Itlternational Journal of Cardiology. 34 (1992) 307-318 ? 1992 Elsevier Science Publishers B.V. All rights reserved
CARD10
0167.5273/92/$05.00
01399
Biological risk factors for sudden death in patients with coronary artery disease and without heart failure Philippe Lauribe ‘, Daniel Benchimol ‘, Jean-Frangois Dartigues ‘?,Sobhi Dada ‘, Hklkne Benchimol ‘, Fransoise Drouillet ‘, Jacques Bonnet ‘A and Henri Bricaud 2 ’ Senkr de Cardiologie, H&ital Nerd, Marseille. France: ’ Chique M6dicale Cardiologique. Hspital Cirrdiologique. Pessac. France: .’ Dt!partement d’lnformatique MPdicale et de Statistiques. UniL,ersite’de Bordeaux II. INSERM Unitti 3.U.l.BordealLv. France; ’ INSERM Unitr’ X de Cardiologie. Pessac, France (Received
30 November
1990; revision
accepted
10 October
lY91)
Lauribe P, Benchimol D, Dartigues J-F, Dada S, Benchimol H, Drouillet F. Bonnet J, Bricaud H. Biological risk factors for sudden death in patients with coronary artery disease and without heart failure. Int J Cardiol 1992;34:307-318. In a study of biological risk factors for sudden death in patients with coronary artery disease, 320 patients were, prospectively, recruited and followed-up over two years. None of the patients had heart failure or recent myocardial infarction. The following variables were recorded: previous acute myocardial infarction, hypertension, smoking habits, ventricular arrhythmia; the angiographic variables included: left ventricular ejection fraction, Jenkins’ and mean atherosclerotic scores; lipid profile: cholesterol, triglycerides, high density lipoprotein cholesterol, low density lipoprotein cholesterol, apolipoproteins Al and B; hemostatic profile: fibrinogen, fibrinopeptide A, antithrombin III, factor VIII antigen, factor VIII coagulant, protein C, plasminogen, a2-antiplasmin, euglobulin clot lysis time and tissue plasminogen activator before and after venous occclusion, tissue plasminogen activator inhibitor, platelet factor 4, @thromboglobulin. During the follow-up period, 12 of the patients died suddenly. In these patients, ejection fraction was lower: 49 _t 16% versus 61 f 14% for the other patients (P < 0.02), fibrinogen higher: 3.9 j, 0.8 g/l versus 3.5 + 0.8 for the living patients (P < 0.05) and protein C lower: 89 + 39% versus 111 ? 39% (P = 0.06) for the other patients. In multivariate analysis: lower ejection fraction (P < O.OOS), older age (P < 0.03) and lower protein C (P < 0.011 were correlated with sudden death. Among the patients with coronary artery disease, the raised fibrinogen and the decreased protein C appeared to be risk factors for sudden cardiac death. These alterations reflected a prothrombotic state which might increase the ischemic risk, due to an acute thrombosis, leading to the fatal ventricular arrhythmia. Determination of these hemostatic variables might be a useful adjunct for assessment of the vital prognosis of patients with coronary artery disease, especially the risk of sudden death in addition to other known clinical, electrocardiographic, hemodynamic risk factors. This would also guide both the instigation of complementary investigations and appropriate therapy in such high risk group of patients.
Correspondence to: Dr. D. Benchimol, HBpital Supported by grants from de Caisse Nationale
Cardiologique, 33604 Pessac. France. d’Assurance Maladie des Travailleurs
Salariks.
308
Key words: Sudden cardiac death; Coronary artery disease; Fibrinogen; Physiological fibrinolysis
Protein C; Hemostatic
factors;
antianginal therapy (associating, at the most, beta-blockers, long-acting nitrates and calcium antagonists); or those having a routine follow-up after a previous myocardial infarction (at least two months ago). Patients had a blood sample for biological tests (lipidic and hemostatic profiles) taken before the coronary angiography which they all underwent to assess their coronary artery disease.
Introduction More than 80% of individuals over 40 years of age who die suddenly are found to have coronary artery disease [ll. It is thus important to assess the risk of sudden death in this group of patients. Numerous studies have attempted to establish risk factors for sudden cardiac death, whether clinical [2,3], electrocardiographic [4,5], electrophysiological [6] or hemodynamic [5,7]. However, little has been reported on possible biological risk factors. The main mechanism of sudden death is tachycardia and/or ventricular fibrillation in the course of an acute ischemic episode. The present prospective study was therefore designed to single out risk factors among various markers of atherogenesis and thrombosis, for sudden death in patients with coronary artery disease but without evidence of heart failure.
Exclusion criteria Patients with unstable angina not stabilized for the last 2 weeks by the oral treatment described, myocardial infarction in the 2 months prior to the admission to avoid the biological perturbations induced in these circumstances; non-coronary heart disease (except for hypertension), clinical heart failure, acute progressive or chronic systemic or neoplastic pathology which might affect the follow-up and those with voluminous hematoma or a recent arterial puncture which might affect the results of the laboratory tests, were excluded.
Patients Inclusion criteria We included patients with stable chest pain or stabilized angina for at least 2 weeks by oral 384 preselected
patients
)c
\ 13 excluded
371 -
coronary
2 refused the arteriography
/
9 incomplete
-
\
anglography
2 died before angiogram and blood sample
- blood sample K 51
320
coronary
artery
normal coronary
disease
50 /
\
alive 1 died of acute
study population
blood sample
hemorrhagic
panCrE!atitiS
Fig. 1. Selection
of the patients.
anglography ‘1 died
Thus, between October 1984 and February 1987, according to the selection criteria, 384 consecutive patients admitted in our cardiology department for coronary angiography were preselected for this study. Among these 384 preselected patients, 320 with an angiographically proven coronary artery disease and a complete blood sample were included (Fig. 1). Fifty-one patients were excluded from the study group because of a strictly normal coronary angiogram. Thirteen patients were excluded: 2 patients refused the coronary angiography, 9 had an incomplete blood sample and 2 died before the inclusion (one before the blood sample and the angiogram, and the other at the first injection in the left main coronary artery which was severely narrowed) and led to an incomplete angiogram and ventriculography. Methods Clinical
and electrocardiographic
data
Routine clinical details were recorded, but the main objective was to identify biological risk factors. We recorded age, previous myocardial infarction (known history of acute myocardial infarction with electrocardiographic and cardiac enzymes modifications), previous and treated hypertension or recent hypertension (systolic blood pressure > 160 mmHg and diastolic blood pressure > 95 mmHg), estimated tobacco consumption in packets/year, ventricular arrhythmia at grade 2 of the Lown classification [8] evaluated from the routine tracings in those patients with abnormal tracings or with a history of arrhythmia. Coronary angiogram analysis
and ventriculography
The severity of the coronary artery lesions was evaluated from two scores: Jenkins’ score [9] which assesses lesions in main proximal segments, and mean atherosclerotic score [lo] over 15 segments of the American Heart Association classification, which evaluated the extent and multiplicity of lesions rather than the direct coronary occlusion. Even the slightest angiographic abnor-
malities were taken as evidence of coronary lesions. The ejection fraction was evaluated from left ventriculograms in the 30” right anterior oblique projection by a quantitative computer-assisted method [ll] based upon the Stanford method. Laboratory
findings
Four main groups of variables were measured. (1) Lipid profile including triglycerides, total cholesterol [ 12-141, high density lipoprotein and low density lipoprotein fractions [1.5,16], and apolipoproteins Al and B [17]. (2) Markers of platelet release: platelet factor 4 (by radioimmunoassay, Abbott kit) [ 181, /3-thromboglobulin (by radioimmunoassay, Amersham kit) [19]. (3) Coagulation parameters: fibrinogen (van Claus chronometric method) [20], fibrinopeptide A (radioimmunoassay kit Mallinckrodt) [21], antithrombin III activity (by chromogenic method, Kabi kit S 2238) [22], antithrombin III antigen (Laurel1 technique) [23], factor VIII antigen (by Laurel1 technique) [23] and factor VIII coagulant (FVIIIc by one-step method with factor VIII deficient plasma), protein C antigen (by Laurel1 technique) [24]. (4) Markers of physiological fibrinolysis: plasminogen and rY2-antiplasmin (chromogenic method, Kabi kit S 2251) [251, euglobulin clot lysis time (Kluft technique) [26], tissue plasminogen activator (Verheijen Technique) [27], plasminogen activator inhibitor (Eriksson method) [28]. Fasting blood samples were taken between X.30 and 10.00 after an overnight rest and 12 h abstinence from smoking. After puncture of a cubital vein, blood was collected without anticoagulant for lipid determinations. Mallinckrodt anticoagulant was used for the fibrinopeptide A assay [21]. and Thrombotect anticoagulant for platelet factor 4 and /3-thromboglobulin assays [l&19]. For hemostatic parameters, nine parts of blood were mixed with one part of 3.8%~ sodium citrate. For the euglobulin clot lysis time. measurement of tissue plasminogen activator and plasminogen activator inhibitor, blood samples were taken from the contralateral cubital vein before and after 10 min occlusion using a cuff inflated to the mid-sys-
310
tolic-diastolic pressure. Plasminogen activator inhibitor was determined before venous occlusion using the technique described by Eriksson et al. [28]. All blood samples were taken on the day before cardiac catheterization. Lipids were measured immediately after blood sampling. To assay the various coagulation factors, fibrinolysis and platelet release, plasma was separated by centrifugation at 3300 rpm for 20 min. Samples were frozen immediately to - 7o”C, and stored at this temperature until assay. Follow-up The follow-up period was 2 years or until death, and was obtained for all patients by means of a questionnaire sent prospectively each year to the patient or family, the family physician, and the cardiologist. For the patients who died in hospital, the relevant clinical findings at the time were also analyzed. The follow-up was obtained for all the 320 patients and also the 51 patients with normal angiograms. Sudden cardiac death was defined as a death occurring less than one hour after onset of symptoms with no other obvious non-cardiac origin [291. Treatment during the follow-up period was not standardized, although it was taken into account in the analysis of the results. Statistical analysis The values of all variables for the patients who died were compared with those of the surviving patients with coronary artery disease. Quantitative variables were compared using the MannWhitney U-test, and the qualitative variables were compared using the chi-square test with the Yates correction for small samples. Differences were considered significant if P < 0.05. A multivariate analysis by logistic regression with respect to sudden death was carried out on all the variables for which the P value in univariate analysis was < 0.10.
