Hemodynamic and metabolic effects of isosorbide

dinitrate in chronic congestive heart failure

Richard Gray, M.D. Kanu Chatterjee, M.B., M.R.C.P.(Lond, and Edin.) John K. Vyden, M.B. William Ganz, M.D., C.Sc. James S. Forrester, M.D. H. J. C. Swan, M.D., Ph.D. Los Angeles, Calif.

Congestive heart failure (CHF) refractory to vigorous conventional treatment with digitalis and diuretics is not uncommon in clinical prac- tice. Recently potent peripheral vasodilators such as nitroprusside and phentolamine have been found useful in improving left ventricular per- formance in patients with left ventricular failure due to acute myocardial infarction (MI)‘e5 and chronic CHF with or without mitral regurgita- tion: The long-term use of these drugs in clinical practice is seriously limited by the need for both continuous infusion and invasive hemodynamic monitoring.

Such limitations are avoided by the use of sublingual nitroglycerin, which has been shown to have beneficial hemodynamic effects in patients with acute MI complicated by left ventricular failure.’ The use of this agent, however, is limited by its duration of action.

The purpose of this study was, therefore, to determine whether a sublingual long-acting peripheral vasodilator could be employed to obtain the beneficial hemodynamic effects of nitroprusside, phentolamine, and nitroglycerin without the limitations for clinical use encoun- tered with these drugs. Isosorbide dinitrate, a

From the Department of Cardiology, Cedars-Sinai Medical Center, and the Department of Medicine, University of California, Los Angeles,

C&f.

This study wss supported in part by Contract No. PH-43-68-1333 under Myocardial Infarction Research Program, National Heart and Lung Institute, National Institutes of Health.

Received for publication Aug. 19, 1974.

Reprint requests to: Richard Gray, M.D., Department of Cardiology, Cedars-Sinai Medical Center, 4833 Fountain Ave., Los Angeles, Calif. 90029.

“long-acting” nitrate which can be administered sublingually or orally, was selected. As a number of patients in this study had ischemic heart disease as the underlying etiology of CHF, the effects of the drug on myocardial metabolism were also studied in five patients by means of coronary sinus blood flow, oxygen, and lactate measurements.

Methods

Patient population. All patients were studied in the Myocardial Infarction Research Unit at Cedars-Sinai Medical Center. Following informed consent, 12 patients, 37 to 77 years of age, with chronic CHF, diagnosed on the basis of dyspnea, cardiomegaly, basilar riles, elevated jugular venous pressure, and signs of pulmonary hyper- tension were studied (Table I). The etiology of heart failure included previous myocardial infarc- tion (8), coronary artery disease (CAD) without infarction (l), primary cardiomyopathy (l), hypertensive cardiovascular disease (l), and rheu- matic heart disease with prosthetic aortic and mitral valves (1). Six patients had associated mitral regurgitation. All patients had been previously treated with standard digitalis, diur- etic, and sodium restriction regimens. Whenever possible previous medications were withheld for 24 hours prior to study.

Hemodynamic and metabolic measurements.

All recordings were made with the patient in the supine position.

Arterial pressure (AP) was monitored contin- uously through a 20 gauge cannula inserted into the radial artery. Right atria1 (RA), pulmonary

346 September, 1975, Vol. 90, No. 3, pp. 346-352

Isosorbide dinitrate in chronr:, h.eart failure

Table I. Clinical data

Patient Age sex Etiology CHF Previous Rx Comments

D. C. 65 G. K. 65 L. K. 77 M. W. 72

H. S. 73 B. K. 37 J. W. 60

M. P. 62 M. T. 77 s. c. 51 M. L. 59

L. c. 67

M CAD, 3 MI + BP* Digoxin furosemide Mitral, tricuspid regurgitation

F CAD, 1 MI Furosemide

F CAD, 1 MI Digoxin furosemide Mitral regurgitation

F Recent MI Digoxin furosemide Mitral regurgitation, severe pulmonary congestion

M CAD, 4 MI Digoxin furosemide Mitral regurgitation

F Cardiomyopathy Digoxin furosemide Normal coronary angiography

F CAD, no MI Digoxin furosemide, NTG, Mitral regurgitation Sorbitrate

I T21M1

Digoxin furosemide Mitral, tricuspid regurgitation Digoxin Severe pulmonary congestion

CAD, 1 MI Digoxin furosemide LBBB F Rheumatic heart disease with Digoxin quinidine, Coumadin Tricuspid regurgitation

aortic and mitral prostheses M CAD, 1 MI Digoxin furosemide

*BP = blood pressure.

artery (PA), and pulmonary capillary wedge (PCW) pressures and cardiac output (CO) were recorded by use of a balloon-tipped, flow-directed, triple-lumen catheter. Cardiac output was mea- sured by the thermodilution technique.*-‘” Ve- nous capacitance (VC) was measured in eight patients with the equilibration method and a mercury-in-rubber strain gauge plethysmo- graph.” The collecting pressure used was 30 mm. I-k.

