- Email: [email protected]
Fibrinopeptide A excretion in urine: A marker of the cumulative thrombin activity in stable versus unstable angina patients
Fibrinopeptide A Excretion in Urine: A Marker of the Cumulative Thrombin Activii in Stable Versus Unstable Angina Patients Diego Ardissino, MD, Maria Grazia Gamba, MD, Piera Angelica Merlini, MD, Albert0 Rolla, MD, Paolo Barberis, MD, Gloria Demicheli, MD, Sophie Testa, MD, Nicola Bruno, MD, and Giuseppe Specchia, MD
Plasma levels and 24-hour urine excretion of fibrinopeptii A were measured in a consecutiie series of 176 patiintswith angina pectoris. Si four patients had stable angina and 115 patients had unstable angina. Urine was collected over 24 hours the day before coronary arteriography, and blood samples were taken at the end of urine collection. When the values of fibrinopeptiie A in plasma and in the 24-hour urine specimens were compared, no significant correlation was found in patients with either stable (es = 0.16, difference not significant) and unstable (es = 0.07, diierence not significant) angina. The concentrations of fibrinopeptkie A in the plasma did not diier significantly when patientswith stable angina (range 0.1 to 62.6, median 7.4 ng/mL) were compared with patients with unstable angina (range 0.2 to 627, median 14 ng/mL, p = 0.066), whereas fibrinopeptii A 24-hour urinary excretion was significantly higher in patients with unstable angina (range 0.3 to 36.5 median 126 pg/24 hr) than in patients with stable angina (range 0.4 to 36.1, median 3.6 pg/24 hr, p < 0.001). Twenty-four-hour urine excretion of fibrinopeptii A in patients with unstable angina and a@ographically documented intracoronary thrombi were hiir than the corresponding values in patients with unstable angina without such angiographii characteristic (p < 0.001). The largest increase in plasma and urine concentratiin of fibrinopeptkie A was observed in patii whose first episode of angina at rest occurred within the previous 46 hours. We conclude that the cumulatlve thrombin activity, assessed by 24-hour urinary excretion of fibrinopeptkle A, is a more useFrom the Division of Cardiology and the Department of Internal Medicine II, IRCCS, Policlinico S. Matteo, University Hospital, Pavia, and Division of Cardiology II, Niguarda Hospital, Milan, Italy. Address for reprints: Diego Ardissino, MD, Divisione di Cardiologia, IRCCS, Policlinico S. Matteo, 27100 Pavia, Italy.
58B
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 68
ful index, compared with single fibrinopeptkle A measurement in plasma, for discriminating between patients with stable and with unstable angina pectoris. (Am J Cardiil1661;68:66B+3lB)
T
he pathophysiology of unstable angina is complex but in most patients it appears to involve an interaction between fixed atherosclerotic coronary stenosis and dynamic coronary obstruction mediated by either coronary vasoconstriction or transient thrombotic occlusion. The hypothesis that the process of plaque complication and thrombus formation is an important pathogenetic mechanism of unstable angina has been strongly supported by results obtained using recent technical improvements, such as serial coronary arteriography’ and high resolution fiberoptic angioscopy,’ and by histopathologic and histochemical studies.3,4Biochemical markers of active thrombotic process have also been documented in patients with acute coronary syndromes.5-8Plasma fibrinopeptide A, a sensitive and specific index of thrombin activity, has been reported to be elevated in patients with acute myocardial infarction and unstable angina pectoris.5-7However, because of the short half-life of fibrinopeptide A (3 to 5 minutes), single measurements in plasma provide data relevant only to a short time period,g”Oand this could be a limitation in the setting of unstable angina where the intermittency of thrombus formation has been clearly documented.3 Moreover, plasma measurements of fibrinopeptide A are subject to sampling and analytic errors due to venipuncture artifacts or to dilutional effect.” For these considerations, we studied prospectiyely patients with stable and unstable angina pectoris using a noninvasive method that assessesthe cumulative effect of thrombin action on fibrinogen and avoids sampling problems-that is, measurement SEPTEMBER 3, 1991
of the amount of fibrinopeptide A excretion in 24-hour urine. In addition, we compared fibrinopeptide A 24-hour excretion of different subsets of patients with unstable angina to determine whether thrombin generation simply reflects frequent venipuncture and possibly other factors associated with intensive care or it is a specific finding related to intracoronary thrombus formation. METHODS Study population: The study population consisted of 179 consecutive patients who were hospitalized in our institution with a clinical diagnosis of either stable or unstable angina pectoris. Patients with unstable angina were, admitted in the coronary care unit, whereas patients with stable angina were referred to the Cardiology Department for elective coronary catheterization. The diagnosis of stable and unstable angina was made independently by 2 experienced cardiologists who were unaware of the results of the biochemical analysis. Unstable angina was defined as: (1) chest pain of recent onset; (2) sudden worsening of preexisting stable angina or postinfarction angina developing after an asymptomatic period of > 12 weeks and classified as Canadian Cardiovascular Society functional class III-IV. Prerequisite for the inclusion in the study was the occurrence of chest pain at rest associated with transient repolarization changes (ST-segment depression or elevation of 2 1 mm, 80 ms after the J point or pseudonormalization of previously negative T waves). Within the group of patients with unstable angina, 25 patients were identified as having Prinzmetal’s angina by the following characteristics: chest pain at rest promptly relieved by nitroglycerin and objective demonstration of transient ST-segment elevation during spontaneous attacks. Patients whose first episode of angina at rest occurred within the previous 48 hours were classified as having acute unstable angina. The following 3 characteristics were required for the diagnosis of stable angina: (1) history of typical chest pain induced by exercise or usual daily activity and lasting r6 months; (2) positive exercise stress test, defined as the occurrence of STsegment depression or elevation of r 1 mm, 80 ms after the J point, with or without chest pain; and (3) angiographically documented coronary artery disease. EXCIUBIO~B: Patients with the following conditions were excluded from the study: (1) concomitant peripheral vascular disorders or valvular heart disease (29 patients); (2) status of postcoronary