Results Over the study period there were 12 sudden deaths (10 men and 2 women), 7 at home (4 in
the first year and 3 in the second year), 5 in hospital (3 in the first year and 2 in the second year). Another patient died of acute hemorrhagic pancreatitis after coronary by-pass surgery one month after hemodynamic investigations and was excluded from the analysis. There were no sudden deaths among the patients with normal angiograms. One patient without coronary artery disease with a dilated myocardiopathy died during the second year with irreversible heart failure. Two groups of patients were thus constituted in the final study population: 307 surviving patients with coronary artery disease, and 12 patients who died suddenly during the observation period. Clinical and angiographic findings The characteristics of the deceased and surviving patients are given in Table 1. The mean age of the patients who died was 62 f 9 versus 57 _t 9 years for the surviving patients with coronary artery disease (NS). Sixty-six percent of the deceased patients had a history of myocardial infarction versus 42% for the surviving patients (NS). Thirty-three percent of the deceased patients had hypertension, but this percentage was not significant from that observed in the other group (47% for the coronary survivors). Tobacco consumption was the same for the patients who died (29 _t 9 packets/year) and for the coronary survivors (29 +_8 packets/year) (NS). There was a higher percentage of arrhythmia in the patients who died: 17% versus 3% for the coronary survivors (NS). The ejection fraction was also lower in the patients who died: 49 -t 16% versus 61 ) 16% for the coronary survivors (P < 0.02). The two scores used to evaluate the extent of coronary artery disease showed no significant differences between the deceased (Jenkins’ score 11.5 + 6.3 and mean atherosclerotic score 209 f 115) and the coronary survivors (Jenkins’ score 8.6 _t 5.6 and mean atherosclerotic score 181 + 109) although Jenkins’ score tended to be higher (P = 0.07). The treatment details for the different groups are summarized in Table 1. There was no differ-
311 TABLE
1
Clinical,
angiographic
and therapeutic
characteristics
of the patients
Sudden death n = 12 Age (years) Previous AM1 Hypertension Tobacco consumption (pack/year) Ventricular arrhythmia (Lawn > 2) Jenkins‘ score Mean atherosclerotic score Ejection fraction Beta-blockers Calcium antagonists Long acting nitrates Anticoagulants Antiplatelets Lipid lowering drugs (fibrates) CAD = coronary
artery
disease:
AMI = acute
patients
62 + 9 h6ri 33%’ 29 f 9 174 11.5 + 6.3 209 _+I15 49 f I6 33% H3Q, 83% 753 8%, 0% myocardial
P
NS NS NS NS NS NS NS c tt.o1! NS
X.6 + 5.h 1x1 _+ 109 hl + lb 37/; 765
NS
73’;
NS
-1zri 73”; 25c;
NS NS NS
infarction.
findings
The lipid profiles are listed in Table 2. There were no significant differences between the groups except for high density lipoprotein cholesterol fraction. This was higher in the patients who died suddenly (1.33 + 0.27 mmol/l) than in the coronary survivors (1.19 _+0.32 mmol/l) (P < 0.05). Similar values for the platelet markers, platelet factor 4 and /3-thromboglobulin (Table 3) were observed in the two groups. Coagulation parameters for the three groups are listed in Table 4. Levels of fibrinogen were higher in the deceased group (3.93 + 0.75 g/l TABLE
with CAD
57 * 9 -12c; -t7? 79 * x 3’;
ence in treatment between the deceased patients and the survivors. Laboratory
Living patients ,r = 307
versus 3.51 ) 0.84 g/l for coronary survivors, P < 0.05). Protein C levels were lower in the patients who died (89.4 f 39% versus 110.3 + 39% in the survivors, P = 0.06). There were no significant differences between the groups in any of the other parameters measured. The data for fibrinolysis are listed in Table 5. The only difference noted was a lower value of tissue plasminogen activator before and after occlusion (90 + 186.7 and 2629 + 2882 W/ml, respectively) in the patients who died, although this was not statistically significant. Multivariate
analysis
Eight variables with a P value < 0.10 in the univariate analysis were selected for the multi-
Z
Lipid profiles
of the patients Sudden death I, = 12
Cholesterol (mmol/lJ Triglycerides tmmol/l) High density lipoproteins (mmol/lJ Low density lipoproteins tmmol/lJ Apolipoprotein Al tmg/lOO ml) Apolipoprotein B (mg/lOO ml)
6.73 1.80 1.33 4.63 142.08 145.33
+ 1.34 * 0.99 + 0.27 * 0.27 + 23.18 + 31.49
patients
Living patients II = 307
P
6.53 1.81 1.19 4.58 131.13 143.20
NS NS < o.t~5 NS NS NS
* I.13 + I.08 * 0.32 f 1.10 * 27.21 + 27.62
312 TABLE
3
Markers
of the platelet
Discussion release
Platelet factor 4 (ng/mlI (ng/mll p-Thromboglobulin
Sudden death patients II = 12
Living patients n = 307
P
14.6+15 50.6 f 34
23.5 +3Y 61.78 + 54
NS NS
variate analysis. The results are given in Table 6. Three variables were correlated with sudden death: a lower ejection fraction (P < 0.008, with an odds ratio of 3.1 for 10% decrease, 95% confidence limits 0.87 to 0.98), a lower value of protein C (I’ < 0.01, with an odds ratio of 1.5 for 10% decrease, 95% confidence limits 0.94 to 0.99) and an older age (P < 0.03, with an odds ratio of 3.1 for 10 years, 95% confidence limits 1 to 1.24). In summary, we found a higher incidence of the usual clinical signs associated with sudden cardiac death in the deceased patients, namely: age, alteration in left ventricular function and ventricular arrhythmia (although not significant for this last one). The patients who died suddenly appeared to have a higher prothrombotic potential, characterized by a higher level of fibrinogen and a lower level of protein C than in the other coronary patients.