Myocardial oxygen consumption (ml./min.) = (arterial - CS OI content [ml./100 ml.]) x CSF (ml./min.) x 10’.

Myocardial lactate extraction (W) = Arterial - coronarv sinus

Arterial lactate (mg./lOO ml.) X 100.

Coronary sinus flow was measured in five cases with the constant infusion thermodilution tech- nique.12 In these five cases, simultaneous samples of systemic arterial, pulmonary artery, and coro- nary sinus blood were analyzed for pH, pCO,, and pOZ by a pH gas analyzer Model 113 and hemo- globin saturation was determined with a co- oximeter Model 182. Samples of blood were also taken for determination of lactate concentration by an automated modification of the enzymatic method of Hohorst.‘3

Protocol. Sublingual isosorbide dinitrate was administered as 5 mg. (three patients), 10 mg. (six patients), or 15 mg. (three patients) based upon the magnitude of heart failure as assessed by mean pulmonary capillary wedge pressure (PC W) and level of arterial pressure (AP). Complete dissolution and absorption of the tablet was assured by having the patient moisten his mouth before drug administration (Fig. 1).

Derived hemodynamic and metabolic parame- ters were calculated as follows:

Cardiac index (CI) = CO/body surface area (BSA) (ml./M.‘!

Control hemodynamic and metabolic data were recorded. After 15 minutes hemodynamic mea- surements were repeated to insure stability of controls. Following drug administration venous capacitance was measured at 5 minute intervals and hemodynamic parameters were recorded at 15 minute intervals. Metabolic studies were performed during peak drug effect. This was defined as that time at which greatest change was seen in cardiac output.

Stroke volume index (SVI) = SV/body surface area (BSA) (ml/M.‘)

Results

Systemic vascular resistance (SVR) = (n - RA) (mm. Hg)/CO (L./min.) x 80 (dynes/sec./cm.?).

AP/At = (DAP - -)/PEP (mm. Hg/sec.),14 where PEP = pre-ejection period (msec.) and is derived from total electromechanical systole (QS,) - left ventricular ejection time (LVET) (msec.) simultaneously recorded from external carotid pulse tracing and electrocardiogram (ECG); DAP = arterial end-diastolic pressure (mm. Hg).

The hemodynamic and peripheral vascular ef- fects of isosorbide dinitra.te became evident within 15 minutes in all patients, reached maximum levels at 15 to 45 minutes, and persist- ed in reduced magnitude for 75 to 90 minutes.

The maximum changes in hemodynamic par- ameters for each individual were defined as occur-

American Heart Journal 347

Gray et al.

60

60 IA

Venous Capacitance

Is i0 45 6b 7k 9’0 MINUTES AFTER ISOSORBIDE

Fig. 1. The effect of isosorbide dinitrate upon five hemody- namic parameters over the course of 90 minutes. These data represent the mean per cent change from control values in 12 patients. A substantial increase in venous capacitance is seen at 5 minutes with peak effect at 15 minutes. All other hemody- namic parameters have peak effect at 15 to 30 minutes. After 75 minutes these effects are markedly reduced. Abbreviations: a mean arterial blood pressure; C.I., cardiac index; PCW, mean pulmonary capillary wedge pressure; TSVR, total systemic vascular resistance.

3.80 40 N .

5 3.00

t 8 d

. 30

5 .E < 2.20 -I

A=--=l L

4 4 SI” 2s

20 *! . f\: %

I*

1.40 ;---• 10 :\;

110 :b&.