artery bypass surgery or coronary angioplasty (31 patients); (3) Q- wave myocardial infarction within the previous 6 months (22 patients); (4) disorder of hemostasis (2 patients); (4) proteinuria or chronic renal failure (3 patients); (5) hematuria (17 patients); (5) other conditions known to increase fibrinopeptide A levels, such as malignancy, infections and chronic inflammatory disease (3 patients). Patients were also excluded from the study if they were taking anticoagulants. Study design: After determination of elegibility, all patients underwent 24-hour urine collection at the end of which a blood sample was taken and coronary arteriography was performed. All patients in this protocol were followed until they were discharged or until another procedure, such as angioplasty or coronary bypass surgery, was performed. All patients who entered the study gave full informed consent and the study protocol was approved by the Institutional Committees. Sample collection and processing: BLOOD: All samples were collected immediately before coronary arteriography by the same investigator, who was unaware of the clinical diagnosis. By means of clean venipuncture, using a 19 gauge needle, without tourniquet and with minimal stasis, the first 3 ml of venous blood was discarded, after which 9 ml of venous blood was drawn into plastic tubes containing 0.1 ml of fibrinopeptide A anticoagulant supplied by the manufacturer of the assay kit. Samples were immediately stored in ice and centrifuged at 2000 g for 20 minutes to obtain platelet-poor plasma and frozen at -70°C. Fibrinopeptide A was measured in paired samples run in duplicate using commercial radioimmunoassay kits (RIA-mat FPA, Mallinckrodt Diagnostica, St Louis, Missouri). The procedures outlined by the manufacturer were followed. Results are expressed as average of the paired samples. Assays of fibrinopeptide A in our laboratory obtained from 18 control subjects selected from among patients hospitalized for electrophysiologic evaluation without clinical evidence of coronary artery disease gave a mean value of 4.5 2 2.36 rig/ml. URINE: Urine was collected over a 24-hour period. The collection started the day before coronary arteriography and ended immediately before blood sampling. Urine volume and pH were recorded and all samples were tested for the presence of blood or protein and immediately stored at -70°C. Before fibrinopeptide A determination, pH was corrected to 8.0 by adding hydrochloric acid 1 M or sodium hydroxide 1 M. Urine was assayed in paired samples for fibrinopeptide A in A SYMPOSIUM:
MANAGEMENT
OF UNSTABLE
ANGINA
596
the same way as for plasma except that bentonite absorption was omitted. Results are expressed as the amount of 24-hour fibrinopeptide A urinary excretion obtained by multiplying the average of the paired samples with the urine volume. Fibrinopeptide A 24-hour urinary excretion in control patients gave a mean value of 6.42 +- 3.10 pg/24 hr. The correlation coefficient of plasma and urinary fibrinopeptide Awas rs = 0.65 (p = 0.003). Coronary angiography Selectivecoronary arteriography was performed using standard Sones’ or Judkins’ technique. Narrowing more than 70% was considered significant coronary stenosis. Left main coronary stenosis 250% was considered significant. Patients were classified as having 1-, 2- ‘or 3-vesseldisease according to the number of vessels with significant stenosis. Qualitative morphologic analysis of the lesions was performed using the method of Ambrose et all2 by two experienced cardiologists who were blinded with respect to the biochemical results. In addition to lesion classification an attempt was made to identify intracoronary thrombi defined as spherical ovoid or irregular intraluminal filling defects surrounded at least on three sides by contrast medium just distal to or within a coronary stenosis.13 In instancesof disagreement between the two observers, intracoronary thrombus was recorded as absent. Statistical analysis: The clinical and angiographic characteristics of the two study groups described in Table I were compared by the unpaired Student t test for continuous data and by the chi-square test for categorical data. The fibrinopeptide A plasma and 24-hour urinary levels were initially assessedfor normality and were found not to be normally distributed. The data were than analyzed by the Mann-Whitney U test. Correlation TABLE I Clinical and Angiographic Characteristics of Patients with Stable and Unstable Angina
Characteristics Age (years) Male (%) Prior myocardial infarction, n (%I Prinzmetal’s angina, n f%) First angina1 attack within the preceding48 hours, n (%I Diseased vessels, n (%) 0 1 2 3 Left main stenosis lntracoronary thrombi, n f%)
Stable Angina (n = 64)
Unstable Angina (n = 115)
5729 46 (72) 27 (42)
59 87 41 25 31
+- 9 (76) (36) (22) (27)
NS NS NS
5 43 36 30 6 39
(5) (37) (31) (26) (5) (34)
NS NS NS NS NS
-
0 17 21 26 2 9
(0) (27) (33) (40) (3) (14)
p Value
NS = difference not significant.
60B
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 68
between urinary and plasma fibrinopeptide A was performed by using the Spearman test. A difference is described as significant if the p value is < 0.05.
