Circadian fluctuations of tissue plasminogen activator antigen and plasminogen activator inhibitor-l antigens in vasospastic angina To elucidate the circadian variation of fibrinolytic components in vasospastic angina, plasma levels of tissue plasminogen activator antigen (t-PA), free plasminogen activator inhibitor antigen (free PAI-I), t-PA/PAI-1 complex, and total PAI- were measured in venous plasma samples. Samples were taken every 6 hours (6:00 AM, noon, 6:00 PM, and midnight) for 24 hours in 14 patients with vasospastic angina, in 9 patients with exertional angina, and in 19 normal subjects. Twenty-four-hour Holter monitoring (Holter monitor, Del Mar Avionics, Irvine, Calif.) was also carried out in all subjects, All of the fibrinolytic components showed circadian variation, with a peak level at 6:00 AM in every study group except for the t-PAIPAI-1 complex in the group of patients with exertional angina. The values for all of the fibrinolytic components at each sampling time were higher in patients with coronary artery disease than in normal subjects. In patiicular, the mean value of free PAI-I at 6:00 AM in patients with vasospastic angina was significantly higher than that in normal subjects and that in patients with exertional angina. This value of free PAL1 in patients with vasospastic angina was closely associated with the duration of ischemic attacks. These results suggested that the circadian fluctuation of fibrinolytic components may be an important factor that leads to coronary thrombosis at the time of coronary spasm, especially in the early morning. (AM HEART J 1992;124:654.)

Kazuyuki Sakata, MD,a Tsuneo Hoshino, MD,” Hiroshi Yoshida, MD,” Norihisa Ono, MD,” Seiji Ohtani, MD,” Shoichi Yokoyama, MD, a Noriko Tsuneo Kaburagi, MD,a Chinori Kurata, MD,b Tetsumei Urano, MD,C Yumiko Takada, MD,C and Akikazu Takada, MD” Shizuoka, Japan

Some patients with vasospastic angina, representing variant angina, exhibit the unique feature of circadian fluctuation of chest pain as a result of vasospasm.l The peak level of chest pain frequency occurs in the early morning, which is also the period of the highest incidence of acute myocardial infarction.2 A large number of coronary angiographic and autopsy studies have shown that in the majority of patients,3-5 acute myocardial infarction and unstable angina are caused by coronary thrombus formation. Therefore determination of a cause-and-effect relationship between spasm and thrombus is a most important problem in this clinical setting. Various systems including the coagulation system, the fibrinolytic system, and the platelet system contribute to thrombus From %he Department of Cardiology, Shizuoka General Hospital; “the Third Department of Internal Medicine; and cthe Department of Physiology; Hamamatsu University School of Medicine, Shizuoka, Japan. Received Reprint Shizuoka

4/l/39869

a54

for publication requests: General

Kazuyuki Hospital,

Jan.

13, 1992;

accepted

Apr.

12, 1992.

Sakata, MD, The Department of Cardiology, 4-27-l Kitaando, Shizuoka 420 Japan.

Mori, MD,a

formation. Recently, Ogawa et a1.l demonstrated that coronary spasm resulted in elevated levels of plasma fibrinopeptide A in patients with variant angina. Although it has been shown that spasm activates the coagulation system, whether fibrin formation leads to coronary obstruction ultimately depends on the fibrinolytic system. Several investigator&l1 have suggested that circadian variation of fibrinolytic activity in patients with coronary artery disease may be related to intracoronary thrombus formation. Therefore to elucidate the relationship between spasm and thrombus, it is important to investigate the circadian variation of fibrinolytic activity. In this study, our aim was to elucidate the circadian variation of fibrinolytic components in vasospastic angina and its relationship to myocardial ischemia that is a result of vasospasm. METHODS Patients.

Forty-two patients who underwent cardiac catheterization in our hospital for the evaluation of chest pain, from November 1990 to June 1991, were prospec-

tively selected for the present

study. The patients

were di-

Volume124 Number4

Fibrinolytic

vided into three groups, patients with vasospasticangina, patients with exertional angina, and normal subjects (Table I). The vasospastic angina group consisted of 14 patients, all of whom had chest pain that occurred mainly in the early morning and at rest. In all of these patients, chest pain wascompletely relieved after sublingual administration of nitroglycerin. Spontaneous chest pain with ST-segment elevation, as determined by ECG, was documented in two patients. Vasospasticangina wasconfirmed by the acetylcholine provocation test during coronary arteriography. None of the patients had significant coronary artery stenosis or prior myocardial infarction. In the present study, although we could not document spontaneous attacks with ST-segment elevation in a large number of patients with vasospasticangina, the clinical features of the patients were compatible with those of patients with variant angina. The exertional angina group consistedof nine patients, all of whom had significant coronary artery stenosiswithout prior myocardial infarction. No coronary vasospasm was found when the acetylcholine provocation test was performed. The normal group consistedof 19 patients with normal coronary arteries, as demonstrated by coronary angiography: thesepatients were the control subjects.All of the patients