CII 90

I a= mE

E 70

o) 7.0 .

a

2 -2 5.0 1, %

.

vo ‘/f

#

* . g 3.0

=o * 8

‘2 1.0 19: l ,

c I

Fig. 2. The hemodynamic response to isosorbide dinitrate in 12 individual patients. Cardiac index increased in 10 patients, remained unchanged in one, and declined in one. Pulmonary capillary wedge pressure decreased in all patients and arterial pressure decreased in 10, remaining unchanged in two. Venous capacitance increased in all eight patients studied. Abbrevia- tions: BP, mean arterial blood pressure (mm. Hg); C.I., cardiac index (L./min./M.*); PCW, mean pulmonary capillary wedge pressure (mm. Hg); Ven. Cup., venous capacitance (c.c./lOO C.C. of tissue).

ring at the time of maximum cardiac output (Table II and Fig. 2). These included a modest reduction in mean arterial pressure (-7 mm. Hg) and substantial changes in right atria1 (-4 mm. Hg) and pulmonary capillary wedge pressure (-7 mm. Hg), cardiac index (+0.38 L./min./M.‘),

stroke volume index (+5 c.c./min./M.2). Notable reductions in systemic and pulmonary vascular resistances, -510 and -110 dynes/sec./cm.-“, respectively, and an increase in venous capaci- tance (+ 1.53 c.c./lOO C.C. of tissue) also occurred. Heart rate remained unchanged. Changes in @/At, an index related to left ventricular dp/dt, were studied in eight patients. In seven patients, all with obstructive coronary artery disease, AP/ At increased during isosorbide dinitrate therapy. The remaining patient, in whom AP/At de- creased, had primary cardiomyopathy without coronary artery disease.

In five patients myocardial metabolic studies were performed before and after drug administra- tion (Table III). In all patients coronary sinus flow (-34 ml./min.) and myocardial oxygen consumption (-3.6 mUmin.) decreased during isosorbide dinitrate therapy. No significant change occurred in arterial-coronary sinus O? difference, or transmyocardial lactate extrac- tion.

Discussion

The principal hemodynamic objectives of man- agement of patients with chronic heart failure are to improve forward cardiac output and decrease pulmonary venous pressure. Intravenous vasodi- lators such as phentolamine or nitroprusside fulfill these objectives and have been successfully applied in the management of patients with congestive heart failure. The obvious disadvan- tages of such therapy are that they cannot be used for long-term management and require closely supervised hemodynamic monitoring.

This study demonstrates that sublingual iso- sorbide dinitrate, a vasodilator agent, may be used to improve left ventricular performance in patients with refractory and severe CHF and may be useful in long-term management. All patients exhibited significant reduction in pulmonary and systemic venous pressures and concomitant re- duction in the symptoms and signs of right and left heart failure.

The data in this study suggest that these changes were a result of changes in both systemic vascular resistance and venous capacitance. Decreased resistance to ejection by the failing heart would be expected to increase ejection fraction and decrease left ventricular volume and filling pressure. An increase in venous capacitance would allow venous pooling, thus reducing both

348 September, 1975, Vol. 90, No. 3

Isosorbide &nitrate in &ronlta heart faiilnre

Table II. Hemodynamic changes during isosorbide dinitrate therapy

T,,* ~..~~~I-. ~~-- .-.

CI SVI SVR

PVR . ../ _. - ! M’liT ! I’C

(beats/ A-P Zi PA PCW (L./min.l (mL.l (dynes/ (dyzes! j(mnc. Hg/ , (c.c.ilOU

Patient min.) (mm. Hg} (mm. Hg) (mm. Hg) (mm. Hg) M-1) min./M.2) sec./cm. :lsec.lcm. .i/ ;si’C., C. l’.) I

D. c. Ci 1

Cr. K.

c

I L. K.

c I

M. W. c I

H. S. c I

R. K.

(:

I

J. W. (‘ I

M. Y. C 1

M. T. (‘ I I

s. c. (‘ , I

M. I,. (: 1

L. c. c 1

Mean (: 1

Change Per cent

P

84 87 12 52 37 1.40 16 2,915 444 2.45

83 81 4 36 28 2.0.3 29 2,029 277 :t.“x

109 84 7 36 30 l.@! 16 2,714 318

103 74 4 28 20 2.18 "2 2,088 1.57

85 89 13 37 27 1.58 19 3,111 350

83 79 7 31 20 1.99 25 2,468 221)

101 75 4 37 27 1.33 13 2,953 4;;