RESULTS
The clinical and angiographic characteristics of the 179 patients constituting the study population are shown in Table I. Sixty-four patients had stable angina pectoris, and the remaining 115 patients were classified as having unstable angina. In 7 stable and in 6 unstable angina patients, blood samples were technically inadequate and the results were excluded from the analysis.The concentrations of fibrinopeptide A in the plasma did not differed significantly when the stable patients (range 0.1 to 82.6, median 7.4 rig/ml) were compared with patients with unstable angina (range 0.2 to 61.7, median 14 rig/ml, p = 0.055), whereas fibrinopeptide A 24-hour urinary excretion was significantly higher in patients with unstable angina (range 0.3 to 38.1, median 11.8 pg/24 hr) than in patients with stable angina (range 0.4 to 38.1, median 3.8 pg/24 hr; p < 0.001) (Figure 1). No correlation was found between plasma fibrinopeptide A and urinary excretion of fibrinopeptide A either in stable (rs = 0.16, p = 0.23) or unstable patients (rs = 0.07, p = 0.46) (Figure 2). Unstable angina patients were then separated into different subgroups according to the presence or absence of Prinzmetal’s angina, angiographic evidence of intracoronary thrombi or acute unstable angina. The fibrinopeptide A plasma levels were found significantly higher in patients with acute unstable angina when compared with those without such characteristic (p = 0.02). Fibrinopeptide A urinary excretion was significantly lower in patients with Prinzmetal’s angina (p
0
8
0
p<.oo1 P=
,055
0
0
. .. 8 . :f
0
. .
0
l
*: .
.: .
N Range Median
FIGURE l. Plasma concentrations unstable (0) anghm pectorls.
:*:t:. . I ,a...A%
.
.‘2 ..
8
I
k .f 1 0 5 5
l .
l;
I
A’.-.
--I.
0
Unstable angina 109 0.2-61.7 14
I
Stable Siable angina 57 0.4-38.1 3.8
N Range Median
and 2441our urinary excretion of flbrhmpeptide
thrombus (range 2.4 to 38.1, median 21.3 ug/24 hr) (difference not significant). DISCUSSION Plasma versus urinary levels of fibrinopeptide A: The syndrome of unstable angina appears to be frequently associated with intracoronary
0
i
0 0 @
t15
i II . ..*-
m.
Stable angina 57 0.1-82.6 7.4
V3 2 20
i l i:) . . .*.. .
0
;3 ‘m 2 5
8
0
:
N =57 rs = 0.16 p =NS 0
0
Unstable U&able angina 109 0.3-38.1 11.8
A (FPA) in patients with stable (0) and
thrombosis and the detection of an ongoing thrombotic process could serve to indicate the need for an aggressive antithrombotic regimen. However, at present a reliable routine technique for the identification of intracoronary thrombi in humans is not available. Coronary arteriography, which is commonly used to visualize thrombi, has been reported
. . I
8.
.
51:.
m
.
... .
O-5.. l * 5 : 0. 5,,
0
.
N =109 IS = 0.07 p = NS .
.
.
.
.
l
. .
I
0
102030405060708090 Plasma FPA (nglml)
0
I
I
,
,
,
,
,
,
,
,
,
,
,
,
,
10 20 30 40 50 60 Plasma FPA (nglml)
FIGURE 2. Correlations between plasma concentrations and 24-hour urinary excretion of fibrhwpeptide tients with stable (0) and unstable (e) anghm pectoris. NS = difference not signkant.
,
,
7-O
A (FPA) In pa-
A SYMPOSIUM: MANAGEMENT OF UNSTABLE ANGINA
6iB
TABLE II Plasma levels and 24-Hour Urinary Excretion of Fibrinopeptide A (FPA) in Different Subgroups of Patients with Unstable Angina Pectoris
Patients (n) Plasma FPA @/ml) Range Median Urinary FPA (pg/24 Range Median
Prinzmetal’s Angina
lntracoronary Thrombus
Acute Unstable Angina
Present
Absent
Present
Absent
Present
Absent
25
90
39
76
31
84
0.2-58.6. 17.7
0.5-61.7 13.4 p=NS
1.6-46.7
0.2-61.7
0.2-61.7
14.3
14
11.7
1.2-60.3 18.9 p = 0.02
0.3-37.7 2.8
1.1-38.1 14.3 p = 0.001
2.4-38.1 21.3
6.9
hr)
to underestimate their real prevalence.’ Biochemical markers of activation of the coagulation system are of interest to determine noninvasively the activity of thrombus formation. Several studie?’ have evaluated various biochemical markers of thrombin generation in patients with coronary artery disease, but at present their value remains controversia1.14’15 Fibrinopeptide A, a breakdown product of the conversion of fibrinogen to fibrin, has been reported to be increased in the plasma of some patients with unstable angina.5-7However, plasma measurements of fibrinopeptide A are subject to artifacts during sampling’l and provide data relevant only to a limited time period due to the short half-life of the peptide.‘,” These difficulties may be overcome by studying urine samples.16,17 In normal persons 0.5% to 1% of the fibrinopeptide A produced is rapidly cleared into urine, where it is measurable by radioimmunoassay technique, and the amount of urine excretion correlates with simultaneously obtained plasma Spot urine17 fibrinopeptide A levels delevels.16’18 tect acute changes of thrombin activity, whereas 24-hour urine excretion provides a cumulative measure of overall thrombin action on fibrinogen. In the present study a correlation between plasma levels and 24-hour urinary excretion of fibrinopeptide A was found only in normal subjects and not in patients with stable and unstable angina. It is possible to hypothesize that in patients with coronary artery disease activation of the coagulation system may occur intermittently; thus, episodic determinations of fibrinopeptide A do not reflect the overall fibrin formation in vivo. Cumulative thrombin activity in unstable angina: This study shows that the 24-hour urinary
excretion of fibrinopeptide A was higher in patients with unstable angina compared with that in a control population of patients with stable angina. Plasma fibrinopeptide A levels did not differ between the two groups. These observations suggest that the cumulative thrombin activity is a more useful index to discriminate between various clini62B