in this group

had negative

results of acetylcholine

provocation tests and no atherosclerotic lesions in their coronary arteries. None of the 42 study patients received anticoagulant agents, antiplatelet agents, steroids, or ,&blockers either before or during this study, and they had no other organic diseases. Study protocol. Venousblood samplesweredrawn every 6 hours from all patients (6:OOAM, noon, 6:00PM, and midnight ), and 24-hour Holter ECG monitoring (Holter monitor, Del Mar Avionics, Irvine, Calif.) werecarried out in the weekbefore cardiac catheterization wasperformed. Within 1 week after diagnostic cardiac catheterization, patients with vasospasticand exertional anginaunderwent exercise myocardial scintigraphy. Blood samples.Blood sampleswere collected after the patients had been lying in bed undisturbed for at least 10 minutes. Blood wasdrawn by venipuncture, without venous occlusion,by a specially trained physician. After collection, 9 parts of blood were mixed with 1 part of 3.1310sodium citrate solution. Plasmawasseparatedby centrifugation at 3300 rpm for 15 minutes and was immediately frozen and stored at, - 70’ C until it wasassayedfor fibrinolytic components. Tissue plasminogen activator antigen. Tissue plasminogen activator (t-PA) antigen was measuredby the enzyme immunoassaymethod.12Anti-t-PA monoclonalantibody (2:2 BIO, IMCO, Stockholm, Sweden) and B-Dgalactosidase-labeledanti-t-PA polyclonal Fab’ fragments were used asthe first and secondantibodies, respectively. The amounts of t-PA were calculated by comparisonwith a standard curve, which wasconstructed with purified t-PA (TD t-PA, Toyobo Co. and Daiichi Seiyaku Co., Tokyo, Japan).

components

855

in uasospastic angina

Table I. Group characteristics Normal

Number Age (yr)* Female/Male

DM Hypertension FH Smoking Hyperlipidemia Triglyceride (mg/dl)* BMI (kg/m2)*

VSA

EAP

14

9 55 * 12

19

54 i 10 5114

1 4 1 11 2 125 & 61 22 2 2

57

i- 10

l/13

1 4 1 9 2 149 + 65 23 + 2

l/8 3 4 1 4 2 165 k 57 23 + 4

VSA, Vasospastic angina; EAP, exertional angina; DM, diabetus mellitus; FH, family history of ischemic heart disease; BMI. body mass index, *Values are expressed as mean f SD. There were no significant differences among the three groups.

Free plasma activator inhibitor antigen (PAI-l), t-PA/ PAL1 complex, and total PAI-1. Plasma levels of free PAI-1, t-PA/PAI-1 complex, and total PAI- were measured by a previously describedmethod.13-16 Briefly, antiPAI- monoclonal antibody wasemployed as the first antibody and anti-t-PA polyclonal antibody-fi-galactosidase conjugate was used as the second antibody. Plasma was usedfor the t-PA/PAI-1 complex assay,and the amounts of t-PApAIcomplex were calculated from the amounts of t-PA in the complex. To measuretotal PAI-1, excess amounts of t-PA were added to the plasmabefore the assay to convert all of the PAI- to the form complexedwith t-PA. The concentration of free PAI- in the plasmawas calculated by subtracting the concentration of the t-PA/ PAI- complex from that of total PAI-1. Cardiac catheterization and acetylcholine provocation test. Coronary angiography was performed by application of a standard Judkins technique to all patients. Before coronary arteriography was performed, a temporary pacing catheter wasinserted into the right ventricle via the femoral vein; this wasconnectedto a pulse generator with a demand-driven rate of 40 beats/min. Blood pressurewas monitored through a Judkins catheter. Standard la-lead ECGs were recorded continuously with a six-channel recorder. After control coronary arteriograms were obtained; acetylcholine, which had been dissolved in saline solution, wasinjected in incremental dosesof 20and 100pg directly into either of the coronary arteries through the Judkins catheter. Coronary arteriography wasperformed 1 minute after acetylcholine was injected over 1 minute. Acetylcholine administration was stopped and coronary arteriography was performed immediately after the onset of chest pain with ischemicECG changes.Chest pain was relieved promptly, either spontaneously or by intracoronary injection of nitroglycerin. The intracoronary injection of acetylcholine was then repeated in the other coronary artery. After the acetylcholine provocation test, coronary angiography was performed after intracoronary injection of 0.3 mg of nitroglycerin. The resultsof coronary arteriography after injection of nitroglycerin were classified ac-

858

Sakata et al.

American

Table II. Characteristics

October 1992 Heart Journal

of vasospastic angina Coronary

Patient no.