TABLE
We employed the usual definition of sudden cardiac death, namely an unexpected witnessed death occurring less than one hour after onset of symptoms with no other non-cardiac origin [29]. We thus isolated a group of 12 patients who died suddenly over the 2 year period, who were compared with the other living coronary patients. Autopsies were not carried out in every case, although patients with acute or chronic noncardiac pathology were expressly excluded, which considerably reduced the risk of diagnostic error. Half the deaths occurred in hospital, and for these patients we could establish the cause of death beyond doubt from the clinical findings and the results of complementary investigations. For the other cases the cause of death was based on information from the family and witnesses. In each case a witness was present, and the presumed cause of death was confirmed by the family physician who was in contact with our department. Epidemiological studies [2,31 have shown that 58% of the mortality of coronary origin in the age range of patients in our study are sudden deaths. The exclusion criteria employed in this study (absence of intercurrent acute or chronic pathology) was probably responsible for the absence of nonsudden and non-cardiac deaths in our population. We observed a 3.8% incidence of sudden deaths in the group with coronary artery disease, which
4
Coagulation
parameters Sudden death n = 12
Fibrinogen (g/l) Fibrinopeptide A (mg/ml) Antithrombin III activity (‘Xl Antithrombin III antigen (‘?%I Factor VIII antigen (o/cl Factor VIII coagulant (%I Protein C antigen (%I
3.93 5.63 100.8 97.5 175.6 146 89.4
patients
+ 0.75 * 11.47 It 11 * 12.7 rt 112 * 72 + 39
Living patients n = 307
P
3.51 7 96.5 98.9 139.6 135.4 110.27
< 0.05 NS NS NS NS NS = 0.06
+ 0.84 + 15 f 11.7 * 15 ? 62 +56 + 39
313 TABLE
5
Factors
of physiological
fibrinolysis Sudden death n= 12
Ptasminogen (%J cy2-antiplasmin (%J Euglobulin clot lysis time before venous occlusion (min) Euglobulin clot lysis time after venous occlusion (mint Tissue plasminogen activator before venous occlusion (ILJ/ml) Tissue plasminogen activator after venous occlusion W/ml) Plasminogen activator inhibitor (W/ml)
patients
P
with CAD
103 + 95.5 f
18 14
108 + 98.2 +
22 14
NS NS
263
+
42
265
f
47
NS
92.5 t
67
143.7 f
98
NS
90.7 +
186.7
164.7k
489
NS
2629
+2882
10.4 +
3220 8.2
+4378
11.5 i
is lower than reported in other studies. Gradman et al. [30] reported 8.1% of sudden deaths in a study of 295 patients over a 16 month period. However, there was marked alteration in left ventricular function in these patients (ejection fraction < 40%), and the percentage of patients with coronary artery disease did not exceed 62%. Rockman et al. [31] reported 21% of sudden deaths in a group of 238 patients with coronary artery disease. This high percentage can be accounted for by the long follow-up period (8 years), and by the selection of patients with frank signs of heart failure (stages III-IV New York Heart Association). A third series of patients followedup over 26 years as part of the Framingham
TABLE
Living patients n = 307
NS 8.7
NS
Study, including 90% of patients with coronary artery disease, gave an overall incidence of sudden cardiac death of 13% [32]. The incidence of sudden death thus appears to depend on the characteristics of the particular population of patients, including such factors as clinical evidence of heart failure, ischemic nature of the cardiac pathology and duration of follow-up (cumulative incidence with increasing duration of follow-up). The low incidence of sudden death in our population, which is, in fact, a limitation of the power of the study, can be accounted for by the relatively short follow-up period (24 months) and the inclusion criteria employed, i.e. patients with no clinical signs of heart failure. Our confidence
6
Multivariate
correlates
of sudden
death
Covariables
Age Euglobulin clot lysis time after venous occlusion High density lipoprotein Protein C Jenkins’ score Apolipoprotein Al Fibrinogen Ejection fraction
p coefficient 0.1 13 0.009 0.763 _ .0.037 0.084 0.033 0.498 0.078
Chi square
P
4.46 2.5 1 0.16 5.89 3.38 1.81 1.29 7.14
0.03 0.11 0.68 0.01 0.06 0.18 0.26 0.008
314
interval 2% is quite comparable to those of Gradman [30] and Rockrnan [31], 3.2% and 5.2%, respectively. The incidence of sudden death we observed can thus be considered to be representative of that of a population of patients with coronary artery disease but without evidence of heart failure, over a two-year period. The clinical and angiographic findings were comparable to those of other studies [5,71. We observed between the two groups of patients (deceased and survivors) a significant difference in ejection fraction. This indicates the prognostic significance of left ventricular function, even in the absence of clinical signs of heart failure. We also observed a higher incidence of ventricular arrhythmia in the patients who died. The absence of a significant difference for this variable might be, at least, partly attributed to the fact that it was based on simple ECG tracings carried out while the patient was in hospital or from telemetered or Holter recordings, but not from systematic Holter investigations. In any case this was not the main objective of the study. Our patients who died suddenly appear, therefore, to be representative of the population generally recognized to be at risk. This is also supported by the results of the multivariate analysis concerning ejection fraction and age which are usual risk factors for sudden death in coronary patients [30-321. We assumed that the alteration in the other parameters could be extrapolated to the general population of patients with stable coronary artery disease without heart failure at risk from sudden death. Alterations in lipid metabolism are now recognized to be implicated in the development of coronary artery disease [33,34]. We therefore looked for a relationship between an abnormality in lipid metabolism and the occurrence of sudden cardiac death. Most parameters were quite comparable between the two groups. This was not unexpected as the severity of the atheroma was also comparable. However, the significant elevation of high density lipoprotein cholesterol (or rather an absence of a reduction) in the patients who died compared to the other patients, associated with the similarities in low density lipoprotein and total cholesterol, was more surprising.
This result seems in agreement with the epidemiological findings of Shestov et al. [35] who studied the influence of lipid risk factors on mortality in the Soviet Union. These authors observed an increased annual total mortality in patients with coronary artery disease who have high levels of high density lipoprotein, but this difference noted in univariate analysis appears namely linked to non-coronary deaths and is reversed in multivariate analysis with respect to coronary mortality. Our univariate result is not confirmed by the univariate analysis which suggests a dependent relationship with another factor, perhaps with age which appears in the multivariate analysis. The heterogeneous nature of the high density lipoprotein fraction whose role is not yet completely understood [361, especially its impact on coagulation processes, may also take part in this result. The study of markers of fibrinolysis, platelet activation and coagulation might provide a better indication of the risk of sudden cardiac death. Various authors have attempted to define a more precise etiology of sudden cardiac death. There appears to be evidence for acute ischemia, sometimes in the absence of frank clinical signs. Acute ischemia is thought to be responsible for the fatal arrhythmia or that occurring on reperfusion of the myocardium with release of metabolites [l] and probably in connection with other triggering factors. Post-mortem examination of patients who died suddenly support this notion, with evidence of an acute coronary lesion in 95% of patients, and a recent intracoronary clot in 74% of patients [37-391. Thrombosis formation commences by platelet aggregation and terminates in fibrinolysis, but we did not detect significant differences in markers of platelet function and fibrinolysis between the two groups of patients. The relatively small size of the group of deceased patients may partly account for this negative result. Systemic measurements are possibly not completely representative of local perturbations of hemostatic factors. However, we observed a decrease in tissue plasminogen activator in the deceased group. It has been reported that there is an increased risk of a second infarction in patients with reduced levels of tissue plasminogen activa-
315
tor [40]. Although the reduction we observed was not statistically significant, it may reflect an enhanced risk of myocardial ischemia. However, we did observe significant alterations in fibrinogen and protein C in the patients who died suddenly. The elevation of fibrinogen is in agreement with the Northwick Park Heart Study [41], where a relationship was noted between elevated fibrinogen and mortality from coronary artery disease. The relationship between this factor, which is involved in coagulation, platelet aggregation and blood viscosity, and the possible development of atheromatous plaques and coronary mortality is perhaps not surprising, although the mechanism is unclear. We observed an elevation of fibrinogen in the deceased patients compared with the other patients. This factor which constitutes a risk factor for progression of coronary artery disease [42], might also enhance the eventuality of sudden death. The disturbances of blood rheology and the enhanced coagulation potential induced by elevated fibrinogen [43] could thus favor thrombus formation in vessels already narrowed by atheroma. This relation, not confirmed in multivariate results, may be dependent on other variables, especially age at recruitment as it is noted in Northwick Park Heart Study [41]. In our study, age appears in multivariate analysis as fibrinogen disappears. We
PROTHROMBOTIC
observed a decrease in protein C in the patients who died suddenly. This protein is involved in coagulation by inactivating prothrombotic factors Va and VIIIa [44]. It is also thought to play a role in profibrinolysis by inhibiting plasminogen activator inhibitor [451. Thus protein C has an antithrombotic effect and a reduced level enhances the risk of venous or arterial thrombosis. Gonzalez et al. [46] have described a link between this factor and myocardial ischemia, and they observed a fall in protein and in the acute phase of the infarction in patients who died within 10 days. A study carried out in our hospital [lo] also indicated that reduced protein C levels might be risk factor for myocardial infarction after coronary by-pass surgery. The relationship between sudden death and protein C levels could be a result of fatal arrhythmia concomitant with acute ischemia caused by a thrombotic process favored by the decreased levels of protein C. It was of interest that there were no differences in Jenkins’ score or mean atherosclerotic score between the two groups of patients with coronary artery disease (deceased or surviving). It would thus seem that a latent prothrombotic potential rather than any difference in severity of atheroma may be related to the risk of sudden cardiac death in this population. This potential, in the absence of progressive thrombotic processes, was suggested by
OTHER IWEMIC
FACTORS
FACTORS \
: SPASM...
J METABOLIC
ISCHEMIA/REPERFUSION x PREMATURE
VENTRICULAR
+
AUTONOMIC
DISTURBANCES
F
t PATHOLOGICAL
TOXIC EFFECT
t
OF DRUGS
MYOCAROIUM
BEATS
NERVOUS SYSTEM
DISTURBANCES
t VENTRICULAR
TACHYCAROIA
and/or VENTRICULAR
FIBRILLATION
t SUDDEN DEATH Fig. 2. Role of prothrombotic
factors
in sudden
death
among
patients
with coronary
artery
disease.
316
the absence of alterations in fibrinopeptide A and plasma levels of the molecules as platelet factor 4, P-thromboglobulin released from platelet (Y granules. The concentrations of these molecules are generally raised during acute clot formation [47,48], but are not altered during the prethrombotic period [49]. This is in line with the results of Meade who found raised levels of fibrinopeptide A in blood samples taken by cardiac puncture just after sudden death [50]. Moreover, the lack of difference in these parameters between the patients who died and the surviving patients indicated that we had selected patients who were not in an acute thrombotic phase, at the time of the inclusion, and that the results were not biased by the presence of an infraclinical thrombosis. We suggest that this latent prothrombotic potential favors acute myocardial ischemia (Fig. 21, and so, represents an associated risk factor for sudden cardiac death in patients with stable coronary artery disease and without heart failure. Anoxia along with other etiological factors such as ventricular arrhythmia, hyperadrenergia, metabolic disorders or toxic actions of drug may trigger the fatal ventricular arrhythmia in patients with pathological myocardium and altered ventricular function [1,51,52]. In fact, disorders of ventricular rhythm are thought to be responsible for the majority of deaths occurring within one hour of onset of symptoms [531.
Conclusion We identified some hemostatic factors, especially protein C, which might indicate an increased risk of sudden death in some groups of patients with stabilized coronary artery disease and without heart failure. These factors were those involved in the regulation of coagulation, reflecting a raised latent prothrombotic potential. Determination of such factors in further prospective studies, particularly the prothrombotic profile, would support the clinical evidence and the results of complementary investigations in assessment of the risk of sudden cardiac death in groups of patients with coronary artery disease. This would also guide instigation of further, pos-
sibly invasive, investigations which could define the vital prognosis of these patients and indicate an appropriate therapeutic strategy.