THE AMERICAN
p=NS
JOURNAL
OF CARDIOLOGY
VOLUME
68
0.3-34.5 p = 0.001
1.8-38.1 23.3
0.3-34.5 7.5 p = 0.001
cal forms of angina pectoris, although the wide overlap does not allow any conclusion to be drawn for a particular patient. The finding that plasma levels of fibrinopeptide A are not significantly different in patients with stable and unstable angina is in contrast with previous studies. This discrepancy may be explained by the fact that our population consisted of a consecutive series of patients with a diagnosis of unstable angina, including patients with Prinzmetal’s and changing pattern angina. The lack of selection may have accounted for the wide variability of activation of the coagulation systemobserved within the group. The subset of patients with unstable angina and angiographically documented intracoronary thrombi had higher values of thrombin generation reflected by increased plasma levels and urinary excretion of fibrinopeptide A. High values were also found in patients with stable angina and intracoronary thrombi. These observations, together with the finding that in patients with vasospasticangina, i.e., Prinzmetal’s angina, the ,overall thrombin activity was low seem to indicate that fibrinopeptide A generation in patients with angina pectoris is a specific finding of activation of the coagulation system and it is not secondary to other factors related to the intensive care, such as frequent venipuncture or intravenous lines. It is interesting to note that the highest values of cumulative thrombin activity were obtained in patients with acute unstable angina, that is, in those with very recent onset of unstable angina who also presented with the highest prevalence of intracoronary thrombi. In conclusion, urinary excretion of fibrinopeptide A better reflects the overall effect of thrombin action on fibrinogen than a single plasma measurement, which gives only information on fibrinogen breakdown at the time of blood sampling. The cumulative thrombin activity, as assessedby 24hour urinary excretion of fibrinopeptide A, is able to discriminate between populations with various
SEPTEMBER
3, 1991
7. Theroux P, L&our JG, Leger-Gauthier C, DeLara J. Fibrinoopeptide A and platelet factor levelsin unstableangina.Circzdation1987;75:1%162. 8. Kruskal JB, Commerford PJ, Franks JJ, Kirsch RE. Fibrin and fibrmogenrelated antigensin patients with stable and unstable coronaty artery disease.N EnglJMed 1987;317:1361-1365. 9. Nossel HL, Yudehnan I, Canfteld E, Butler VP, SpanondisK, Wihner GD, Qureshi GD. Measurement of fibrinopeptide A in human blood. .I C&r Invest 1974;54:4>54. 10. Ariga T, Oshiba S, Akashi Y, Endoh E, Imai H, Sawai H. Turnover of fibrinopeptide A (FPA) in rabbits. ZIzrombRes 1985;38:137-147. 1F Nossel HL, Ti M, Kaplan KL, Sponondis K, Soland T, Butler VP. The generationoffibrinopeptide Amclinical blood samples.JC[inInvest 1976;58:1136REFERENCES 1144. L M&e A, Theroux P, Taeymans Y, Descoings B, Lesperance J, Waters D, l2. Ambrose JA, Winters SL, Stem A, Eng A, Teicholz LE, Gorlin R, Fuster V. Pelletier G, BourassaM. Unstable anginaand progressionof coronary atheroscle- Angiographicmorphologyand the pathogenesisof unstableanginapectoris.JAm rosis.NE~~JMed1983;309:685-689. Cdl Cardiol1985;5:609616. 2. ShermanCT, Litvack F, Grundfest W, Lee M, Hickey A, Chaux A, Kass R, 13. AmbroseJA, Hjemdah-Monsen C, Borrico S,ShermanW, Cohen M, Gorlin Blanche R, Matloff J, Morgenstem L, Ganz W, SwanHJ, Forrester .I. Coronary R, Fuster V. Quantitative and qualitative effects of irmacoronary streptokinasein angioscopyin patients with unstableanginapectoris.NEngl JMed 1986;315:913unstableanginaand non-Q wave infarction. JAm Coil Ca&l1987;9:1156-1165. 919. 3. Falk E. Unstable angina with fatal outcome: dynamic coronary thrombosis l4.Ambrose JA, Alexopoulos D. Thrombolysis in unstable angina: will the leading to infarction and/or suddendeath. Autopsy evidenceof recurrent mural beneficial effects of thrombolytic therapy in myocardial infarction apply to thrombosiswith peripheral embolization cuhninating in total vascular occlusion. patientswith unstableangina?JAm CONC&o1 1989;13:16&171. 15. Alexopoulos D, Ambrose JA, Stump D, Borrico S, Gorlin R, DeshmukhP, Circu[atin 1985;71:699-708. Fisher EA. Thrombosis-relatedmarkers in unstable angina pectoris.JAm Co11 4. Davies MJ, Thomas A. Thrombosii and acute coronary artery lesions in Car& 1991;17:866-871. suddencardiac ischemicdeath.NE& JMed 1984;310:1137-1140. 5. Neri Semeri GG, Gensini GF, Abbate R, Laureano R, Parodi 0. Is raised 16. Galhno A, Haeberli A, Straub PW. Fibrinopeptide A excretion in patients with atheroscleroticartery disease.ThrombRes 1985;38:237-244. plasma fibrinopeptide A a marker of acute coronary insufficiency? Lancet 17. Wilensky RL, Zeller JA, Wish M, Tulchinsky M. Urinary fibrinipeptide A 1980;2:982. 6. Eisemberg PR, Sherman LA, SchectmanK, Perez J, Sobel BE, Jatfe AS. levelsin ischemicheart disease.JAm Co11Cardiol1989;14:597-603. Fibrinopeptide A: a marker of acute coronary thronmbosis.Circulation 198371: iB. Alkjaersing N, Fletcher AP. Catabolism and excretion of fibrinopeptide A. 912-918. Blood 1982;60:148-156.
clinical forms of angina pectoris, although the result cannot be taken as an indication of angina1 activity in the individual patient. Further studies will be necessary to clarify the clinical value of urinary fibrinopeptide A in identifying high-risk patients for cardiac events to be treated with an aggressive antithrombotic therapy.
