL 2009;21:34–39

Key words: saphenous vein graft, percutaneous coronary intervention, post procedural myonecrosis, nitroprusside, cardiac enzyme elevation

________________________________________________
Approximately 6% of all percutaneous coronary interventions (PCIs) involve treatment of a saphenous vein graft (SVG) lesion.¹ Diseased SVGs have more friable and thrombotic lesions compared to native coronary arteries.² Consequently, SVG PCIs carry a higher risk of procedural complications such as myocardial infarction (MI) and no-reflow resulting from distal particulate embolization and platelet clumping that mechanically obstruct the microcirculation, or due to spasm of the microcirculation from local responses to embolization.^2-6 The no-reflow phenomenon occurs in 8% and post-procedural non-Q-wave MI (NQWMI) occurs in 17–20% of SVG PCIs.^3,7,8 The use of embolic protection devices (EPDs) during SVG PCI to protect the microcirculation from distal embolization has decreased the incidence of post-procedural MI to 7–10%, which is still higher than that observed with native-vessel PCI.^7,8 Patients who experience the no-reflow phenomenon or NQWMI following SVG PCI have an increased risk of subsequent MI and death.^3,9,10 Sodium nitroprusside (NTP), a direct nitric oxide donor, vasodilates the microcirculation and has been shown to improve coronary blood flow in patients with no-reflow/slow-flow phenomena developing during PCI.¹¹ We hypothesized that the prophylactic administration of NTP during SVG PCI would ameliorate the detrimental effects of distal embolization on the microcirculation, particularly vasoconstriction, and reduce the frequency and magnitude of post-procedural myonecrosis.

Materials and Methods

Study design and patient selection. This was a retrospective, matched, case-control study of patients who underwent SVG PCIs at our institution without the use of EPD and who had normal preprocedural cardiac enzymes. The study group consisted of 64 consecutive patients who underwent SVG PCI and who received prophylactic intragraft NTP before device activation. For each case, 2 matched control patients from the same time frame were selected in reverse chronologic order from our SVG PCI database. Among the candidate control patients with normal preprocedural cardiac enzymes and without EPD use, our prespecified hierarchal matching criteria for the 2 controls for each case were: 1) presence or absence of stent implantation; 2) thromboatherectomy device use (AngioJet thrombectomy [Possis Medical, Inc., Minneapolis, Minnesota] and excimer laser); 3) clinical presentation (MI during this admission or no MI); 4) thrombus presence, and 5) pre-PCI thrombolysis in myocardial infarction (TIMI) flow in the graft (TIMI 3 or TIMI Study protocol. Heparin was given intravenously during all PCIs to achieve an activated clotting time > 250 seconds or > 200 seconds if a glycoprotein (GP) IIb/IIIa inhibitor was administered. GP IIb/IIIa inhibitors were administered according to operator discretion. Clopidogrel or ticlopidine were administered in patients who had a stent placed either pre- or post- PCI. Intragraft NTP (50­–300 µgm) was delivered in divided doses via the guiding catheter before initial device activation. As part of our routine care pathway in all PCI patients, cardiac enzymes were obtained 6–8 hours post procedure and then at approximately 8-hour intervals until two sets were normal or the creatinine kinase (CK)-MB level was declining. Angiographic analysis. The coronary angiograms were analyzed by 2 observers who were blinded to the clinical data. Quantitative coronary analysis was performed using an automated edge-detection algorithm. The outer diameter of the contrast-filled catheter was used as the calibration standard. The view showing the most severe stenosis was used to measure the pre- and post-lesion minimum luminal diameters (MLD) and reference vessel diameters (RVD). The percent (%) diameter stenosis was automatically calculated using the RVD and the MLD. Lesion lengths were measured using the least foreshortened view. Pre- and post-procedural flow was also assessed from the initial and final coronary angiograms using the standard TIMI flow criteria as well as assessment for the presence of the no-reflow phenomenon.