References 1 Myerburg R, Kessler K, Bassett A, Castellanos A. A biological approach to sudden cardiac death: structure, function, and cause. Am J Cardiol 1989;15:1512-1516. 2 Kannel WB, Doyle JT, McNamara PM, Quickenton P. Gordon T. Precursors of sudden coronary death: factors related to the incidence of sudden death. Circulation 1975;51:606-613. 3 Doyle JT, Kannel WB, McNamara PM, Quickenton P, Gordon T. Factors related to suddenness of death from coronary disease: combined Albany-Framingham studies. Am J Cardiol 1976;37:1073-1078. 4 Follansbee WP. Michelson EL, Morganroth J. Non-sustained ventricular tachycardia in ambulatory patients. Characteristics and association with sudden death. Ann Intern Med 1980;92:741-747. 5 Bigger JT Jr, Fleiss JL, Kleiger R et al. The relationships among ventricular arrhythmias, left ventricular dysfunction, and mortality in the 2 years after myocardial infarction. Circulation 1984:54:250-258. 6 Wilber DJ. Gram H, Finkelstein D et al. Out of hospital cardiac arrest: use of electrophysiologic testing in the prediction of long-term outcome. N Engl J Med 1988; 318:19-24. 7 Schulze RA, Strauss HW, Pitt B. Sudden death in the year following myocardial infarction: relationship of ventricular premature contraction in the late phase and left ventricular ejection fraction. Am J Med 1977;62:192-199. to sudden death from 8 Lown B. Wolf M. Approaches coronary heart disease. Circulation 1971;44:130-142. 9 Jenkins PJ, Harper RW, Nestel PJ. Severity of coronary atherosclerosis related to lipoprotein concentration. Br Med J 1978;2:388-391. C et al. Risk factor for 10 Oysel N, Bonnet J, Vergnes myocardial infarction during coronary artery bypass graft surgery. Eur Heart J 1989;10:806-815. 11 Colle JP, Rahal S, Ohayon J, et al. Global left ventricular function and regional wall motion in pure mitral stenosis. Clin Cardiol 1984;7:573-580. 12 Frinder P. Determination of glucose in blood using glucose oxydase with an alternative oxygen acceptor. Ann Clin Biochem 1969;6:24. S, Richmond W, Fu PC. 13 Albain CC, Poowl S, Chant Enzymatic determination of total serum cholesterol. Clin Chem 1974:20:470-477. 14 Wahlefeld AW. Bermeyer HU. Methoden der enzymatischen Analyse. 3rd ed. Tome II. Weiheim: Verlagchemie 1974. 15 Burnstein M, Schelnick HR, Marfin R. Rapid method for the isolation of lipoprotein from human serum by precipitation with polyions. J Lipid Res 1970;11:583-595.
317 16 Lopes-Virella MF, Stone P, Ellis P, Colwell JA. Cholesterol determination in high density lipoproteins separated by three different methods. Clin Chem 1977;23:882-884. 17 Henck CC, Schlierf G. Nephelometty of apoprotein B in human serum. Clin Chem 1979;25:221-226. 18 Handin RI, Mc Donough M, Lesch M. Elevation of platelet factor IV in acute myocardial infarction: measurement by radioimmunoassay. J Lab Clin Med 1978;91:340-349. 19 Ludlam CA, Moore S. Bolton AE, Pepper DS, Cash JD. The release of a human platelet specific protein measured by radioimmunoassay. Thromb Res 1975;6:543-548. 20 Claus A. Gerinnungsphysiologische Schnellmethode zur Bestimmung des Fibrinogens. Acta Haematol 1957;17: 237-246. 21 Hockum C. Radioimmunoassay of fibrinopeptide A. Clinical applications. Thromb Res 1976;8:225-236. 23 Abildgaard U, Lie M, Odegard OR. Antithrombin (heparin cofactor) assay with new chromogenic substrates (S-2238 and chromozym TH). Thromb Res 1977;11:549-553. 23 Laurel1 CB. Quantitative estimation of proteins by electrophoresis in agarose gel containing antibodies. Ann Biochem 1966:15:57-65. 24 Gonzalez R, Alberta I, Sala N. Vicente V. Congenital deficiency of protein C. Response to danazol and DDAVP. Thromb Haemost 1983:53:320-322. 25 Frieherger P, Knos M. Gustavsson S. Aurell L, Cleason G. Methods for the determination of plasmin, antiplasmin and plasminogen by means of substrate S-2251. Haemostasis lY78;7:138-145. 26 Kluft C. Brakman P, Veldhyyzn-Stolk EC. Screening of fibrinolytic activity in the plasma euglobulin fractions on the fibrin plate. In: Davidson JF, Samama M, Desnoyerse PC, eds. Progress in chemical fibrinolysis and thrombosis. New York: Raven Press 19762157-65. 27 Verheijen JH, Mullaart E, Chang GTG, Kluft C. Wijngaards G. A simple sensitive spectrophotometric assay for extrinsic (tissue type) plasminogen activator applicable to measurement in plasma. Thromb Haemost 1982;48:266269. 28 Ericksson E, Ranby M, Grysander E, Risberg B. Determination of plasminogen activator inhibitor in plasma using tPA and chromogenic single point poly-D-lysine stimulated assay. Thromb Res 1988;50:91-101. 29 Goldstein S. The necessity of a uniform definition of sudden coronary death: witnessed death within 1 hour of the onset of acute symptoms. Am Heart J 1982;102:156159. 30 Gradman A, Deedwania P, Cody R et al. Predictors of total mortality in sudden death in mild to moderate heart failure. J Am Co11 Cardiol 1989;14:564-570. 31 Rockman HA, Juneau C. Chatterjee K, Rouleau JL. Long-term predictors of sudden and low output death in chronic congestive heart failure secondary to coronary artery disease. Am J Cardiol 1989;64:1344-1348. 32 Schatzkin A, Cupples LA, Heeren T, Morelock S, Muatel M, Kannel WB. The epidemiology of sudden unexpected
33
34
35
36
37
38
39
40
41
42
43 44
45
46
47
48
49
death: risk factors for men and women in the Framingham Heart Study. Am Heart J 1984;107:1300-1306. Arntzenius AC, Kromhout D, Barth JD et al. Diet. lipoproteins, and the progression of atherosclerosis. The Leiden Interventional Trial. N Engl J Med lY85;312:805811. Raichlen JS. Healy B. Achuff SC, Pearson TA. Importance of risk factors in the angiographic progression of coronary artery disease. Am J Cardiol 1986:57:66-77. Shestov DB, Deev AD, Zhukovsky GS. Thron MD, Davis CE, Klimov AN. Coronary heart disease risk factors and mortality in the USSR lipid research clinics follow-up study. Atherosclerosis Rev 1988;37:261-275. Gwynne JT. High-density lipoprotein cholesterol levels as a marker of reverse cholesterol transport. Am J Cardiol 1989;64:1OG-176. Davies MJ, Thomas AC. Thrombosis and acute coronary artery lesion in sudden cardiac ischemic death. N Engl J Med 1984;310:1137-1140. Davies MJ, Bland JM, Hangartner JRW. Angelini A. Thomas AC. Factors influencing the presence or absence of acute coronary artery thrombi in sudden ischaemic death. Eur Heart J 1989;10:203-2fl6. Baroldi G. Falzi G. Mariani F. Sudden coronary death. A postmortem study in 208 cases compared to 97 “control” subjects. Am Heart J 1979;98:20-31. Gram J, Jespersen J. A selective depression of tissue plasminogen activator ftPA) activity in euglobulins characterize a risk group among survivors of acute myocardial infarction. Thromb Haemost lY87:57: 137- 13Y. Meade TW, Chkrabarti R. Haines AP et al. Haemostatic function and cardiovascular death: early results of a prospective study. Lancet 1980;17: lOSO- 1053. Kannel WB. Wolf PA, Castelli WP, D’Agostino RB. Fihrinogen and risk of cardiovascular disease. The Framingham Study. JAMA 1987:258:1183-l 186. Merill EW. Rheology of blood. Physiol Rev 1969:49:X63887. Verhaeghe R, Verstraete M. Schetz F. Vanhove P. Suy R, Vermylen J. Clinical trial of urinase and anticoagulants as a method of treatment for advanced limb ischemia. Eur J Clin Pharmacol lY79;16:165-170. Comp PC, Esmon CT. Generation of fibrinolytic activity of actived protein C into dogs. J Clin Invest IM1:68:12211228. Gonzalez R, Vincente R. Alegre A, Pagon P. Alberta I. Protein C and antithrombin III in acute myocardial infarction. Thromb Res 1986;43:681-685. Eisenberg PR, Sherman LA. Schectman K, Perez J. Sobel B, Jaffe AS. Fibrinopeptide A: a marker of acute coronary thrombosis. Circulation 1985;5:912-918. Theroux P. Latour JG, Leger-Gauthier C, De Lara J. Fibrinopeptide A and platelet factor levels in unstable angina pectoris. Circulation 1987;1:156-162. Van Hulsteign H, Kolff J. Briet E, Van der Laarse A, Bertina R. Fibrinopeptide A and beta thromboglobulin in
318 patients with angina pectoris and acute myocardial infarction. Am Heart J 1984;107:39-45. 50 Meade TW, Howarth DJ, Striling Y, Welch TP, Crompton MR. Fibrinopeptide A and sudden cardiac death. Lancet 1984;15:607-609. 51 Goldstein S. Toward a new understanding of the mechanism and prevention of sudden death in coronary heart disease. Circulation 1990;82:284-288.
52 Verheugt FWA, Brugada P. Sudden death after acute myocardial infarction: the forgotten thrombotic view. Am J Cardiol 1991;67:1130-1134. 53 Hinkle LE, Thaler HT. Clinical classification of cardiac deaths. Circulation 1982:65:457-464.