A SYMPOSIUM: MANAGEMENT OF UNSTABLE ANGINA
63B
Plasma levels and 24-hour urine excretion of fibrinopeptii A were measured in a consecutiie series of 176 patiintswith angina pectoris. Si four patients had stable angina and 115 patients had unstable angina. Urine was collected over 24 hours the day before coronary arteriography, and blood samples were taken at the end of urine collection. When the values of fibrinopeptiie A in plasma and in the 24-hour urine specimens were compared, no significant correlation was found in patients with either stable (es = 0.16, difference not significant) and unstable (es = 0.07, diierence not significant) angina. The concentrations of fibrinopeptkie A in the plasma did not diier significantly when patientswith stable angina (range 0.1 to 62.6, median 7.4 ng/mL) were compared with patients with unstable angina (range 0.2 to 627, median 14 ng/mL, p = 0.066), whereas fibrinopeptii A 24-hour urinary excretion was significantly higher in patients with unstable angina (range 0.3 to 36.5 median 126 pg/24 hr) than in patients with stable angina (range 0.4 to 36.1, median 3.6 pg/24 hr, p < 0.001). Twenty-four-hour urine excretion of fibrinopeptii A in patients with unstable angina and a@ographically documented intracoronary thrombi were hiir than the corresponding values in patients with unstable angina without such angiographii characteristic (p < 0.001). The largest increase in plasma and urine concentratiin of fibrinopeptkie A was observed in patii whose first episode of angina at rest occurred within the previous 46 hours. We conclude that the cumulatlve thrombin activity, assessed by 24-hour urinary excretion of fibrinopeptkle A, is a more useFrom the Division of Cardiology and the Department of Internal Medicine II, IRCCS, Policlinico S. Matteo, University Hospital, Pavia, and Division of Cardiology II, Niguarda Hospital, Milan, Italy. Address for reprints: Diego Ardissino, MD, Divisione di Cardiologia, IRCCS, Policlinico S. Matteo, 27100 Pavia, Italy.
58B
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 68
ful index, compared with single fibrinopeptkle A measurement in plasma, for discriminating between patients with stable and with unstable angina pectoris. (Am J Cardiil1661;68:66B+3lB)
T
he pathophysiology of unstable angina is complex but in most patients it appears to involve an interaction between fixed atherosclerotic coronary stenosis and dynamic coronary obstruction mediated by either coronary vasoconstriction or transient thrombotic occlusion. The hypothesis that the process of plaque complication and thrombus formation is an important pathogenetic mechanism of unstable angina has been strongly supported by results obtained using recent technical improvements, such as serial coronary arteriography’ and high resolution fiberoptic angioscopy,’ and by histopathologic and histochemical studies.3,4Biochemical markers of active thrombotic process have also been documented in patients with acute coronary syndromes.5-8Plasma fibrinopeptide A, a sensitive and specific index of thrombin activity, has been reported to be elevated in patients with acute myocardial infarction and unstable angina pectoris.5-7However, because of the short half-life of fibrinopeptide A (3 to 5 minutes), single measurements in plasma provide data relevant only to a short time period,g”Oand this could be a limitation in the setting of unstable angina where the intermittency of thrombus formation has been clearly documented.3 Moreover, plasma measurements of fibrinopeptide A are subject to sampling and analytic errors due to venipuncture artifacts or to dilutional effect.” For these considerations, we studied prospectiyely patients with stable and unstable angina pectoris using a noninvasive method that assessesthe cumulative effect of thrombin action on fibrinogen and avoids sampling problems-that is, measurement SEPTEMBER 3, 1991
of the amount of fibrinopeptide A excretion in 24-hour urine. In addition, we compared fibrinopeptide A 24-hour excretion of different subsets of patients with unstable angina to determine whether thrombin generation simply reflects frequent venipuncture and possibly other factors associated with intensive care or it is a specific finding related to intracoronary thrombus formation. METHODS Study population: The study population consisted of 179 consecutive patients who were hospitalized in our institution with a clinical diagnosis of either stable or unstable angina pectoris. Patients with unstable angina were, admitted in the coronary care unit, whereas patients with stable angina were referred to the Cardiology Department for elective coronary catheterization. The diagnosis of stable and unstable angina was made independently by 2 experienced cardiologists who were unaware of the results of the biochemical analysis. Unstable angina was defined as: (1) chest pain of recent onset; (2) sudden worsening of preexisting stable angina or postinfarction angina developing after an asymptomatic period of > 12 weeks and classified as Canadian Cardiovascular Society functional class III-IV. Prerequisite for the inclusion in the study was the occurrence of chest pain at rest associated with transient repolarization changes (ST-segment depression or elevation of 2 1 mm, 80 ms after the J point or pseudonormalization of previously negative T waves). Within the group of patients with unstable angina, 25 patients were identified as having Prinzmetal’s angina by the following characteristics: chest pain at rest promptly relieved by nitroglycerin and objective demonstration of transient ST-segment elevation during spontaneous attacks. Patients whose first episode of angina at rest occurred within the previous 48 hours were classified as having acute unstable angina. The following 3 characteristics were required for the diagnosis of stable angina: (1) history of typical chest pain induced by exercise or usual daily activity and lasting r6 months; (2) positive exercise stress test, defined as the occurrence of STsegment depression or elevation of r 1 mm, 80 ms after the J point, with or without chest pain; and (3) angiographically documented coronary artery disease. EXCIUBIO~B: Patients with the following conditions were excluded from the study: (1) concomitant peripheral vascular disorders or valvular heart disease (29 patients); (2) status of postcoronary
