Role of nitrates in acute myocardial infarction

Roleof Nitrates in Acute MyocardialInfarction JOHN

T.

FLAHERTY,

Management of patients after myocardial infarction includes several therapeutic options. Lysis of the coronary thrombosis with intravenous or intracoronary administration of streptokinase or intravenous administration of one of the newer, currently experimental agents, such as tissue plasminogen activation or prourokinase, can directly restore oxygen and substrate delivery to potentially salvageable myocardium. Percutaneous transluminal coronary angioplasty can likewise restore vessel patency with potential salvage of ischemic myocardium, if perfused sufficiently early after symptom onset. Another strategy is to administer intravenous thrombolytic therapy and then perform early angioplasty on patients with acute myocardial infarctbn who reach the hospital within 4 holirs of symptom onset. These patients should have intravenous nitroglycerin begun

D

espite warnings in Friedberg’s Diseases of the Heart as recently as the 1966 edition1 that nitroglycerin administration was contraindicated in patients with acute myocardial infarction [AMI), cardiologists at The Johns Hopkins Hospital began studying thti shortterm hemodynamic and antiischemic effects of intravenous nitroglycerin (NTG) in patients with AM1 more than 14 years ago.2-5In order to be sure that by lowering coronary perfusion pressure regional ischemia was not being made worse, precordial ST segments were monitored in those patients with anterior wall infarction. In the first 12 patients studied the mean blood pressure was lowered by only 7 mm Hg, and in all 6 patients with anterior infarction a reduction in the sum of the ST-segment voltages was documented.2 Encouraged by our initial experience, a wider range of infusion rates was used in the next 30 patients.3 Beneficial antiischemic effects were found in all hemodynamic subgroups, irrespective of the presence or absence of left ventricular (LV) failure. At lower infusion rates nitroglycerin had acted principalFrom the Cardiology Division, Johns Hopkins University School of Medicine, Baltimore, Maryland. Address for reprints: John T. Flaherty, MD, Cardiology Division, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205. 35H

MD

before or simultaneously with beginning thrombolytic therapy. The infusion is titrated to ldwer systolic arterial pressure by 10% to 15 % , and then maintained at a constant rate for up to 48 hours. Patients seen more than 4 hours after symptom onset, with evidence of viable myocardium (e.g., persistent R waves in those electrocardiographic leads demonstrating ST-segment elevation) may also receive intravenous nitroglycerin and thrombolytic or percutaneous transluminal coronary angioplasty therapy. The combined results of the,several clinical trials df intravenous nitroglycerin in acute myocardial infarction would support its use in patients seen 4 to 12 hours after onset of symptoms or in patients seen earlier, in whom thrombolytic or percutaneous translumihal coronary angioplasty th&apy cannot be utilized. (Am J Cardiol 1987;80:35H-38H)

ly as a venodilator, lowering LV filling pressure by 10 mm Hg, but lowering mean arterial pressure by only 7 mm Hg. At higher infuSion rates, more balanced venous and arterial dilating effects w&e evident, with a similar lowering of LV filling pressure but with a greater [mean 21 mm Hg) lowering of meati arterial pressure. In the hemodynamic subgroup of patients without LV failure a decrease in stroke volume and a decrease in the LV filling pressure were demonstrated, evidence of venodilation and a predominant preload lowering or “diuretic-like effect.” In the subgroup of patients with mild LV failure, as evident from an elevated LV filling pressure and a more normal stroke vblume, a similar lowering of LV filling pressure was observed but with maintenance of stroke volume, suggesting that NTG was inducing a degree of afterload lowering as well.‘In the subgroup of patients manifesting the most severe degree of LV failure, the most beneficial hemodynamic effects were observed. In these patients, LV filling pressure was again lowered but, in addition, stroke volume increased, suggesting both preload and afterload lowering and a balanced venous and arterial dilating effect. Thus, short-term administration of intravenous NTG to patients with AM1 could be shown to improve regional ischemia, as evident from precordial ST-segment mapping in all hemodynamic subgroups. The

36H

A SYMPOSIUM:

NITROGLYCERIN

THERAPY-A

CONTEMPORARY

subgroup of patients with the most severe degree of LV failure obtained the most beneficial hemodynamic response, demonstrating both reduction in pulmonary venous congestion and improvement in forward cardiac output. Slow upward titration of the infusion rate allowed these beneficial antiischemic and hemodynamic effects to be obtained without excessive lowering of blood pressure and without reflex tachycardia. In fact, the mean heart rates were lower than baseline in all hemodynamic subgroups, most notable in subgroups with evidence of LV failure.