CARD10
0167.5273/92/$05.00
01399
Biological risk factors for sudden death in patients with coronary artery disease and without heart failure Philippe Lauribe ‘, Daniel Benchimol ‘, Jean-Frangois Dartigues ‘?,Sobhi Dada ‘, Hklkne Benchimol ‘, Fransoise Drouillet ‘, Jacques Bonnet ‘A and Henri Bricaud 2 ’ Senkr de Cardiologie, H&ital Nerd, Marseille. France: ’ Chique M6dicale Cardiologique. Hspital Cirrdiologique. Pessac. France: .’ Dt!partement d’lnformatique MPdicale et de Statistiques. UniL,ersite’de Bordeaux II. INSERM Unitti 3.U.l.BordealLv. France; ’ INSERM Unitr’ X de Cardiologie. Pessac, France (Received
30 November
1990; revision
accepted
10 October
lY91)
Lauribe P, Benchimol D, Dartigues J-F, Dada S, Benchimol H, Drouillet F. Bonnet J, Bricaud H. Biological risk factors for sudden death in patients with coronary artery disease and without heart failure. Int J Cardiol 1992;34:307-318. In a study of biological risk factors for sudden death in patients with coronary artery disease, 320 patients were, prospectively, recruited and followed-up over two years. None of the patients had heart failure or recent myocardial infarction. The following variables were recorded: previous acute myocardial infarction, hypertension, smoking habits, ventricular arrhythmia; the angiographic variables included: left ventricular ejection fraction, Jenkins’ and mean atherosclerotic scores; lipid profile: cholesterol, triglycerides, high density lipoprotein cholesterol, low density lipoprotein cholesterol, apolipoproteins Al and B; hemostatic profile: fibrinogen, fibrinopeptide A, antithrombin III, factor VIII antigen, factor VIII coagulant, protein C, plasminogen, a2-antiplasmin, euglobulin clot lysis time and tissue plasminogen activator before and after venous occclusion, tissue plasminogen activator inhibitor, platelet factor 4, @thromboglobulin. During the follow-up period, 12 of the patients died suddenly. In these patients, ejection fraction was lower: 49 _t 16% versus 61 f 14% for the other patients (P < 0.02), fibrinogen higher: 3.9 j, 0.8 g/l versus 3.5 + 0.8 for the living patients (P < 0.05) and protein C lower: 89 + 39% versus 111 ? 39% (P = 0.06) for the other patients. In multivariate analysis: lower ejection fraction (P < O.OOS), older age (P < 0.03) and lower protein C (P < 0.011 were correlated with sudden death. Among the patients with coronary artery disease, the raised fibrinogen and the decreased protein C appeared to be risk factors for sudden cardiac death. These alterations reflected a prothrombotic state which might increase the ischemic risk, due to an acute thrombosis, leading to the fatal ventricular arrhythmia. Determination of these hemostatic variables might be a useful adjunct for assessment of the vital prognosis of patients with coronary artery disease, especially the risk of sudden death in addition to other known clinical, electrocardiographic, hemodynamic risk factors. This would also guide both the instigation of complementary investigations and appropriate therapy in such high risk group of patients.
Correspondence to: Dr. D. Benchimol, HBpital Supported by grants from de Caisse Nationale
Cardiologique, 33604 Pessac. France. d’Assurance Maladie des Travailleurs
Salariks.
308
Key words: Sudden cardiac death; Coronary artery disease; Fibrinogen; Physiological fibrinolysis
Protein C; Hemostatic
factors;
antianginal therapy (associating, at the most, beta-blockers, long-acting nitrates and calcium antagonists); or those having a routine follow-up after a previous myocardial infarction (at least two months ago). Patients had a blood sample for biological tests (lipidic and hemostatic profiles) taken before the coronary angiography which they all underwent to assess their coronary artery disease.
Introduction More than 80% of individuals over 40 years of age who die suddenly are found to have coronary artery disease [ll. It is thus important to assess the risk of sudden death in this group of patients. Numerous studies have attempted to establish risk factors for sudden cardiac death, whether clinical [2,3], electrocardiographic [4,5], electrophysiological [6] or hemodynamic [5,7]. However, little has been reported on possible biological risk factors. The main mechanism of sudden death is tachycardia and/or ventricular fibrillation in the course of an acute ischemic episode. The present prospective study was therefore designed to single out risk factors among various markers of atherogenesis and thrombosis, for sudden death in patients with coronary artery disease but without evidence of heart failure.
Exclusion criteria Patients with unstable angina not stabilized for the last 2 weeks by the oral treatment described, myocardial infarction in the 2 months prior to the admission to avoid the biological perturbations induced in these circumstances; non-coronary heart disease (except for hypertension), clinical heart failure, acute progressive or chronic systemic or neoplastic pathology which might affect the follow-up and those with voluminous hematoma or a recent arterial puncture which might affect the results of the laboratory tests, were excluded.
Patients Inclusion criteria We included patients with stable chest pain or stabilized angina for at least 2 weeks by oral 384 preselected
patients
)c
\ 13 excluded
371 -
coronary
2 refused the arteriography
/
9 incomplete
-
\
anglography
2 died before angiogram and blood sample
- blood sample K 51
320
coronary
artery
normal coronary
disease
50 /
\
alive 1 died of acute
study population
blood sample
hemorrhagic
panCrE!atitiS
Fig. 1. Selection
of the patients.
anglography ‘1 died
Thus, between October 1984 and February 1987, according to the selection criteria, 384 consecutive patients admitted in our cardiology department for coronary angiography were preselected for this study. Among these 384 preselected patients, 320 with an angiographically proven coronary artery disease and a complete blood sample were included (Fig. 1). Fifty-one patients were excluded from the study group because of a strictly normal coronary angiogram. Thirteen patients were excluded: 2 patients refused the coronary angiography, 9 had an incomplete blood sample and 2 died before the inclusion (one before the blood sample and the angiogram, and the other at the first injection in the left main coronary artery which was severely narrowed) and led to an incomplete angiogram and ventriculography. Methods Clinical
and electrocardiographic
data
Routine clinical details were recorded, but the main objective was to identify biological risk factors. We recorded age, previous myocardial infarction (known history of acute myocardial infarction with electrocardiographic and cardiac enzymes modifications), previous and treated hypertension or recent hypertension (systolic blood pressure > 160 mmHg and diastolic blood pressure > 95 mmHg), estimated tobacco consumption in packets/year, ventricular arrhythmia at grade 2 of the Lown classification [8] evaluated from the routine tracings in those patients with abnormal tracings or with a history of arrhythmia. Coronary angiogram analysis
and ventriculography
The severity of the coronary artery lesions was evaluated from two scores: Jenkins’ score [9] which assesses lesions in main proximal segments, and mean atherosclerotic score [lo] over 15 segments of the American Heart Association classification, which evaluated the extent and multiplicity of lesions rather than the direct coronary occlusion. Even the slightest angiographic abnor-
malities were taken as evidence of coronary lesions. The ejection fraction was evaluated from left ventriculograms in the 30” right anterior oblique projection by a quantitative computer-assisted method [ll] based upon the Stanford method. Laboratory
findings
Four main groups of variables were measured. (1) Lipid profile including triglycerides, total cholesterol [ 12-141, high density lipoprotein and low density lipoprotein fractions [1.5,16], and apolipoproteins Al and B [17]. (2) Markers of platelet release: platelet factor 4 (by radioimmunoassay, Abbott kit) [ 181, /3-thromboglobulin (by radioimmunoassay, Amersham kit) [19]. (3) Coagulation parameters: fibrinogen (van Claus chronometric method) [20], fibrinopeptide A (radioimmunoassay kit Mallinckrodt) [21], antithrombin III activity (by chromogenic method, Kabi kit S 2238) [22], antithrombin III antigen (Laurel1 technique) [23], factor VIII antigen (by Laurel1 technique) [23] and factor VIII coagulant (FVIIIc by one-step method with factor VIII deficient plasma), protein C antigen (by Laurel1 technique) [24]. (4) Markers of physiological fibrinolysis: plasminogen and rY2-antiplasmin (chromogenic method, Kabi kit S 2251) [251, euglobulin clot lysis time (Kluft technique) [26], tissue plasminogen activator (Verheijen Technique) [27], plasminogen activator inhibitor (Eriksson method) [28]. Fasting blood samples were taken between X.30 and 10.00 after an overnight rest and 12 h abstinence from smoking. After puncture of a cubital vein, blood was collected without anticoagulant for lipid determinations. Mallinckrodt anticoagulant was used for the fibrinopeptide A assay [21]. and Thrombotect anticoagulant for platelet factor 4 and /3-thromboglobulin assays [l&19]. For hemostatic parameters, nine parts of blood were mixed with one part of 3.8%~ sodium citrate. For the euglobulin clot lysis time. measurement of tissue plasminogen activator and plasminogen activator inhibitor, blood samples were taken from the contralateral cubital vein before and after 10 min occlusion using a cuff inflated to the mid-sys-
310
tolic-diastolic pressure. Plasminogen activator inhibitor was determined before venous occlusion using the technique described by Eriksson et al. [28]. All blood samples were taken on the day before cardiac catheterization. Lipids were measured immediately after blood sampling. To assay the various coagulation factors, fibrinolysis and platelet release, plasma was separated by centrifugation at 3300 rpm for 20 min. Samples were frozen immediately to - 7o”C, and stored at this temperature until assay. Follow-up The follow-up period was 2 years or until death, and was obtained for all patients by means of a questionnaire sent prospectively each year to the patient or family, the family physician, and the cardiologist. For the patients who died in hospital, the relevant clinical findings at the time were also analyzed. The follow-up was obtained for all the 320 patients and also the 51 patients with normal angiograms. Sudden cardiac death was defined as a death occurring less than one hour after onset of symptoms with no other obvious non-cardiac origin [291. Treatment during the follow-up period was not standardized, although it was taken into account in the analysis of the results. Statistical analysis The values of all variables for the patients who died were compared with those of the surviving patients with coronary artery disease. Quantitative variables were compared using the MannWhitney U-test, and the qualitative variables were compared using the chi-square test with the Yates correction for small samples. Differences were considered significant if P < 0.05. A multivariate analysis by logistic regression with respect to sudden death was carried out on all the variables for which the P value in univariate analysis was < 0.10.