104 69 1 28 23 1.44 14 2,093 2.56

72 101 7 24 19 2.58 36 2,589 185

76 96 5 15 12 2.40 30 2,541 100

2.90 3.20

89 81 13 31 17 2.24 26 2,088 189

78 80 8 22 14 2.68 36 1,708 96

:3.15

4.18

74 82 6 48 30 2.19 31 1,901

74 72 6 34 22 2.15 30 1,902

0.92

1.95

64 85 16 40 25 1.60 26 2,523

62 68 13 38 18 2.06 32 1,481

1.66

4.22

110 107 16 44 29 2.25 21 3,194

121 97 13 45 28 3.49 28 1,884

112 80 10 44 39 2.73 24 1,386

102 80 5 30 30 3.04 30 1,333

62 73 14 37 32 2.41 39 1,314 63 70 8 32 26 2.66 42 1,258

1.47

2.33

4.37

5.74

2.80

7.00

90 76 5 36 29 2.116 23 1,986

82 68 3 23 14 2.34 27 1,600

349

:ju1 d

42:1

324

407

230

99

71

13 1

161

“25

256

88 i 5 85 t 3 10 c 1 39 k 4 28 f 2 1.99 -+ 0.13 24 -+ 2 2,428 + 168 302 i- 37 253 z 31 2.46+0.16

85 -+ 5 78 i 2 6-cl 30 f 4 21 f 2 2.37 t 0.15 29 t 2 1,918 +- 99 192 -f- 26 29x -_+ 39 3.99zto.24

-3

-3

-7

-8

< 0.02

-4

-40

< 0.02

-9 -7 + 0.38

-23 -25 i-20

NS < 0.02 -=z 0.02 < 0.02

+5

+21

< 0.02

-510 -110

-“I --:ir;

< 0.02 -c 0.0'2 -- --.-_ -. ~.. ..-- -~~- ..-.__ -

*HR, heart rate; s, mean arterial pressure; E, mean right atria1 pressure; PA. mean pulmonary arterial pressure; PCW, mean pulmonary capillary wedge pressure; CI, cardiac index; SVI, stroke volume index; SVR, systemic vascular resistance; PVR, pulmonary vascular resistance; VC, VPDOUS capacitance. tC, control; I, lsosorbide dinitrate.

the systemic and pulmonary venous pressures. observed following isosorbide dinitrate therapy in This venous pooling effect, even in the absence of 10 of the 12 patients. An increased stroke volume CHF, has been described previously.‘” with decreased left ventricular filling pressure

Increased forward cardiac output, the other suggests an enhanced cardiac performance. The important objective of therapy of CHF, was also mechanism of increased cardiac output is likely to

American Heart Journal 349

Gray et al.

Table III. Myocardial metabolism, five patients

CSF* A-C’S 0, MVO, Lactate extraction (c.c./min.) (vol. %) (c.c./min.) @J

Control 165 + 13 11.4 + 1.0 18.1 f 1.6 36 +- 5.5 Isoaorbide dinitrate 131 f 15 11.2 I 1.0 14.5 k 1.6 40 -e 6.1 Per cent -34 -0.2 -3.6 +4 P < 0.02 NS < 0.02 NS

‘CSF, coronary sinus flow; MVO,, myocardial oxygen consumption.

Table IV. Maintained improvement after 2 months of isosorbide treatment

HR* AP zi ir;i PCW CZ svz SVR PVR

5173 c 79 103 8 44 23 1.8 23 2,610 307 5173 I 80 99 6 37 18 2.0 25 2,286 193

7173 I 68 80 3 21 9 2.3 34 1,717 124

‘For abbreviations see Table II.

be a decrease in resistance to left ventricular ejection, as previously described.‘” In two patients, the cardiac index did not increase. In one patient (H. S.) who had sustained four previous myocardial infarctions with long- standing heart failure, PCW pressure decreased markedly early in the study suggesting that left ventricular filling pressure was reduced below that which was optimal.” The other patient (J. W.), whose cardiac index did not increase (2.19 to 2.15 ml./min./M.*), was the only patient treated previously with long-acting nitrates (Sor- bitrate for chronic angina). Isosorbide dinitrate caused limb flow to drop sharply and limb vascular resistance to increase (not shown in figures).

AP/At, an index closely related to left ventric- ular dp/dt, increased during isosorbide dinitrate therapy, although both preload and afterload decreased and heart rate remained unchanged. While not conclusive, these unexpected findings are consistent with improved over-all contractile state which might be contributory to improved cardiac performance. The mechanism of any such improved over-all contractility, in the face of a known lack of direct inotropic effect of isosorbide dinitrate, remains conjectural. Reflex increase in inotropy is unlikely in the absence of an increase in heart rate. Relief of myocardial ischemia and recruitment of previously hypofunctioning myo- cardial segments may result in an improved con- tractile state, as suggested by recent angiographic studies before and after nitroglycerin administra- tion.18 Such improvement, however, would not be

350

expected in patients without coronary artery disease. In this study, the only patient who did not show an increase in AP/At was the one who had primary cardiomyopathy without obstruc- tive coronary artery disease.