artery bypass surgery or coronary angioplasty (31 patients); (3) Q- wave myocardial infarction within the previous 6 months (22 patients); (4) disorder of hemostasis (2 patients); (4) proteinuria or chronic renal failure (3 patients); (5) hematuria (17 patients); (5) other conditions known to increase fibrinopeptide A levels, such as malignancy, infections and chronic inflammatory disease (3 patients). Patients were also excluded from the study if they were taking anticoagulants. Study design: After determination of elegibility, all patients underwent 24-hour urine collection at the end of which a blood sample was taken and coronary arteriography was performed. All patients in this protocol were followed until they were discharged or until another procedure, such as angioplasty or coronary bypass surgery, was performed. All patients who entered the study gave full informed consent and the study protocol was approved by the Institutional Committees. Sample collection and processing: BLOOD: All samples were collected immediately before coronary arteriography by the same investigator, who was unaware of the clinical diagnosis. By means of clean venipuncture, using a 19 gauge needle, without tourniquet and with minimal stasis, the first 3 ml of venous blood was discarded, after which 9 ml of venous blood was drawn into plastic tubes containing 0.1 ml of fibrinopeptide A anticoagulant supplied by the manufacturer of the assay kit. Samples were immediately stored in ice and centrifuged at 2000 g for 20 minutes to obtain platelet-poor plasma and frozen at -70°C. Fibrinopeptide A was measured in paired samples run in duplicate using commercial radioimmunoassay kits (RIA-mat FPA, Mallinckrodt Diagnostica, St Louis, Missouri). The procedures outlined by the manufacturer were followed. Results are expressed as average of the paired samples. Assays of fibrinopeptide A in our laboratory obtained from 18 control subjects selected from among patients hospitalized for electrophysiologic evaluation without clinical evidence of coronary artery disease gave a mean value of 4.5 2 2.36 rig/ml. URINE: Urine was collected over a 24-hour period. The collection started the day before coronary arteriography and ended immediately before blood sampling. Urine volume and pH were recorded and all samples were tested for the presence of blood or protein and immediately stored at -70°C. Before fibrinopeptide A determination, pH was corrected to 8.0 by adding hydrochloric acid 1 M or sodium hydroxide 1 M. Urine was assayed in paired samples for fibrinopeptide A in A SYMPOSIUM:
MANAGEMENT
OF UNSTABLE
ANGINA
596
the same way as for plasma except that bentonite absorption was omitted. Results are expressed as the amount of 24-hour fibrinopeptide A urinary excretion obtained by multiplying the average of the paired samples with the urine volume. Fibrinopeptide A 24-hour urinary excretion in control patients gave a mean value of 6.42 +- 3.10 pg/24 hr. The correlation coefficient of plasma and urinary fibrinopeptide Awas rs = 0.65 (p = 0.003). Coronary angiography Selectivecoronary arteriography was performed using standard Sones’ or Judkins’ technique. Narrowing more than 70% was considered significant coronary stenosis. Left main coronary stenosis 250% was considered significant. Patients were classified as having 1-, 2- ‘or 3-vesseldisease according to the number of vessels with significant stenosis. Qualitative morphologic analysis of the lesions was performed using the method of Ambrose et all2 by two experienced cardiologists who were blinded with respect to the biochemical results. In addition to lesion classification an attempt was made to identify intracoronary thrombi defined as spherical ovoid or irregular intraluminal filling defects surrounded at least on three sides by contrast medium just distal to or within a coronary stenosis.13 In instancesof disagreement between the two observers, intracoronary thrombus was recorded as absent. Statistical analysis: The clinical and angiographic characteristics of the two study groups described in Table I were compared by the unpaired Student t test for continuous data and by the chi-square test for categorical data. The fibrinopeptide A plasma and 24-hour urinary levels were initially assessedfor normality and were found not to be normally distributed. The data were than analyzed by the Mann-Whitney U test. Correlation TABLE I Clinical and Angiographic Characteristics of Patients with Stable and Unstable Angina
Characteristics Age (years) Male (%) Prior myocardial infarction, n (%I Prinzmetal’s angina, n f%) First angina1 attack within the preceding48 hours, n (%I Diseased vessels, n (%) 0 1 2 3 Left main stenosis lntracoronary thrombi, n f%)
Stable Angina (n = 64)
Unstable Angina (n = 115)
5729 46 (72) 27 (42)
59 87 41 25 31
+- 9 (76) (36) (22) (27)
NS NS NS
5 43 36 30 6 39
(5) (37) (31) (26) (5) (34)
NS NS NS NS NS
-
0 17 21 26 2 9
(0) (27) (33) (40) (3) (14)
p Value
NS = difference not significant.
60B
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 68
between urinary and plasma fibrinopeptide A was performed by using the Spearman test. A difference is described as significant if the p value is < 0.05.
RESULTS
The clinical and angiographic characteristics of the 179 patients constituting the study population are shown in Table I. Sixty-four patients had stable angina pectoris, and the remaining 115 patients were classified as having unstable angina. In 7 stable and in 6 unstable angina patients, blood samples were technically inadequate and the results were excluded from the analysis.The concentrations of fibrinopeptide A in the plasma did not differed significantly when the stable patients (range 0.1 to 82.6, median 7.4 rig/ml) were compared with patients with unstable angina (range 0.2 to 61.7, median 14 rig/ml, p = 0.055), whereas fibrinopeptide A 24-hour urinary excretion was significantly higher in patients with unstable angina (range 0.3 to 38.1, median 11.8 pg/24 hr) than in patients with stable angina (range 0.4 to 38.1, median 3.8 pg/24 hr; p < 0.001) (Figure 1). No correlation was found between plasma fibrinopeptide A and urinary excretion of fibrinopeptide A either in stable (rs = 0.16, p = 0.23) or unstable patients (rs = 0.07, p = 0.46) (Figure 2). Unstable angina patients were then separated into different subgroups according to the presence or absence of Prinzmetal’s angina, angiographic evidence of intracoronary thrombi or acute unstable angina. The fibrinopeptide A plasma levels were found significantly higher in patients with acute unstable angina when compared with those without such characteristic (p = 0.02). Fibrinopeptide A urinary excretion was significantly lower in patients with Prinzmetal’s angina (p
0
8
0
p<.oo1 P=
,055
0
0
. .. 8 . :f
0
. .