Results Over the study period there were 12 sudden deaths (10 men and 2 women), 7 at home (4 in
the first year and 3 in the second year), 5 in hospital (3 in the first year and 2 in the second year). Another patient died of acute hemorrhagic pancreatitis after coronary by-pass surgery one month after hemodynamic investigations and was excluded from the analysis. There were no sudden deaths among the patients with normal angiograms. One patient without coronary artery disease with a dilated myocardiopathy died during the second year with irreversible heart failure. Two groups of patients were thus constituted in the final study population: 307 surviving patients with coronary artery disease, and 12 patients who died suddenly during the observation period. Clinical and angiographic findings The characteristics of the deceased and surviving patients are given in Table 1. The mean age of the patients who died was 62 f 9 versus 57 _t 9 years for the surviving patients with coronary artery disease (NS). Sixty-six percent of the deceased patients had a history of myocardial infarction versus 42% for the surviving patients (NS). Thirty-three percent of the deceased patients had hypertension, but this percentage was not significant from that observed in the other group (47% for the coronary survivors). Tobacco consumption was the same for the patients who died (29 _t 9 packets/year) and for the coronary survivors (29 +_8 packets/year) (NS). There was a higher percentage of arrhythmia in the patients who died: 17% versus 3% for the coronary survivors (NS). The ejection fraction was also lower in the patients who died: 49 -t 16% versus 61 ) 16% for the coronary survivors (P < 0.02). The two scores used to evaluate the extent of coronary artery disease showed no significant differences between the deceased (Jenkins’ score 11.5 + 6.3 and mean atherosclerotic score 209 f 115) and the coronary survivors (Jenkins’ score 8.6 _t 5.6 and mean atherosclerotic score 181 + 109) although Jenkins’ score tended to be higher (P = 0.07). The treatment details for the different groups are summarized in Table 1. There was no differ-
311 TABLE
1
Clinical,
angiographic
and therapeutic
characteristics
of the patients
Sudden death n = 12 Age (years) Previous AM1 Hypertension Tobacco consumption (pack/year) Ventricular arrhythmia (Lawn > 2) Jenkins‘ score Mean atherosclerotic score Ejection fraction Beta-blockers Calcium antagonists Long acting nitrates Anticoagulants Antiplatelets Lipid lowering drugs (fibrates) CAD = coronary
artery
disease:
AMI = acute
patients
62 + 9 h6ri 33%’ 29 f 9 174 11.5 + 6.3 209 _+I15 49 f I6 33% H3Q, 83% 753 8%, 0% myocardial
P
NS NS NS NS NS NS NS c tt.o1! NS
X.6 + 5.h 1x1 _+ 109 hl + lb 37/; 765
NS
73’;
NS
-1zri 73”; 25c;
NS NS NS
infarction.
findings
The lipid profiles are listed in Table 2. There were no significant differences between the groups except for high density lipoprotein cholesterol fraction. This was higher in the patients who died suddenly (1.33 + 0.27 mmol/l) than in the coronary survivors (1.19 _+0.32 mmol/l) (P < 0.05). Similar values for the platelet markers, platelet factor 4 and /3-thromboglobulin (Table 3) were observed in the two groups. Coagulation parameters for the three groups are listed in Table 4. Levels of fibrinogen were higher in the deceased group (3.93 + 0.75 g/l TABLE
with CAD
57 * 9 -12c; -t7? 79 * x 3’;
ence in treatment between the deceased patients and the survivors. Laboratory
Living patients ,r = 307
versus 3.51 ) 0.84 g/l for coronary survivors, P < 0.05). Protein C levels were lower in the patients who died (89.4 f 39% versus 110.3 + 39% in the survivors, P = 0.06). There were no significant differences between the groups in any of the other parameters measured. The data for fibrinolysis are listed in Table 5. The only difference noted was a lower value of tissue plasminogen activator before and after occlusion (90 + 186.7 and 2629 + 2882 W/ml, respectively) in the patients who died, although this was not statistically significant. Multivariate
analysis
Eight variables with a P value < 0.10 in the univariate analysis were selected for the multi-
Z
Lipid profiles
of the patients Sudden death I, = 12
Cholesterol (mmol/lJ Triglycerides tmmol/l) High density lipoproteins (mmol/lJ Low density lipoproteins tmmol/lJ Apolipoprotein Al tmg/lOO ml) Apolipoprotein B (mg/lOO ml)
6.73 1.80 1.33 4.63 142.08 145.33
+ 1.34 * 0.99 + 0.27 * 0.27 + 23.18 + 31.49
patients
Living patients II = 307
P
6.53 1.81 1.19 4.58 131.13 143.20
NS NS < o.t~5 NS NS NS
* I.13 + I.08 * 0.32 f 1.10 * 27.21 + 27.62
312 TABLE
3
Markers
of the platelet
Discussion release
Platelet factor 4 (ng/mlI (ng/mll p-Thromboglobulin
Sudden death patients II = 12
Living patients n = 307
P
14.6+15 50.6 f 34
23.5 +3Y 61.78 + 54
NS NS
variate analysis. The results are given in Table 6. Three variables were correlated with sudden death: a lower ejection fraction (P < 0.008, with an odds ratio of 3.1 for 10% decrease, 95% confidence limits 0.87 to 0.98), a lower value of protein C (I’ < 0.01, with an odds ratio of 1.5 for 10% decrease, 95% confidence limits 0.94 to 0.99) and an older age (P < 0.03, with an odds ratio of 3.1 for 10 years, 95% confidence limits 1 to 1.24). In summary, we found a higher incidence of the usual clinical signs associated with sudden cardiac death in the deceased patients, namely: age, alteration in left ventricular function and ventricular arrhythmia (although not significant for this last one). The patients who died suddenly appeared to have a higher prothrombotic potential, characterized by a higher level of fibrinogen and a lower level of protein C than in the other coronary patients.
TABLE
We employed the usual definition of sudden cardiac death, namely an unexpected witnessed death occurring less than one hour after onset of symptoms with no other non-cardiac origin [29]. We thus isolated a group of 12 patients who died suddenly over the 2 year period, who were compared with the other living coronary patients. Autopsies were not carried out in every case, although patients with acute or chronic noncardiac pathology were expressly excluded, which considerably reduced the risk of diagnostic error. Half the deaths occurred in hospital, and for these patients we could establish the cause of death beyond doubt from the clinical findings and the results of complementary investigations. For the other cases the cause of death was based on information from the family and witnesses. In each case a witness was present, and the presumed cause of death was confirmed by the family physician who was in contact with our department. Epidemiological studies [2,31 have shown that 58% of the mortality of coronary origin in the age range of patients in our study are sudden deaths. The exclusion criteria employed in this study (absence of intercurrent acute or chronic pathology) was probably responsible for the absence of nonsudden and non-cardiac deaths in our population. We observed a 3.8% incidence of sudden deaths in the group with coronary artery disease, which
4
Coagulation
parameters Sudden death n = 12
Fibrinogen (g/l) Fibrinopeptide A (mg/ml) Antithrombin III activity (‘Xl Antithrombin III antigen (‘?%I Factor VIII antigen (o/cl Factor VIII coagulant (%I Protein C antigen (%I
3.93 5.63 100.8 97.5 175.6 146 89.4
patients
+ 0.75 * 11.47 It 11 * 12.7 rt 112 * 72 + 39
Living patients n = 307
P
3.51 7 96.5 98.9 139.6 135.4 110.27
< 0.05 NS NS NS NS NS = 0.06
+ 0.84 + 15 f 11.7 * 15 ? 62 +56 + 39
313 TABLE
5
Factors
of physiological
fibrinolysis Sudden death n= 12
Ptasminogen (%J cy2-antiplasmin (%J Euglobulin clot lysis time before venous occlusion (min) Euglobulin clot lysis time after venous occlusion (mint Tissue plasminogen activator before venous occlusion (ILJ/ml) Tissue plasminogen activator after venous occlusion W/ml) Plasminogen activator inhibitor (W/ml)
patients
P
with CAD
103 + 95.5 f
18 14
108 + 98.2 +
22 14
NS NS
263
+
42
265
f
47
NS
92.5 t
67
143.7 f
98
NS
90.7 +
186.7
164.7k
489
NS
2629
+2882
10.4 +
3220 8.2
+4378
11.5 i
is lower than reported in other studies. Gradman et al. [30] reported 8.1% of sudden deaths in a study of 295 patients over a 16 month period. However, there was marked alteration in left ventricular function in these patients (ejection fraction < 40%), and the percentage of patients with coronary artery disease did not exceed 62%. Rockman et al. [31] reported 21% of sudden deaths in a group of 238 patients with coronary artery disease. This high percentage can be accounted for by the long follow-up period (8 years), and by the selection of patients with frank signs of heart failure (stages III-IV New York Heart Association). A third series of patients followedup over 26 years as part of the Framingham
TABLE
Living patients n = 307
NS 8.7
NS
Study, including 90% of patients with coronary artery disease, gave an overall incidence of sudden cardiac death of 13% [32]. The incidence of sudden death thus appears to depend on the characteristics of the particular population of patients, including such factors as clinical evidence of heart failure, ischemic nature of the cardiac pathology and duration of follow-up (cumulative incidence with increasing duration of follow-up). The low incidence of sudden death in our population, which is, in fact, a limitation of the power of the study, can be accounted for by the relatively short follow-up period (24 months) and the inclusion criteria employed, i.e. patients with no clinical signs of heart failure. Our confidence
6
Multivariate
correlates
of sudden
death
Covariables
Age Euglobulin clot lysis time after venous occlusion High density lipoprotein Protein C Jenkins’ score Apolipoprotein Al Fibrinogen Ejection fraction
p coefficient 0.1 13 0.009 0.763 _ .0.037 0.084 0.033 0.498 0.078
Chi square
P
4.46 2.5 1 0.16 5.89 3.38 1.81 1.29 7.14
0.03 0.11 0.68 0.01 0.06 0.18 0.26 0.008
314