¹² Steps involved in the interventional procedure were identified by reviewing the time-stamped procedural records and the cineangiograms. Procedural steps were defined as the use of angioplasty alone, stenting with or without angioplasty, or the use of atherectomy devices (extraction catheters, directional coronary atherectomy or AngioJet thrombectomy). Outcomes and definitions. The main outcome of the study is the incidence of post-PCI MI defined by the magnitude of CK-MB elevation. Post-procedural MI was defined by the two commonly accepted thresholds of CK-MB elevation > 3 x and > 5 x the upper limit of normal (ULN) following the procedure, whereas the CK-MB ULN in our laboratory is 5 ng/ml.¹²Statistical analysis. Statistical analysis was performed using SPSS software, version 10.0.5 (SPSS, Inc., Chicago, Illinois). Categorical data are presented as % and were compared between the two groups by the chi-square test. Continuous variables are presented as mean ± 1 standard deviation and were compared between the two groups using the unpaired t-test. The clinical, angiographic and procedural predictors of post-PCI MI (CK-MB > 3 x and > 5 x ULN) were determined using multivariate binary logistic regression analysis. A p-value Results The prespecified matching characteristics are shown in Table 1. When precise matches were not available, controls without the higher-risk characteristic were chosen, hence there is a slightly higher but nonsignificant frequency of the presence of thrombus and 3 x ULN occurred in 6.3% of patients treated with NTP versus 16.4% of controls (p = 0.049). CK-MB > 5 x ULN occurred in 1.6% of NTP cases versus 10.9% of controls (p = 0.022) (Figure 1 and Table 5). Self-limited hypotension has previously been described with the use of intracoronary NTP to treat the no-reflow phenomenon.11 In this series of prophylactic treatment in stable patients, hypotension with intracoronary NTP was transient and self-limited. No patient had a serious event related to hypotension or required treatment with pressors. A multivariate binary logistic regression model correcting for baseline differences between the groups that included stent use, ISR, thrombus presence, preprocedural TIMI 3 flow, MI as procedural indication, NTP and GP IIb/IIIa use showed that NTP was the only independent and significant predictor of reduced post-procedural CK-MB elevation > 5 x ULN with a RR = 0.11 (CI: 0.02–0.97; p = 0.046) and of borderline significance for > 3 x ULN with RR = 0.33 (CI: 0.10–1.05; p = 0.061). In a second more parsimonious model examining ISR and NTP, NTP was the only independent and significant predictor of reduced post-procedural CK-MB elevation > 5 x ULN with a RR = 0.12 (CI: 0.01–0.96; p = 0.045) and of borderline significance for > 3 x ULN with RR = 0.35 (CI: 0.11–1.09; p = 0.071).

Discussion

The main finding of this case-control study is that prophylactic NTP during SVG PCI without EPD resulted in a significant decrease in the frequency and magnitude of post-procedural MI. Prophylactic NTP use was the single independent predictor of reduced post-procedural MI defined as CK-MB > 5 x ULN. We chose myonecrosis rather than angiographic parameters as the endpoint of the study because the association between post-procedural cardiac enzyme elevation and subsequent clinical outcome is well defined. In addition, post-procedural enzyme elevation has been the accepted primary endpoint in evaluating other strategies such as EPD use in studies of SVG PCI. Also, there are no clearly accepted or validated quantitative angiographic criteria such as TIMI frame counts, myocardial blush scores or grading of no-reflow phenomenon in SVG PCI. SVG PCIs frequently result in distal embolization of particulate material and post-procedural myonecrosis.¹⁵ In a pooled analysis of all published SVG stent studies with systematic enzyme collection from 1995 to 2006, the risk of post-procedural myonecrosis defined as CK-MB elevation > 3 x ULN (n = 5,310) was 18.5 ± 2.1% and > 5 x ULN (n = 3,430) was 16.1 ± 3.3%,¹⁶ similar to the rates observed in the control group in this study. Even with the use of EPDs, the observed rate of CK-MB elevation > 3 x ULN (n = 1335) was 9.9 ± 0.7% compared to 6.6% for > 3 x ULN and 1.6% for > 5 x ULN in our study.