A Randomized Trial to AssessEffectsof IntravenousNitroglycerinon Infarct Size Based on these favorable results with short-term treatment, a randomized placebo-controlled trial of longer term infusion of intravenous NTG in patients with AM1 was undertaken.6 Patients received 48 hours of either NTG or placebo therapy followed by administration of cutaneous NTG or placebo for an additional 72 hours, Included were patients with electrocardiographic changes suggestive of AM1 who could be randomized within 12 hours of the onset of chest pain (Table I). Excluded were patients with systolic blood pressure <95 mm Hg, heart rate <55 beats/min or age greater than 75 years. Also excluded were patients

TABLE I

Patient Population

Inclusion criteria Electrocardiographic changes suggestive of AM1 Randomized within 12 hours of onset of chest pain Exclusion criteria Systolic blood pressure <95 mm Hg Heart rate <55 beats/min Age greater than 75 years Significant arterial hypertension AMI = acute myocardial

infarcton.

5,?50% PLACEBO
PLACEBO L IO hrs

S50% u

NTG < IO hrs

NTG 2 10 hrs

FIGURE 1. Change in left ventricular ejection fraction (in absolute percent) as measured by gated blood pool scintigraphy between day 1 and days 7 to 14 in patients receiving intravenous nitroglycerin (NTG) or placebo early (
PERSPECTIVE

with significant arterial hypertension in whom intravenous vasodilator therapy could not ethically be withheld. Before beginning the NTG infusion a thallium-201 perfusion scan and a technetium-99m labeled gated blood pool scan were obtained. A computer-assisted scoring technique quantified the extent and severity of myocardial perfusion defects7 This thallium defect score provided an estimate of the total area at risk and included in some patients a previously infarcted myocardium. The gated blood pool scan defined pretreatment LV ejection fraction as well as pretreatment segmental wall motion. Infusion of intravenous NTG was begun at 5 to 10 pg/min. The infusion rate was increased in a stepwise manner until a 10% lowering of mean arterial pressure was obtained, monitored by an ultrasonic blood pressure cuff. This final infusion rate was then continued for 48 hours, unless side effects or hemodynamic instability required reduction or discontinuation of the infusion. The mean infusion rate required to obtain the 10% mean arterial pressure lowering was 90 /*g/min. Seven to 14 days after admission, the thallium-201 myocardial perfusion scan and the gated blood pool scan were repeated. Fifty-six patients were randomized to NTG and 48 patients were randomized to placebo therapy. No significant differences in admission clinical, laboratory or scintigraphic parameters were found between the NTG- and placebo-treated patient groups. Early and late treatment subgroups were then defined by the time interval between the onset of chest pain and the initiation of treatment (mean 10 hours]. Patients treated less than 10 hours from the onset of symptoms were considered to have been treated early, and those treated 110 hours after the onset of chest pain were considered to have been treated late. Although not anticipated to be a major endpoint variable owing to the relatively small patient population studied, the &month mortality rate in the early NTG treatment subgroup was only 14% compared with 21% to 28% for the other 3 treatment subgroups; this difference approached but did not quite reach statistical significance. When the clinical outcomes of in-hospital death, infarct extension or the development of new congestive heart failure were examined separately among treatment subgroups, differences strongly favored early NTG treatment, but again did not quite reach statistical significance. However, when the incidence of any 1 of these 3 unfavorable outcomes was combined, a highly significant reduction was noted in those patients receiving early nitroglycerin treatment. Patients with an abnormal ejection fraction on admission and who received early treatment with NTG demonstrated a mean improvement in LV ejection fraction of 11% (Fig. 1). Patients with initially abnormal ejection fractions in the other 3 treatment subgroups demonstrated either no change or a small decrease in ejection fraction. Patients with initially normal ejection fractions (>50%] demonstrated de-

November

TABLE II Infarction

Comparison

of Intravenous

18, 1987

Nitroglycerin

THE AMERICAN

with Nitroprusside

JQIJRNAC

MBF = myocardial

blood flow; 1 = decrease;

tMBF

371-l

Nitroprusside

tST elevations

JST elevations IST elevations

Volume 60

in Acute Myocardial

Nitroglycerin Chiarello” Patients with anterior transmural myocardial infarction Open chest anesthetized dog model with acute coronary ligation Mann’* Patients with angiographically visible intercoronary collateral channels

OF CARDICLCGY

and tMBF

$ST elevations

and JMBF

~MBF

1 = increase

creases in ejection fraction irrespective of their therapeutic subgroup. It would seem likely that admission ejection fractions were artificially augmented by the high circulating catecholamine levels. Seven to 14 days later, in the absence of elevated catecholamines, a lower overall ejection fraction might be expected. Beneficial effects of early treatment with intravenous NTG are evident despite this underlying tendency for ejection fraction to decline both times.

anterior transmural infarctions revealed significantly better preservation of precordial R wave voltage.