The metabolic cost of enhanced mechanical performance, although investigated in only five patients, was consistent in all cases. Since coronary blood flow decreased in these patients, reduced MVO, could be related to decreased coronary perfusion pressure. In such an instance increased transmyocardial A-V oxygen difference or a decrease in transmyocardial lactate extrac- tion would be expected, but neither was observed in this study. The decrease in coronary blood flow and myocardial 0, consumption, therefore, are most likely related to decreased over-all 0, demand, which in turn reflects the observed de- crease in preload and afterload despite some suggestion of increase in contractile state.

Clinical application of isosorbide dinitrate therapy was begun in all of our patients following their hemodynamic study, and they were dis- charged from the hospital on chronic isosorbide therapy. Nine patients have been treated with isosorbide dinitrate for 4 to 8 months and all of them are judged clinically to be improved. Con- tinued hemodynamic benefit has been docu- mented in one 55-year-old man studied after 2 months of such therapy (Table IV).

Although not investigated directly in this study, another important application of this mode of therapy may be in the treatment of left ventricular failure during acute myocardial

September, 1975, Vol. 90, No. 3

lsosorbide dinitrate in chronic heurt failure

infarction. This potential use is highlighted by well-founded reluctance of many physicians to use digitalis in this setting and the recent emer- gence of potent vasodilators as effective therapy. Our results suggest that study of the use of isosorbide in this clinical situation is warranted.

Although this study demonstrates that isosor- bide dinitrate may be effective in the manage- ment of refractory congestive heart failure, some disadvantages should be remembered. The dura- tion of action does not usually exceed 90 minutes so that frequent drug administration may be a source of patient inconvenience. Sudden unex- pected hypotension and marked decrease in left ventricular filling pressure, although infrequent, may produce an adverse response. For these rea- sons, the arterial pressure, left ventricular filling pressure, and cardiac output should be monitored during the initiation of therapy. If these facilities are not available, the initial dose should not exceed 2.5 mg. The chance of an adverse response is further minimized if the patient has clear-cut evidence of increased left ventricular filling pres- sure. If dizziness, undue tachycardia, or hypoten- sion occur, therapy should be discontinued.

Summary

To assess the potential beneficial effects of a nonparenteral vasodilator, sublingual isosorbide dinitrate (5 to 15 mg.) was administered in 12 patients with chronic congestive heart failure refractory to conventional therapy. Hemody- namic measurements were performed before and at 15 minute intervals after drug administration for 90 minutes. Venous capacitance was measured at 5 minute intervals. Myocardial metabolism was also studied in five patients before and after drug administration.

Hemodynamic effects were characterized by a modest decrease in mean arterial pressure (85 + 3 to 78 -+ 2 [S.E.M.] mm. Hg) and substantial decrease in right atria1 (10 + 1 to 6 f l), pulmonary arterial (39 ? 4 to 30 + 4) and pulmonary capillary wedge pressures (28 + 2 to 21 + 2). These changes were accompanied by an increase in venous capacitance (2.46 +- 0.16 to 3.99 ? 0.24 c.c./lOO C.C. of tissue). Along with a decrease in left ventricular filling pressure, cardiac index increased (1.99 +- 0.13 to 2.37 + 0.15 L./min./M.‘). No significant effect on heart rate was seen. AP/At, an index related to left ventricular dp/dt, increased in all but one

American Heart Journal

patient (253 +- 31 to 298 t 39 mm. Hg/sec.) (p < 0.02 for all changes) in the face of decreased preload and afterload and unchanged heart rate, suggesting unproved contractile state. A decrease in coronary blood flow (165 i 13 t.o 131 + 15 c.c./min.) and myocardial oxygen consumption (18.1 + 1.6 to 14.5 k 1.6 c.c./min.) was noted (p < 0.02). No change in arterial-coronary sinus oxygen difference or lactate extraction was observed.

These data demonstrate that the objectives of therapy in congestive heart failure, namely im- proved forward output and decreased ventricular filling pressures, can be achieved without meta- bolic deterioration by using sublingual isosor- bide. The mechanisms responsible are related to diminished impedance to ventricular ejection and peripheral pooling of blood. Since the duration of action does not usually exceed 90 minutes, fre- quent drug administration may be a source of patient inconvenience.