¹⁶ Given the demonstrated efficacy of EPDs, we anticipated unprotected patients would have a higher event rate. This would be expected to improve our power to examine the hypothesis as to whether prophylactic microvascular vasodilator therapy would have any detectable effect on outcome in SVG PCI. A CK-MB > 3 x ULN was present in 16.4% of the control group patients compared to 18.5% in the pooled English language published experience. Since this is a retrospective study, the decision to not use an EPD was at the individual operator’s discretion. EPD use was limited by anatomic and lesion characteristics. In a study of 624 real-world patients, only 77% of patients had anatomy that was suitable for the use of these devices.¹⁷ Lack of suitable anatomy is also the typical reason for not using an EPD at our institution. Of 19,546 SVG PCI procedures in the American College of Cardiology-National Cardiovascular Data Registry from January 1, 2004, through March 30, 2006, EPDs were used in only 22% of patients.¹ Mechanisms of myocardial injury. The embolized material mainly arises from the necrotic plaque core and consists of cholesterol clefts, lipid-rich macrophages and fibrin material which obstruct the microcirculation, disrupt its endothelial integrity, release vasoactive amines from activated platelets, release oxygen-derived free radicals, potentiate platelet thrombus formation and cause vasospasm of the distal microcirculation.² This microvascular obstruction, spasm, and dysfunction further leads to myocardial necrosis and inflammation.^2,18 Additionally, analysis of blood aspirated during PCIs performed using the PercuSurge GuardWire EPD (Medtronic, Inc., Minneapolis, Minnesota) has shown a significant release of vasoconstrictive factors such as endothelin and serotonin, thrombotic factors such as tissue factor and plasminogen activator inhibitor, thrombin/antithrombin III complex, prothrombin fragment F1.2 and inflammatory factors such as soluble CD40 ligand and soluble P-selectin compared to pre-PCI.¹⁹ Aspirated blood during PCI with distal balloon occlusion was shown to have more vasoconstrictive substances compared to peripheral venous and arterial blood contents.²⁰ The marked release of soluble vasoconstrictors during SVG PCI explains why filter protection is not fully protective, suggesting the need for a pharmacologic vasodilator to protect the microcirculation from the adverse effects of the released factors. Mechanical strategies to reduce myocardial injury. The use of an EPD during SVG PCI remains the only mechanical strategy proven in randomized, controlled studies to significantly reduce the rate of post-PCI myonecrosis by trapping embolic material before it reaches the microcirculation. The use of covered stents such as the Symbiot (Boston Scientific Corp., Natick, Massachusetts) or Jostent (Abbott Vascular, Abbott Park, Illinois) did not significantly reduce distal embolization or post-procedural myonecrosis when compared to noncovered stents.^21,22 Similarly, there is no randomized trial evidence that thromboatherectomy devices reduce distal embolization or post-procedural myonecrosis. Despite the use of EPDs, the rate of myonecrosis remains higher than that observed with native-vessel PCI. Embolization can still occur while crossing the lesion and before deployment of the EPD due to inadequate seal or aspiration with balloon occlusion systems or incomplete apposition of filter-based devices.²³ The most contemporary large multicenter study (PRIDE) documented a 10.2% rate of CK-MB ≥ 3 x ULN with EPD use.²⁴Pharmacologic strategies to reduce myocardial injury. Antithrombotic strategies to reduce or mitigate the effects of distal embolization in SVG PCI have generally been disappointing. A pooled analysis of five randomized clinical trials did not show any short- or long-term clinical benefit with the use of GP IIb/IIIa inhibitors in SVG PCI.²⁵ In a study of 1,537 patients who underwent SVG PCI, the use GP IIb/IIIa inhibitors with or without an EPD did not reduce post-PCI myonecrosis after adjustment using a propensity model.²⁶ The use of bivalirudin with an EPD tended to decrease the incidence of myonecrosis as compared to heparin.²⁷ Fischell et al recently reported a series of SVG PCI patients receiving prophylactic intragraft nicardipine, a potent microvascular vasodilator.²⁸ Although no specifically matched control group was presented, nicardipine appeared to have a similar protective effect as we have observed with NTP in reducing the frequency and magnitude of post-procedural myonecrosis. In a small randomized study, 10 patients undergoing SVG PCI received 200 mg of intragraft verapamil and 12 patients received placebo immediately prior to PCI.