0
l
*: .
.: .
N Range Median
FIGURE l. Plasma concentrations unstable (0) anghm pectorls.
:*:t:. . I ,a...A%
.
.‘2 ..
8
I
k .f 1 0 5 5
l .
l;
I
A’.-.
--I.
0
Unstable angina 109 0.2-61.7 14
I
Stable Siable angina 57 0.4-38.1 3.8
N Range Median
and 2441our urinary excretion of flbrhmpeptide
thrombus (range 2.4 to 38.1, median 21.3 ug/24 hr) (difference not significant). DISCUSSION Plasma versus urinary levels of fibrinopeptide A: The syndrome of unstable angina appears to be frequently associated with intracoronary
0
i
0 0 @
t15
i II . ..*-
m.
Stable angina 57 0.1-82.6 7.4
V3 2 20
i l i:) . . .*.. .
0
;3 ‘m 2 5
8
0
:
N =57 rs = 0.16 p =NS 0
0
Unstable U&able angina 109 0.3-38.1 11.8
A (FPA) in patients with stable (0) and
thrombosis and the detection of an ongoing thrombotic process could serve to indicate the need for an aggressive antithrombotic regimen. However, at present a reliable routine technique for the identification of intracoronary thrombi in humans is not available. Coronary arteriography, which is commonly used to visualize thrombi, has been reported
. . I
8.
.
51:.
m
.
... .
O-5.. l * 5 : 0. 5,,
0
.
N =109 IS = 0.07 p = NS .
.
.
.
.
l
. .
I
0
102030405060708090 Plasma FPA (nglml)
0
I
I
,
,
,
,
,
,
,
,
,
,
,
,
,
10 20 30 40 50 60 Plasma FPA (nglml)
FIGURE 2. Correlations between plasma concentrations and 24-hour urinary excretion of fibrhwpeptide tients with stable (0) and unstable (e) anghm pectoris. NS = difference not signkant.
,
,
7-O
A (FPA) In pa-
A SYMPOSIUM: MANAGEMENT OF UNSTABLE ANGINA
6iB
TABLE II Plasma levels and 24-Hour Urinary Excretion of Fibrinopeptide A (FPA) in Different Subgroups of Patients with Unstable Angina Pectoris
Patients (n) Plasma FPA @/ml) Range Median Urinary FPA (pg/24 Range Median
Prinzmetal’s Angina
lntracoronary Thrombus
Acute Unstable Angina
Present
Absent
Present
Absent
Present
Absent
25
90
39
76
31
84
0.2-58.6. 17.7
0.5-61.7 13.4 p=NS
1.6-46.7
0.2-61.7
0.2-61.7
14.3
14
11.7
1.2-60.3 18.9 p = 0.02
0.3-37.7 2.8
1.1-38.1 14.3 p = 0.001
2.4-38.1 21.3
6.9
hr)
to underestimate their real prevalence.’ Biochemical markers of activation of the coagulation system are of interest to determine noninvasively the activity of thrombus formation. Several studie?’ have evaluated various biochemical markers of thrombin generation in patients with coronary artery disease, but at present their value remains controversia1.14’15 Fibrinopeptide A, a breakdown product of the conversion of fibrinogen to fibrin, has been reported to be increased in the plasma of some patients with unstable angina.5-7However, plasma measurements of fibrinopeptide A are subject to artifacts during sampling’l and provide data relevant only to a limited time period due to the short half-life of the peptide.‘,” These difficulties may be overcome by studying urine samples.16,17 In normal persons 0.5% to 1% of the fibrinopeptide A produced is rapidly cleared into urine, where it is measurable by radioimmunoassay technique, and the amount of urine excretion correlates with simultaneously obtained plasma Spot urine17 fibrinopeptide A levels delevels.16’18 tect acute changes of thrombin activity, whereas 24-hour urine excretion provides a cumulative measure of overall thrombin action on fibrinogen. In the present study a correlation between plasma levels and 24-hour urinary excretion of fibrinopeptide A was found only in normal subjects and not in patients with stable and unstable angina. It is possible to hypothesize that in patients with coronary artery disease activation of the coagulation system may occur intermittently; thus, episodic determinations of fibrinopeptide A do not reflect the overall fibrin formation in vivo. Cumulative thrombin activity in unstable angina: This study shows that the 24-hour urinary
excretion of fibrinopeptide A was higher in patients with unstable angina compared with that in a control population of patients with stable angina. Plasma fibrinopeptide A levels did not differ between the two groups. These observations suggest that the cumulative thrombin activity is a more useful index to discriminate between various clini62B
THE AMERICAN
p=NS
JOURNAL
OF CARDIOLOGY
VOLUME
68
0.3-34.5 p = 0.001
1.8-38.1 23.3
0.3-34.5 7.5 p = 0.001