The authors would like to acknowledge with gratitude the technical assistance of Docela Edwards, Luciann Robinson, Roman Kulczycky, David Mickle, Koichi Nagasawa, Marsha Groseth-Dittrich. Alma Aldredge, Lance Laforteza, dohn Albus, and Marge Raymond, and the edit,orial assistance of Sharman .Jamison.

REFERENCES

1.

2.

3.

4.

5.

6.

7.

a.

9.

Franciosa, J. A., Guiha, N. H., Limas. C. J., et al.: Improved left ventricular function during nitroprusside infusion in acute infarction, Lancet 1:650, 1972. Majid, P. A.. Sharma, B., and Taylor, S. H.: Phentol- amine for vasodilator treatment of severe heart failure. Lancet 2:719, 1971. Gould, L., Zahir, M., and Ettinger, S.: Phentolamine and (cardiovascular performance, Br. Heart J. 3 1:154, 1969. Wahnsky, P., Chatterjee, K., Forrester. J., Parmley, W. W., and Swan, H. J. C.: Enhanced left ventricular performance with phentolamine in acute myocardial infarction, Am. J. Cardiol. 33:37, 1974. Chatterjee, K., Parmley, W. W., Ganz, R., Forrester, J., Walinsky, P., Crexells, C., and Swan, H. J. C.: Hemody- namic and metabolic responses to vasodilator therapy in acute myocardial infarction, Circulation 48:1183, 1973. Chatter& K.. Parmlev. W. W.. Swan. H. J. C.. Berman. G.. Forrester, J., and Marcus, H. S.: Beneficial’effects of vasodilator agents in severe mitral regurgitation due t,o dysfunction of subvalvar apparatus, Circulation 48:684, 1973. Gold, H. K., Leinbach, R. C., and Sanders, C. A.: IJse of nitroglycerin in congestive failure following acute myo- cardial infarction, Circulation 46:839, 19X. Ganz. W.. Donoso. R.. Marcus. H. S.. Forrester. J. S.. and , , I ,

Swan, H. J. C.: A new technique for measurement of car- diac output by thermodilution in man, Am. ,J. Cardiol. 27392, 1971. Forrester, J., Ganz, W., Diamond, G., McHugh, T., Chonette, D., and Swan, H. J. C.: Thermodilution cardiac output determination with a single flow-directed catheter, AM. HEART J. 83:306, 1972.

351

Gray et al.

10.

11.

12.

13.

14.

Ganz, W., and Swan, H. J. C.: Measurement of blood flow by thermodilution, Am. J. Cardiol. 29:241, 1972. Mason, D. T., and Braunwald, E.: The effects of nitro- glycerin and amyl nitrate on arteriolar and venous tone in the human forearm circulation, Circulation 32:755, 1965. Ganz, W., Tamura, K., Marcus, H. S., Donoso, R., Yoshida, S., and Swan, H. J. C.: Measurement of coro- nary sinus blood flow by continuous thermodilution in man, Circulation 44:181, 1971. Marbach, E. P., and Weil, M. H.: Rapid enzymatic measurement of blood lactate and pyruvate use and significance of metaphosphoric acid as a common precip- itant, Clin. Chem. 13:X4, 1967. Diamond, G., Forrester, J. S., Chatterjee, K., Wegner, S., and Swan, H. J. C.: Mean electromechanical AP/At, Am. J. Cardiol. 30:338, 1972.

15.

16.

17.

18.

Zelis, R., and Mason, D. T.: Demonstration of nitrate tolerance attenuation of the venodilator response to nitroglycerin by the chronic administration of isosorbide dinitrate, Circulation 40:11-221, 1969. Sweatman, T., Strals, G., Arthur, S., et al.: The long acting hemodynamic effects of isosorbide dinitrate, Am. J. Cardiol. 29:480, 1972. Crexells, C., Chatterjee, K., Forrester, J. S., Dikshit, K., and Swan, H. J. C.: Optimal level of filling pressure in the left side of the heart in acute myocardial infarction, N. Engl. J. Med. 289:1263, 1973. Pine, R., Meister, S. G., Banka, V. S., et al.: Detection of reversible ventricular contraction abnormalities with ni- troglycerin: Correlation with post coronary bypass ventriculography, Circulation 48:IV-104, 1973.

352 September, 1975, Vol. 90, No. 3