Comparisonof the Effectsof Nitroglycerinand Nitroprussideon RegionalIschemia

Several studies have compared the hemodynamic and antiischemic effects of NTG and nitroprusside (Table II]. Chiarello et all1 reported a study in which 10 patients with acute anterior transmural myocardial infarction first received an intravenous infusion of sodiOtherClinicalTrials AssessingEffectsof um nitroprusside at a rate sufficient to lower mean IntravenousNitroglycerinon Infarct Size arterial pressure by 25 mm Hg. During nitroprusside Several other clinical studies of intravenous NTG infusion, all 10 patients showed an increase in their in patients with AM1 have been carried out. Bussman precordial ST segment elevations, suggesting worsenet al8 compared hemodynamics and cardiac enzyme ing of regional myocardial ischemia. After discontinurelease in 31 patients who received intravenous NTG ing the nitroprusside infusion, 5 of these patients were for 48 hours with 29 patients who did not. Bussman et given sublingual NTG in a dose that lowered mean al8 studied only patients with transmural infarcts and arterial pressure by 14 mm Hg. All 5 patients demonLV filling pressures of 112 mm Hg. Treatment was strated a decrease in their precordial ST segment instituted a mean of 10 hours after the onset of symp- elevations, suggesting an improvement in regional toms. NTG was titrated to lower the pulmonary dia- ischemia. In order to define the mechanism responsible for stolic pressure, specifically avoiding excessive lowering of arterial pressure. A mean infusion rate of 47 pg/ these opposite effects of NTG and nitroprusside on min was used. Total and myocardial specific (MB] cre- regional ischemia in patients, these same investigators atine-kinase infarct size was found to be significantly carried out an analogous study in an open-chest aneslower for the NTG-treated compared with the control thetized canine model. After ligation of the left anterior descending coronary artery, nitroprusside was inpatients. A randomized prospective study in 85 patients fused at a rate sufficient to lower mean arterial treated with a %-hour infusion of NTG or placebo, a pressure by 20 mm Hg. After discontinuing the nitromean of 6 hours after onset of symptoms, was reported prusside infusion, NTG was infused at a rate that lowered mean arterial pressure by an equal amount. Usby Jaffe et a1.g NTG infusion was titrated to lower systolic blood pressure by lo%, unless heart rate in- ing the radioactive microsphere technique, Chiarello creased by 20 beats/min, or to a maximum dose of 200 et al demonstrated that myocardial blood flow distal to pg/min. The mean infusion rate required was 54 pug/ the ligated left anterior descending coronary artery min. Creatine-kinase infarct size was lower with NTG decreased and epicardial ST segment voltages intreatment, but only in the subgroup of patients with creased during nitroprusside infusion, In contrast, transmural inferior infarctions. during NTG infusion, regional myocardial blood flow In another randomized prospective study, Derrida increased and epicardial ST segments decreased. The et allo reported a significant reduction in in-hospital improvement in regional ischemia observed with NTG mortality in 74 patients given NTG treatment com- was evidence of an increase in intercoronary collaterpared with control patients. NTG infusion was titrated al blood flow. During nitroprusside infusion, in conto lower systolic arterial pressure 20 mm Hg, and then trast, collateral flow to the ischemic zone actually demaintained for 1 to 7 days. The mean NTG infusion creased as flow to the nonischemic zone increased, rate was 51 ,ug/min, and the mean time between the suggesting the development of a “coronary steal.” onset of symptoms and the initiation of therapy was 10 In the cardiac catheterization laboratory Mann et hours. Precordial mapping studies in 46 patients with alI2 compared intravenous NTG and nitroprusside in a

38H

A SYMPOSIUM:

NITROGLYCERIN

THERAPY-A

CONTEMPORARY

group of patients with fixed coronary artery disease. In the subgroup of patients with angiographically visible intercoronary collaterals, myocardial blood flow, measured by the xenon washout technique, distal to a severe coronary artery stenosis, likewise increased with NTG and decreased with nitroprusside. In summary, although NTG and nitroprusside exert similar effects on hemodynamics, they appear to exert substantially dissimilar effects on intercoronary collateral blood flow and thereby on the severity of regional myocardial ischemia. These differences would appear to make intravenous NTG preferable to nitroprusside for treatment of patients with acute ischemia or infarction, especially in the presence of severe fixed coronary disease.