E&R E&R R R R

2 3

6

11 12 13 14

Segments

Anterior Inferior No change Leads I aVL ST elevation

R E&R E&R R R E&R R R R

8 9 10

Ex Tl-201

ECG changes during attack

Type of chest pain

of the coronary ECT,

Exercise

arteries thallium-201

Tl-201 ECT location of transient defect

angiogram

Er

I aVL

are defined in accordance emission

computed

S7 s14 S2,6,13 S6 SS S6

cording to the reporting system of the American Heart Association.17 Coronary artery spasm that was induced by acetylcholine was considered to be present if total or subtotal occlusion of the involved artery occurred in association with chest pain and ischemic ST-segment changes. Twenty-four-hour Holter ECG. All patients underwent Holter monitoring with a two-channel recorder (model 453 A, Del Mar Avionics, Irvine, Calif.) for at least 24 hours on the day of blood sampling. An ischemic attack was defined as the occurrence of more than 2.0 mm ST-segment elevation or the occurrence of more than 2.0 mm ST-segment depression 80 msec after the J point was reached that lasted for at least 1 minute. Exercise myocardial scintigraphy. To investigate exercise-induced vasospasm and exercise-induced ischemia, patients with vasospastic angina and those with exertional angina were subjected to a symptom-limited treadmill exercise test within 1 week after cardiac catheterization. In this test the Bruce protocol was employed, and the test was carried out while patients fasted. The exercise was continued until progressive chest pain or leg fatigue developed. At this point, 111 MBq of thallium-201 was injected through an intravenous cannula, which had been previously inserted into a peripheral vein. All patients continued exercising at the same work load for another minute. Patients were routinely instructed to discontinue antianginal drugs for 24 hours before the test. ECG monitoring was performed throughout the exercise, and both blood pressure and heart rate were recorded before exercise, every 3 minutes during exercise, and at the time of thallium-201 injection. Standard 12-lead ECGs were recorded before and immediately after exercise while patients were in the supine position. Tomographic images of the heart were obtained by exercise thallium-201 emission computed tomography with the

Heart Association acetylchofine; NTG,

Normal 50 “i 50c; 50 r‘,l Normal Normal Normal Normal 25 “A

Sl 4PL

S6 S2,8 s9 S6

and VI-4

Committee nitroglycerin;

Report. E. effort;

NTG

Normal Normal Normal Normal Normal

Sl

and VI-V4

with the American tomography; Ach.

After

S6

Anterior Anterior Inferior Inferior Anterior Leads

Ach-induced spasm

R, rest.

use of a large field-of-view rotating gamma camera (Shimadzu Nuclear Camera, 500R, Shimadzu Corp., Kyoto, Japan), which was fitted with a high-resolution collimeter, which was fed to an on-line computer system (Scintipac 700, Shimadzu Corp.). Thirty-two projections over a range of 180 degrees, from the 45 degree right anterior oblique position to the 45 degree left posterior oblique position, were acquired on a 64 X 64 matrix of a 38 cm field of view for 30 seconds per image. Initial imaging was begun within 5 minutes after completion of exercise. Delayed images were obtained 4 hours after injection of thallium-201 chloride. Neither attenuation nor scatter correction was used. Orthogonal images were generated by oblique angle reconstruction, which produced vertical long-axis, short-axis, and horizontal long-axis slices of 6 mm in thickness. Quantitative analysis of the myocardial perfusion defect was performed with a computerized two-dimensional polar map of the three-dimensional myocardial radionuclide activity. Maps were generated by means of circumferential profile analysis, whereby the pixel count activity from the center to the outer boundary of each short axis was determined along radians spaced at B-degree intervals over 360 degrees. Count values at each point in the profile were then normalized to the maximal counts in each profile image. Individual slices from the cardiac apex to the base were displayed in the polar map as concentric rings from the center to the periphery of the map. The means and standard deviations of the normal values at each point were determined from the values that were calculated for 15 normal subjects. The exercise-induced ischemic area in each patient was defined as the area that showed counts below 2 standard deviations of the normal values on both the stress and washout polar maps. Statistical analysis. The results were expressed as means ? SD. Data were analyzed by one-way analysis of

Volume Number

Table

124 4

Fibrinolytic

components

in vasospastic angina

857

III. Total duration of ischemicattack Midnight-f?:00 (min) WA

48 2 47 16 i 36

EAP p Value Values are expressed

6.40 M-Noon (min)

AM

Noon-6:00 (mini

PM

Noon-Midnight (min)

23 k 17 46 +- 37

26 + 38 44 k 18

26 + 50 30 + 21

NS

NS

NS

0.01 as mean c SD.

NS,Norsigniticant; otherabbreviations asin TableI. Table

IV. Plasma levels of fibrinolytic components 6~00

t-PA (rig/ml) Normal VSA EAP Free

PAI-

Normal VSA

EAP t-PApAINormal

VSA EAP Total

PAI-

Normal

VSA EAP

AM

Noon

8.2 + 3.7 12.7 k 5.F

6.9 + 2.6 11.8 * 5.3t

12.2+ 3.8*

6:OO

PM

Midnight

p Value