interval 2% is quite comparable to those of Gradman [30] and Rockrnan [31], 3.2% and 5.2%, respectively. The incidence of sudden death we observed can thus be considered to be representative of that of a population of patients with coronary artery disease but without evidence of heart failure, over a two-year period. The clinical and angiographic findings were comparable to those of other studies [5,71. We observed between the two groups of patients (deceased and survivors) a significant difference in ejection fraction. This indicates the prognostic significance of left ventricular function, even in the absence of clinical signs of heart failure. We also observed a higher incidence of ventricular arrhythmia in the patients who died. The absence of a significant difference for this variable might be, at least, partly attributed to the fact that it was based on simple ECG tracings carried out while the patient was in hospital or from telemetered or Holter recordings, but not from systematic Holter investigations. In any case this was not the main objective of the study. Our patients who died suddenly appear, therefore, to be representative of the population generally recognized to be at risk. This is also supported by the results of the multivariate analysis concerning ejection fraction and age which are usual risk factors for sudden death in coronary patients [30-321. We assumed that the alteration in the other parameters could be extrapolated to the general population of patients with stable coronary artery disease without heart failure at risk from sudden death. Alterations in lipid metabolism are now recognized to be implicated in the development of coronary artery disease [33,34]. We therefore looked for a relationship between an abnormality in lipid metabolism and the occurrence of sudden cardiac death. Most parameters were quite comparable between the two groups. This was not unexpected as the severity of the atheroma was also comparable. However, the significant elevation of high density lipoprotein cholesterol (or rather an absence of a reduction) in the patients who died compared to the other patients, associated with the similarities in low density lipoprotein and total cholesterol, was more surprising.
This result seems in agreement with the epidemiological findings of Shestov et al. [35] who studied the influence of lipid risk factors on mortality in the Soviet Union. These authors observed an increased annual total mortality in patients with coronary artery disease who have high levels of high density lipoprotein, but this difference noted in univariate analysis appears namely linked to non-coronary deaths and is reversed in multivariate analysis with respect to coronary mortality. Our univariate result is not confirmed by the univariate analysis which suggests a dependent relationship with another factor, perhaps with age which appears in the multivariate analysis. The heterogeneous nature of the high density lipoprotein fraction whose role is not yet completely understood [361, especially its impact on coagulation processes, may also take part in this result. The study of markers of fibrinolysis, platelet activation and coagulation might provide a better indication of the risk of sudden cardiac death. Various authors have attempted to define a more precise etiology of sudden cardiac death. There appears to be evidence for acute ischemia, sometimes in the absence of frank clinical signs. Acute ischemia is thought to be responsible for the fatal arrhythmia or that occurring on reperfusion of the myocardium with release of metabolites [l] and probably in connection with other triggering factors. Post-mortem examination of patients who died suddenly support this notion, with evidence of an acute coronary lesion in 95% of patients, and a recent intracoronary clot in 74% of patients [37-391. Thrombosis formation commences by platelet aggregation and terminates in fibrinolysis, but we did not detect significant differences in markers of platelet function and fibrinolysis between the two groups of patients. The relatively small size of the group of deceased patients may partly account for this negative result. Systemic measurements are possibly not completely representative of local perturbations of hemostatic factors. However, we observed a decrease in tissue plasminogen activator in the deceased group. It has been reported that there is an increased risk of a second infarction in patients with reduced levels of tissue plasminogen activa-
315
tor [40]. Although the reduction we observed was not statistically significant, it may reflect an enhanced risk of myocardial ischemia. However, we did observe significant alterations in fibrinogen and protein C in the patients who died suddenly. The elevation of fibrinogen is in agreement with the Northwick Park Heart Study [41], where a relationship was noted between elevated fibrinogen and mortality from coronary artery disease. The relationship between this factor, which is involved in coagulation, platelet aggregation and blood viscosity, and the possible development of atheromatous plaques and coronary mortality is perhaps not surprising, although the mechanism is unclear. We observed an elevation of fibrinogen in the deceased patients compared with the other patients. This factor which constitutes a risk factor for progression of coronary artery disease [42], might also enhance the eventuality of sudden death. The disturbances of blood rheology and the enhanced coagulation potential induced by elevated fibrinogen [43] could thus favor thrombus formation in vessels already narrowed by atheroma. This relation, not confirmed in multivariate results, may be dependent on other variables, especially age at recruitment as it is noted in Northwick Park Heart Study [41]. In our study, age appears in multivariate analysis as fibrinogen disappears. We
PROTHROMBOTIC
observed a decrease in protein C in the patients who died suddenly. This protein is involved in coagulation by inactivating prothrombotic factors Va and VIIIa [44]. It is also thought to play a role in profibrinolysis by inhibiting plasminogen activator inhibitor [451. Thus protein C has an antithrombotic effect and a reduced level enhances the risk of venous or arterial thrombosis. Gonzalez et al. [46] have described a link between this factor and myocardial ischemia, and they observed a fall in protein and in the acute phase of the infarction in patients who died within 10 days. A study carried out in our hospital [lo] also indicated that reduced protein C levels might be risk factor for myocardial infarction after coronary by-pass surgery. The relationship between sudden death and protein C levels could be a result of fatal arrhythmia concomitant with acute ischemia caused by a thrombotic process favored by the decreased levels of protein C. It was of interest that there were no differences in Jenkins’ score or mean atherosclerotic score between the two groups of patients with coronary artery disease (deceased or surviving). It would thus seem that a latent prothrombotic potential rather than any difference in severity of atheroma may be related to the risk of sudden cardiac death in this population. This potential, in the absence of progressive thrombotic processes, was suggested by
OTHER IWEMIC
FACTORS
FACTORS \
: SPASM...
J METABOLIC
ISCHEMIA/REPERFUSION x PREMATURE
VENTRICULAR
+
AUTONOMIC
DISTURBANCES
F
t PATHOLOGICAL
TOXIC EFFECT
t
OF DRUGS
MYOCAROIUM
BEATS
NERVOUS SYSTEM
DISTURBANCES
t VENTRICULAR
TACHYCAROIA
and/or VENTRICULAR
FIBRILLATION
t SUDDEN DEATH Fig. 2. Role of prothrombotic
factors
in sudden
death
among
patients
with coronary
artery
disease.
316
the absence of alterations in fibrinopeptide A and plasma levels of the molecules as platelet factor 4, P-thromboglobulin released from platelet (Y granules. The concentrations of these molecules are generally raised during acute clot formation [47,48], but are not altered during the prethrombotic period [49]. This is in line with the results of Meade who found raised levels of fibrinopeptide A in blood samples taken by cardiac puncture just after sudden death [50]. Moreover, the lack of difference in these parameters between the patients who died and the surviving patients indicated that we had selected patients who were not in an acute thrombotic phase, at the time of the inclusion, and that the results were not biased by the presence of an infraclinical thrombosis. We suggest that this latent prothrombotic potential favors acute myocardial ischemia (Fig. 21, and so, represents an associated risk factor for sudden cardiac death in patients with stable coronary artery disease and without heart failure. Anoxia along with other etiological factors such as ventricular arrhythmia, hyperadrenergia, metabolic disorders or toxic actions of drug may trigger the fatal ventricular arrhythmia in patients with pathological myocardium and altered ventricular function [1,51,52]. In fact, disorders of ventricular rhythm are thought to be responsible for the majority of deaths occurring within one hour of onset of symptoms [531.
Conclusion We identified some hemostatic factors, especially protein C, which might indicate an increased risk of sudden death in some groups of patients with stabilized coronary artery disease and without heart failure. These factors were those involved in the regulation of coagulation, reflecting a raised latent prothrombotic potential. Determination of such factors in further prospective studies, particularly the prothrombotic profile, would support the clinical evidence and the results of complementary investigations in assessment of the risk of sudden cardiac death in groups of patients with coronary artery disease. This would also guide instigation of further, pos-
sibly invasive, investigations which could define the vital prognosis of these patients and indicate an appropriate therapeutic strategy.