²⁹ The verapamil group had 0% no-reflow compared to 33.3% in the placebo group, as well as better TIMI myocardial perfusion grade and TIMI frame counts.²⁹ However, there was no difference in the incidence of cardiac biomarker release following PCI.²⁹ The results of our study with a different potent vasodilator agent strongly support the more general hypothesis that prophylactic use of potent microvascular vasodilators may have a significant protective effect. While the half-lives of these agents during intraarterial and intracoronary administration are unknown, nicardipine has a significantly longer half-life as an intravenous (IV) agent (8 hours) than nitroprusside (2 minutes). Whether a longer IV half-life will translate into greater clinical efficacy with adequate safety for intracoronary use is unknown. The mechanism of this protective effect likely includes mitigating the effects of distal embolization since it seems unlikely that pharmacologic agents would reduce distal embolization like a distal protection device. In addition to directly preventing or reversing microvascular spasm, these agents may enhance endogenous protective mechanisms. Endogenous adenosine initially attenuates the effect of embolization until it becomes overwhelmed by the embolic mass after which its protective effect decreases and ischemia ensues.³⁰ Adenosine, verapamil and nicorandil have been shown to improve microvascular perfusion during acute MI with improvement in perfusion defects, ventricular function and a reduction in mortality.^31,32 Vessels diseased with atherosclerosis have impaired nitric oxide release in both the epicardial artery or graft and the resistance arterioles.³³ Nitric oxide positively affects latent collaterals or collateral blood flow by eliciting vasodilation and inhibiting platelet aggregation in the vascular bed distal to the target lesion.³⁴ Animal studies have demonstrated that NTP, a direct nitric oxide donor that does not require metabolism by the resistance arterioles, reduces platelet aggregation in the setting of injured endothelium and enhances ischemic preconditioning, both of which may contribute salutary effects.³⁵ Intracoronary and intragraft NTP administration has also previously been successfully used to treat the no-reflow phenomenon in both PCI and acute MI settings without significant hemodynamic side effects.¹¹ Thus, NTP may significantly alter the pathophysiology underlying the potential microvascular dysfunction resulting during PCI. The main finding of this study is that prophylactic intragraft administration of the microvascular vasodilator NTP was associated with a significantly reduced frequency and magnitude of post-PCI myonecrosis. Study limitations. This study has several important limitations. This was a retrospective, nonrandomized, single-center and small observational study. Therefore, the results and conclusions are subject to the limitations inherent in all such studies. A selection bias could be present despite matching the cases and controls for various variables that are important risk factors for SVG PCI distal embolization and post-procedural myonecrosis. The control group patients were less likely to have preprocedural target lesion thrombus and were more likely to have TIMI 3 flow than the patients treated with NTP. Hence, the NTP patients may have been at higher risk for post-procedural myonecrosis than the controls, potentially leading to an underestimation of the treatment effect. The results of our study cannot be generalized to SVG PCI with an EPD since we included only patients who had SVG PCI without an EPD.

Conclusion

Prophylactic administration of NTP during SVG PCI was associated with a significantly lower frequency and magnitude of post-procedural cardiac enzyme elevation. A larger, prospective, randomized, controlled trial to further evaluate this promising result of prophylactic administration of NTP or other potent microvascular vasodilators is warranted in SVG PCI. Given the proven efficacy of EPDs in appropriate patients, ideally, this prospective study to evaluate the utility of microvascular vasodilator prophylaxis would enroll both EP-eligible and ineligible patients.