cal forms of angina pectoris, although the wide overlap does not allow any conclusion to be drawn for a particular patient. The finding that plasma levels of fibrinopeptide A are not significantly different in patients with stable and unstable angina is in contrast with previous studies. This discrepancy may be explained by the fact that our population consisted of a consecutive series of patients with a diagnosis of unstable angina, including patients with Prinzmetal’s and changing pattern angina. The lack of selection may have accounted for the wide variability of activation of the coagulation systemobserved within the group. The subset of patients with unstable angina and angiographically documented intracoronary thrombi had higher values of thrombin generation reflected by increased plasma levels and urinary excretion of fibrinopeptide A. High values were also found in patients with stable angina and intracoronary thrombi. These observations, together with the finding that in patients with vasospasticangina, i.e., Prinzmetal’s angina, the ,overall thrombin activity was low seem to indicate that fibrinopeptide A generation in patients with angina pectoris is a specific finding of activation of the coagulation system and it is not secondary to other factors related to the intensive care, such as frequent venipuncture or intravenous lines. It is interesting to note that the highest values of cumulative thrombin activity were obtained in patients with acute unstable angina, that is, in those with very recent onset of unstable angina who also presented with the highest prevalence of intracoronary thrombi. In conclusion, urinary excretion of fibrinopeptide A better reflects the overall effect of thrombin action on fibrinogen than a single plasma measurement, which gives only information on fibrinogen breakdown at the time of blood sampling. The cumulative thrombin activity, as assessedby 24hour urinary excretion of fibrinopeptide A, is able to discriminate between populations with various
SEPTEMBER
3, 1991
7. Theroux P, L&our JG, Leger-Gauthier C, DeLara J. Fibrinoopeptide A and platelet factor levelsin unstableangina.Circzdation1987;75:1%162. 8. Kruskal JB, Commerford PJ, Franks JJ, Kirsch RE. Fibrin and fibrmogenrelated antigensin patients with stable and unstable coronaty artery disease.N EnglJMed 1987;317:1361-1365. 9. Nossel HL, Yudehnan I, Canfteld E, Butler VP, SpanondisK, Wihner GD, Qureshi GD. Measurement of fibrinopeptide A in human blood. .I C&r Invest 1974;54:4>54. 10. Ariga T, Oshiba S, Akashi Y, Endoh E, Imai H, Sawai H. Turnover of fibrinopeptide A (FPA) in rabbits. ZIzrombRes 1985;38:137-147. 1F Nossel HL, Ti M, Kaplan KL, Sponondis K, Soland T, Butler VP. The generationoffibrinopeptide Amclinical blood samples.JC[inInvest 1976;58:1136REFERENCES 1144. L M&e A, Theroux P, Taeymans Y, Descoings B, Lesperance J, Waters D, l2. Ambrose JA, Winters SL, Stem A, Eng A, Teicholz LE, Gorlin R, Fuster V. Pelletier G, BourassaM. Unstable anginaand progressionof coronary atheroscle- Angiographicmorphologyand the pathogenesisof unstableanginapectoris.JAm rosis.NE~~JMed1983;309:685-689. Cdl Cardiol1985;5:609616. 2. ShermanCT, Litvack F, Grundfest W, Lee M, Hickey A, Chaux A, Kass R, 13. AmbroseJA, Hjemdah-Monsen C, Borrico S,ShermanW, Cohen M, Gorlin Blanche R, Matloff J, Morgenstem L, Ganz W, SwanHJ, Forrester .I. Coronary R, Fuster V. Quantitative and qualitative effects of irmacoronary streptokinasein angioscopyin patients with unstableanginapectoris.NEngl JMed 1986;315:913unstableanginaand non-Q wave infarction. JAm Coil Ca&l1987;9:1156-1165. 919. 3. Falk E. Unstable angina with fatal outcome: dynamic coronary thrombosis l4.Ambrose JA, Alexopoulos D. Thrombolysis in unstable angina: will the leading to infarction and/or suddendeath. Autopsy evidenceof recurrent mural beneficial effects of thrombolytic therapy in myocardial infarction apply to thrombosiswith peripheral embolization cuhninating in total vascular occlusion. patientswith unstableangina?JAm CONC&o1 1989;13:16&171. 15. Alexopoulos D, Ambrose JA, Stump D, Borrico S, Gorlin R, DeshmukhP, Circu[atin 1985;71:699-708. Fisher EA. Thrombosis-relatedmarkers in unstable angina pectoris.JAm Co11 4. Davies MJ, Thomas A. Thrombosii and acute coronary artery lesions in Car& 1991;17:866-871. suddencardiac ischemicdeath.NE& JMed 1984;310:1137-1140. 5. Neri Semeri GG, Gensini GF, Abbate R, Laureano R, Parodi 0. Is raised 16. Galhno A, Haeberli A, Straub PW. Fibrinopeptide A excretion in patients with atheroscleroticartery disease.ThrombRes 1985;38:237-244. plasma fibrinopeptide A a marker of acute coronary insufficiency? Lancet 17. Wilensky RL, Zeller JA, Wish M, Tulchinsky M. Urinary fibrinipeptide A 1980;2:982. 6. Eisemberg PR, Sherman LA, SchectmanK, Perez J, Sobel BE, Jatfe AS. levelsin ischemicheart disease.JAm Co11Cardiol1989;14:597-603. Fibrinopeptide A: a marker of acute coronary thronmbosis.Circulation 198371: iB. Alkjaersing N, Fletcher AP. Catabolism and excretion of fibrinopeptide A. 912-918. Blood 1982;60:148-156.
clinical forms of angina pectoris, although the result cannot be taken as an indication of angina1 activity in the individual patient. Further studies will be necessary to clarify the clinical value of urinary fibrinopeptide A in identifying high-risk patients for cardiac events to be treated with an aggressive antithrombotic therapy.
A SYMPOSIUM: MANAGEMENT OF UNSTABLE ANGINA
63B