Randomized Trials to AssessEffectsof Nitroprussideon Infarct Sizeand Mortality Two randomized prospective placebo-controlled trials of nitroprusside compared with placebo in patients with AM1 have been reported. Cohn et all3 reported the results of a large multicenter Veterans Administration Cooperative Trial. The nitroprusside infusion was titrated to reach 1 of 4 endpoints: a reduction of LV filling pressure to <60% of control, a lowering of systolic arterial pressure to 20% of the control pressure plus 76 mm Hg, the development of significant side effects or a maximum infusion rate of 200 pg/ min. Onset of chest pain less than 24 hours before admission was required for a patient to be a candidate for the study (812 patients with transmural infarcts entered the study). Patients with LV pressures X12 mm Hg were excluded, as were patients with hypertension requiring vasodilator therapy, or a systolic blood pressure
PERSPECTIVE

rate >120 beats/min. The mean interval between the onset of symptoms and the initiation of therapy was 5 hours. Durrer et alI4 found a significant reduction in mortality after 23 days of follow-up between nitroprusside- and placebo-treated patients. Eighteen patients died in the placebo-treated group compared with only 5 in the nitroprusside-treated group (p <0.05]. Noteworthy, however, was the finding that the cause of death in 9 of 18 placebo-treated patients was rupture of the left ventricular free wall, a papillary muscle or the interventricular septum. This relatively high incidence of myocardial rupture of one form or another in a population of patients that included both subendocardial and transmural infarctions would seem higher than expected. Since initially hypertensive patients appeared not to be excluded, nitroprusside’s beneficial effect on mortality could be a reflection of its ability to reduce blood pressure during the early hours of AMI, thereby preventing myocardial rupture. Thus, for different reasons, both of these studies raise questions regarding the advisability or the benefits of nitroprusside treatment in patients with AM1 and would appear to favor the use of intravenous NTG in this clinical setting.

References 1. Friedberg CK. Diseases of the Heart, third ed. Philadelphia: WB Saunders, 1966293. 2. Flaherty IT, Reid PR, Kelly DT, Taylor DR, Weisfeldt ML, Pitt B. Intravenous nitroglycerin in acute myocardial infarction. Circulation 1975;51:132139. 3. Flahery JT, Come PC, Baird MG, Rouleau J, Taylor DR, Weisfeldt ML, Greene HL, Becker LC, Pitt B. Effects of intravenous nitroglycerin on left ventricular function and ST segment changes in acture myocardial info&tion. Br Heart J 1976;38:612-621. 4. Come PC, Flaherty JT, Baird MG, Rouleau JR, Weisfeldt ML, Greene HL, Becker L, Pitt B. Reversal by phenylephrine of the beneficial effects of intravenous nitroglycerin in patients with acute myocardiaf infarction. N Engl J Med 1975;293:1003-1007. 5. Come PC, Flaherty JT, Becker LC, Weisfeldt ML, Greene HL, Weiss JL, Pitt B. Combined administration of nitroglycerin and propranolol to patients with acute myocordial infarction. Chest 1981;80:416-424. 6. Flaherty JT, Becker LC, Bulkley BH, Weiss JL, Gerstenhlith G, Kallman CH, Silverman LK, Wei J, Pitt B, Weisfeldt ML. A randomized prospective trial of intravenous nitroglycerin in patients with acute myocardial infarction. Circulation 1983;68:576-588. 7. Burow RD, Pond M, Schaeffer AW, Becker L. “Circumferential profiles”: a method for computer analysis of thallium 201 myocardium perfusion images. J Nucl Med 1979;20:771-777. 8. Bussman WD, PassekD, Seidel W, Kaltenbach M. Reduction of CK and CKMB indexes of infarct size by intravenous nitroglycerin. Circulation 1981; 63:615-622. 9. Jaffe AS, Geltman EM, Tiefenbrunn AF, Dieter Ambos H, Strauss HD, Sobel BE, Roberts R. Relation of the extent of inferior myocardial infarction with intravenous nitroelvcerin: a randomized .urouective studv. Br Heart T 1983;49:452-460.

- '

10. Derrida ]P, Sal R, Chiche P. Effects of prolonged nitroglycerin infusion in patients with acute myocardial infarction (abstr). Am J Cardiol 1978;41:407. 11. Chiarello M, Gold HK, Leinbach RC, Davis MA, Maroko PR. Comparison between the effects of nitroprusside and nitroglycerin on ischemid injury during acute myocardial infarction. Circulation 1976;54:766-773. 12. Mann T, Cohn PF, Holman BL, Green LH, Markis JE, Phillips DA. Effect of nitroprusside on regional myocardial blood flow in coronary disease: results in 25 patients and comparison with nitroglycerin. Circulation 1978; 57:732-738. 13. Cohn JN, Franciosa JA, Francis GS, Archibald D, Tristani F, Fletcher R, Montero A, Cintron G, Clarke J, Hager D, Saunders R, Cobb F, Smith R, Loeb H. Settle H. Effect of short-term infusion of sodium nitromusside on mortalitv r(lte in acute $yoca;diaJ infarction compljcated by Ieft ventricular failure. 2 Engl J Med 1982;306:1129-1135. 14. Durrer JD, Lie KL, Van Cape11FJT, Durrer D. Effect of sodium nitroprusside on mortality in acute myocardial infarction. N Engl J Med 1982;306:11211128.