References

1. Mehta SK, Frutkin AD, Milford-Beland S, et al. Utilization of distal embolic protection in saphenous vein graft interventions (an analysis of 19,546 patients in the American College of Cardiology-National Cardiovascular Data Registry). Am J Cardiol 2007;100:1114–1118. 2. Topol EJ, Yadav JS. Recognition of the importance of embolization in atherosclerotic vascular disease. Circulation 2000;101:570–580. 3. Hong MK, Mehran R, Dangas G, et al. Creatine kinase-MB enzyme elevation following successful saphenous vein graft intervention is associated with late mortality. Circulation 1999;100:2400–2405. 4. Bhargava B, Kornowski R, Mehran R, et al. Procedural results and intermediate clinical outcomes after multiple saphenous vein graft stenting. J Am Coll Cardiol 2000;35:389–397. 5. Hong MK, Mehran R, Dangas G, et al. Are we making progress with percutaneous saphenous vein graft treatment? A comparison of 1990 to 1994 and 1995 to 1998 results. J Am Coll Cardiol 2001;38:150–154. 6. Rogers C, Huynh R, Seifert PA, et al. Embolic protection with filtering or occlusion balloons during saphenous vein graft stenting retrieves identical volumes and sizes of particulate debris. Circulation 2004;109:1735–1740. 7. Stone GW, Rogers C, Hermiller J, et al. Randomized comparison of distal protection with a filter-based catheter and a balloon occlusion and aspiration system during percutaneous intervention of diseased saphenous vein aorto-coronary bypass grafts. Circulation 2003;108:548–553. 8. Baim DS, Wahr D, George B, et al. Randomized trial of a distal embolic protection device during percutaneous intervention of saphenous vein aorto-coronary bypass grafts. Circulation 2002;105:1285–1290. 9. Holmes DR Jr, Topol EJ, Califf RM, et al. A multicenter, randomized trial of coronary angioplasty versus directional atherectomy for patients with saphenous vein bypass graft lesions. CAVEAT-II Investigators. Circulation 1995;91:1966–1974. 10. Abbo KM, Dooris M, Glazier S, et al. Features and outcome of no-reflow after percutaneous coronary intervention. Am J Cardiol 1995;75:778–782. 11. Hillegass WB, Dean NA, Liao L, et al. Treatment of no-reflow and impaired flow with the nitric oxide donor nitroprusside following percutaneous coronary interventions: Initial human clinical experience. J Am Coll Cardiol 2001;37:1335–1343. 12. Smith SC Jr, Feldman TE, Hirshfeld JW Jr, et al. ACC/AHA/SCAI 2005 guideline update for percutaneous coronary intervention: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/SCAI Writing Committee to Update the 2001 Guidelines for Percutaneous Coronary Intervention). J Am Coll Cardiol 2006;47:e1–e121. 13. Browner WS, Black D, Newman TB, Hulley SB. Estimating sample size and power. In: Hulley SB, Cummings SR (eds). Designing Clinical Research. Baltimore: Williams and Wilkins. 1988, pp. 139–150. 14. Schlesselman JJ. Case-Control Studies. New York: Oxford University Press, 1982. 15. de Feyter PJ. Percutaneous treatment of saphenous vein bypass graft obstructions: A continuing obstinate problem. Circulation 2003;107:2284–2286. 16. Hillegass WB, Lyerly ML, Patel NJ, et al. Frequency and predictors of post-procedural myonecrosis in saphenous vein graft intervention in the stent era: A meta-regression. Circulation 2006;114(Suppl II­):733. 17. Webb LA, Dixon SR, Safian RD, O’Neill WW. Usefulness of embolic protection devices during saphenous vein graft intervention in a nonselected population. J Interv Cardiol 2005;18:73–75. 18. Webb JG, Carere RG, Virmani R, et al. Retrieval and analysis of particulate debris after saphenous vein graft intervention. J Am Coll Cardiol 1999;34:468–475. 