References 1 Myerburg R, Kessler K, Bassett A, Castellanos A. A biological approach to sudden cardiac death: structure, function, and cause. Am J Cardiol 1989;15:1512-1516. 2 Kannel WB, Doyle JT, McNamara PM, Quickenton P. Gordon T. Precursors of sudden coronary death: factors related to the incidence of sudden death. Circulation 1975;51:606-613. 3 Doyle JT, Kannel WB, McNamara PM, Quickenton P, Gordon T. Factors related to suddenness of death from coronary disease: combined Albany-Framingham studies. Am J Cardiol 1976;37:1073-1078. 4 Follansbee WP. Michelson EL, Morganroth J. Non-sustained ventricular tachycardia in ambulatory patients. Characteristics and association with sudden death. Ann Intern Med 1980;92:741-747. 5 Bigger JT Jr, Fleiss JL, Kleiger R et al. The relationships among ventricular arrhythmias, left ventricular dysfunction, and mortality in the 2 years after myocardial infarction. Circulation 1984:54:250-258. 6 Wilber DJ. Gram H, Finkelstein D et al. Out of hospital cardiac arrest: use of electrophysiologic testing in the prediction of long-term outcome. N Engl J Med 1988; 318:19-24. 7 Schulze RA, Strauss HW, Pitt B. Sudden death in the year following myocardial infarction: relationship of ventricular premature contraction in the late phase and left ventricular ejection fraction. Am J Med 1977;62:192-199. to sudden death from 8 Lown B. Wolf M. Approaches coronary heart disease. Circulation 1971;44:130-142. 9 Jenkins PJ, Harper RW, Nestel PJ. Severity of coronary atherosclerosis related to lipoprotein concentration. Br Med J 1978;2:388-391. C et al. Risk factor for 10 Oysel N, Bonnet J, Vergnes myocardial infarction during coronary artery bypass graft surgery. Eur Heart J 1989;10:806-815. 11 Colle JP, Rahal S, Ohayon J, et al. Global left ventricular function and regional wall motion in pure mitral stenosis. Clin Cardiol 1984;7:573-580. 12 Frinder P. Determination of glucose in blood using glucose oxydase with an alternative oxygen acceptor. Ann Clin Biochem 1969;6:24. S, Richmond W, Fu PC. 13 Albain CC, Poowl S, Chant Enzymatic determination of total serum cholesterol. Clin Chem 1974:20:470-477. 14 Wahlefeld AW. Bermeyer HU. Methoden der enzymatischen Analyse. 3rd ed. Tome II. Weiheim: Verlagchemie 1974. 15 Burnstein M, Schelnick HR, Marfin R. Rapid method for the isolation of lipoprotein from human serum by precipitation with polyions. J Lipid Res 1970;11:583-595.
317 16 Lopes-Virella MF, Stone P, Ellis P, Colwell JA. Cholesterol determination in high density lipoproteins separated by three different methods. Clin Chem 1977;23:882-884. 17 Henck CC, Schlierf G. Nephelometty of apoprotein B in human serum. Clin Chem 1979;25:221-226. 18 Handin RI, Mc Donough M, Lesch M. Elevation of platelet factor IV in acute myocardial infarction: measurement by radioimmunoassay. J Lab Clin Med 1978;91:340-349. 19 Ludlam CA, Moore S. Bolton AE, Pepper DS, Cash JD. The release of a human platelet specific protein measured by radioimmunoassay. Thromb Res 1975;6:543-548. 20 Claus A. Gerinnungsphysiologische Schnellmethode zur Bestimmung des Fibrinogens. Acta Haematol 1957;17: 237-246. 21 Hockum C. Radioimmunoassay of fibrinopeptide A. Clinical applications. Thromb Res 1976;8:225-236. 23 Abildgaard U, Lie M, Odegard OR. Antithrombin (heparin cofactor) assay with new chromogenic substrates (S-2238 and chromozym TH). Thromb Res 1977;11:549-553. 23 Laurel1 CB. Quantitative estimation of proteins by electrophoresis in agarose gel containing antibodies. Ann Biochem 1966:15:57-65. 24 Gonzalez R, Alberta I, Sala N. Vicente V. Congenital deficiency of protein C. Response to danazol and DDAVP. Thromb Haemost 1983:53:320-322. 25 Frieherger P, Knos M. Gustavsson S. Aurell L, Cleason G. Methods for the determination of plasmin, antiplasmin and plasminogen by means of substrate S-2251. Haemostasis lY78;7:138-145. 26 Kluft C. Brakman P, Veldhyyzn-Stolk EC. Screening of fibrinolytic activity in the plasma euglobulin fractions on the fibrin plate. In: Davidson JF, Samama M, Desnoyerse PC, eds. Progress in chemical fibrinolysis and thrombosis. New York: Raven Press 19762157-65. 27 Verheijen JH, Mullaart E, Chang GTG, Kluft C. Wijngaards G. A simple sensitive spectrophotometric assay for extrinsic (tissue type) plasminogen activator applicable to measurement in plasma. Thromb Haemost 1982;48:266269. 28 Ericksson E, Ranby M, Grysander E, Risberg B. Determination of plasminogen activator inhibitor in plasma using tPA and chromogenic single point poly-D-lysine stimulated assay. Thromb Res 1988;50:91-101. 29 Goldstein S. The necessity of a uniform definition of sudden coronary death: witnessed death within 1 hour of the onset of acute symptoms. Am Heart J 1982;102:156159. 30 Gradman A, Deedwania P, Cody R et al. Predictors of total mortality in sudden death in mild to moderate heart failure. J Am Co11 Cardiol 1989;14:564-570. 31 Rockman HA, Juneau C. Chatterjee K, Rouleau JL. Long-term predictors of sudden and low output death in chronic congestive heart failure secondary to coronary artery disease. Am J Cardiol 1989;64:1344-1348. 32 Schatzkin A, Cupples LA, Heeren T, Morelock S, Muatel M, Kannel WB. The epidemiology of sudden unexpected
33
34
35
36
37
38
39
40
41
42
43 44
45
46
47
48
49
death: risk factors for men and women in the Framingham Heart Study. Am Heart J 1984;107:1300-1306. Arntzenius AC, Kromhout D, Barth JD et al. Diet. lipoproteins, and the progression of atherosclerosis. The Leiden Interventional Trial. N Engl J Med lY85;312:805811. Raichlen JS. Healy B. Achuff SC, Pearson TA. Importance of risk factors in the angiographic progression of coronary artery disease. Am J Cardiol 1986:57:66-77. Shestov DB, Deev AD, Zhukovsky GS. Thron MD, Davis CE, Klimov AN. Coronary heart disease risk factors and mortality in the USSR lipid research clinics follow-up study. Atherosclerosis Rev 1988;37:261-275. Gwynne JT. High-density lipoprotein cholesterol levels as a marker of reverse cholesterol transport. Am J Cardiol 1989;64:1OG-176. Davies MJ, Thomas AC. Thrombosis and acute coronary artery lesion in sudden cardiac ischemic death. N Engl J Med 1984;310:1137-1140. Davies MJ, Bland JM, Hangartner JRW. Angelini A. Thomas AC. Factors influencing the presence or absence of acute coronary artery thrombi in sudden ischaemic death. Eur Heart J 1989;10:203-2fl6. Baroldi G. Falzi G. Mariani F. Sudden coronary death. A postmortem study in 208 cases compared to 97 “control” subjects. Am Heart J 1979;98:20-31. Gram J, Jespersen J. A selective depression of tissue plasminogen activator ftPA) activity in euglobulins characterize a risk group among survivors of acute myocardial infarction. Thromb Haemost lY87:57: 137- 13Y. Meade TW, Chkrabarti R. Haines AP et al. Haemostatic function and cardiovascular death: early results of a prospective study. Lancet 1980;17: lOSO- 1053. Kannel WB. Wolf PA, Castelli WP, D’Agostino RB. Fihrinogen and risk of cardiovascular disease. The Framingham Study. JAMA 1987:258:1183-l 186. Merill EW. Rheology of blood. Physiol Rev 1969:49:X63887. Verhaeghe R, Verstraete M. Schetz F. Vanhove P. Suy R, Vermylen J. Clinical trial of urinase and anticoagulants as a method of treatment for advanced limb ischemia. Eur J Clin Pharmacol lY79;16:165-170. Comp PC, Esmon CT. Generation of fibrinolytic activity of actived protein C into dogs. J Clin Invest IM1:68:12211228. Gonzalez R, Vincente R. Alegre A, Pagon P. Alberta I. Protein C and antithrombin III in acute myocardial infarction. Thromb Res 1986;43:681-685. Eisenberg PR, Sherman LA. Schectman K, Perez J. Sobel B, Jaffe AS. Fibrinopeptide A: a marker of acute coronary thrombosis. Circulation 1985;5:912-918. Theroux P. Latour JG, Leger-Gauthier C, De Lara J. Fibrinopeptide A and platelet factor levels in unstable angina pectoris. Circulation 1987;1:156-162. Van Hulsteign H, Kolff J. Briet E, Van der Laarse A, Bertina R. Fibrinopeptide A and beta thromboglobulin in
318 patients with angina pectoris and acute myocardial infarction. Am Heart J 1984;107:39-45. 50 Meade TW, Howarth DJ, Striling Y, Welch TP, Crompton MR. Fibrinopeptide A and sudden cardiac death. Lancet 1984;15:607-609. 51 Goldstein S. Toward a new understanding of the mechanism and prevention of sudden death in coronary heart disease. Circulation 1990;82:284-288.
52 Verheugt FWA, Brugada P. Sudden death after acute myocardial infarction: the forgotten thrombotic view. Am J Cardiol 1991;67:1130-1134. 53 Hinkle LE, Thaler HT. Clinical classification of cardiac deaths. Circulation 1982:65:457-464.