19. Salloum J, Tharpe C, Vaughan D, Zhao DX. Release and elimination of soluble vasoactive factors during percutaneous coronary intervention of saphenous vein grafts: Analysis using the PercuSurge GuardWire distal protection device. J Invasive Cardiol 2005;17:575–579. 20. Leineweber K, Bose D, Vogelsang M, et al. Intense vasoconstriction in response to aspirate from stented saphenous vein aortocoronary bypass grafts. J Am Coll Cardiol 2006;47:981–986. 21. Stankovic G, Colombo A, Presbitero P, et al. Randomized evaluation of polytetrafluoroethylene-covered stent in saphenous vein grafts: The Randomized Evaluation of polytetrafluoroethylene COVERed stent in Saphenous vein grafts (RECOVERS) Trial. Circulation 2003;108:37–42. 22. Blackman DJ, Choudhury RP, Banning AP, Channon KM. Failure of the Symbiot PTFE-covered stent to reduce distal embolization during percutaneous coronary intervention in saphenous vein grafts. J Invasive Cardiol 2005;17:609–612. 23. Gorog DA, Foale RA, Malik I. Distal myocardial protection during percutaneous coronary intervention: When and where? J Am Coll Cardiol 2005;46:1434–1445. 24. Carrozza JP Jr, Mumma M, Breall JA, et al. Randomized evaluation of the TriActiv balloon-protection flush and extraction system for the treatment of saphenous vein graft disease. J Am Coll Cardiol 2005;46:1677–1683. 25. Roffi M, Mukherjee D, Chew DP, et al. Lack of benefit from intravenous platelet glycoprotein IIb/IIIa receptor inhibition as adjunctive treatment for percutaneous interventions of aortocoronary bypass grafts: A pooled analysis of five randomized clinical trials. Circulation 2002;106:3063–3067. 26. Karha J, Gurm HS, Rajagopal V, et al. Use of platelet glycoprotein IIb/IIIa inhibitors in saphenous vein graft percutaneous coronary intervention and clinical outcomes. Am J Cardiol 2006;98:906–910. 27. Rha SW, Kuchulakanti PK, Pakala R, et al. Bivalirudin versus heparin as an antithrombotic agent in patients who undergo percutaneous saphenous vein graft intervention with a distal protection device. Am J Cardiol 2005;96:67–70. 28. Fischell TA, Subraya RG, Ashraf K, et al. “Pharmacologic” distal protection using prophylactic, intragraft nicardipine to prevent no-reflow and non-Q-wave myocardial infarction during elective saphenous vein graft intervention. J Invasive Cardiol 2007;19:58–62. 29. Michaels AD, Appleby M, Otten MH, et al. Pretreatment with intragraft verapamil prior to percutaneous coronary intervention of saphenous vein graft lesions: Results of the randomized, controlled vasodilator prevention on no-reflow (VAPOR) trial. J Invasive Cardiol 2002;14:299–302. 30. Hori M, Inoue M, Kitakaze M, et al. Role of adenosine in hyperemic response of coronary blood flow in microembolization. Am J Physiol 1986;250:H509–518. 31. Taniyama Y, Ito H, Iwakura K, et al. Beneficial effect of intracoronary verapamil on microvascular and myocardial salvage in patients with acute myocardial infarction. J Am Coll Cardiol 1997;30:1193–1199. 32. Ito H, Taniyama Y, Iwakura K, et al. Intravenous nicorandil can preserve microvascular integrity and myocardial viability in patients with reperfused anterior wall myocardial infarction. J Am Coll Cardiol 1999;33:654–660. 33. Chilian WM, Dellsperger KC, Layne SM, et al. Effects of atherosclerosis on the coronary microcirculation. Am J Physiol 1990;258:H529–539. 34. Traverse JH, Kinn JW, Klassen C, et al. Nitric oxide inhibition impairs blood flow during exercise in hearts with a collateral-dependent myocardial region. J Am Coll Cardiol 1998;31:67–74. 35. Willerson JT, Igo SR, Yao SK, et al. Localized administration of sodium nitroprusside enhances its protection against platelet aggregation in stenosed and injured coronary arteries. Tex Heart Inst J 1996;23:1–8.