(Circulation. 1999;100:483-489.)
© 1999 American Heart Association, Inc.
Clinical Investigation and Reports |
From the Departments of Heart Disease (K.V., J.E.N.) and Clinical Biochemistry (M.F.), Haukeland University Hospital, Bergen, Norway.
Correspondence to Dr K. Vikenes, Department of Heart Disease, Haukeland University Hospital, N-5021 Bergen, Norway.
| Abstract |
|---|
|
|
|---|
Methods and ResultsOf 121 consecutive male patients (mean age 65, range 41 to 90 years) undergoing angiography, 96 had coronary artery stenosis and 25 had normal angiograms. Serotonin, blood platelet count, and conventional biochemical risk factors for CAD were determined in the morning the day before the angiography. High serotonin (cut-point 1000 nmol/L) was significantly associated with CAD with an odds ratio (OR) of 3.4 (95% confidence interval 1.2 to 9.8). The corresponding OR for smokers was 4.8 (1.9 to 12.2); hypercholesterolemia (>7 mmol/L), 2.9 (1.1 to 7.6); high platelet count (cut-point 325 109/L), 3.0 (1.0 to 9.5); and family history of heart disease, 2.3 (1.0 to 5.2). After adjustment with conventional risk factors, the OR for CAD was 3.8 (1.1 to 13.1), comparing high and low values of serotonin. The relation between serotonin and CAD was strengthened only when patients <70 years (n=82) were included in the analysis. In this age group, the occurrence of cardiac events during a mean of 3.7 years of follow-up was significantly associated with high serotonin values.
ConclusionsThe study suggests that serotonin is associated with coronary artery disease and occurrence of cardiac events, particularly in younger age groups. This association seems to persist after adjustment for conventional risk factors.
Key Words: risk factors atherosclerosis ischemia coronary disease platelets
| Introduction |
|---|
|
|
|---|
Serotonin release during coronary angioplasty and infusion of serotonin at coronary angiography have both contributed to clinically relevant vasoconstriction in patients with coronary artery disease.18 19 The effects were seen with plasma levels of serotonin equivalent to those resulting from local platelet activation and were attenuated by ketanserin. Until this study, cardiovascular complications have not been convincingly reduced by ketanserin in patients with various manifestations of arteriosclerotic disease, but a reduction of mortality has been suggested in nonrandomized studies.20 21 Although an effect on restenosis after coronary angioplasty has been shown,22 a larger study failed to confirm this.23 Ketanserin reduces the number of cases of preeclampsia and severe hypertension.24
Thus, serotonin clearly has important vascular actions and may also be involved in atherogenesis. In this study, we assessed serotonin as a clinical marker of coronary artery disease in patients with stable symptoms admitted to elective coronary angiography for chest pain.
| Methods |
|---|
|
|
|---|
Hypercholesterolemia was classified according to pretreatment values (26% of the study population were on cholesterol lowering medication), or cholesterol values >7.0 mmol/L in untreated patients. Hypertension was classified according to previously diagnosed hypertension and use of current antihypertensive medication.
Biochemical Measurements
Blood samples were drawn in the morning between 8 AM
and noon the day before angiography. Venous blood was collected in
Vacutainer tubes with ACD solution (22g Na citrate, 8 g citrate,
and 24.5 g dextrose per liter) 9:1.5 vol/vol. The samples
were centrifuged at 450g for 5 minutes and
platelet rich plasma (PRP) was pipetted off the supernatant.
Platelets were counted with a Coulter Counter, Model S-Plus III
(Coulter Electronics). Because all patients had a hematocrit within
reference values, no correction for hematocrite was done.
Total serotonin in PRP and serotonin content in platelets were measured. Four hundred fifty microliters of PRP were mixed with 150 ìL of 2.8 mol/L per chloric acid, and serotonin was measured by a HPLC method with a Zipax SCX column (DuPont) eluted with a 150 mmol/L sodium acetate (pH 5.2), 10 mL n-propanol/L, and fluorometry with excitation at 304 nm and detection at 340 nm as previously described.25
Precision for the serotonin determination was better than ±5% both within and between runs. Serotonin (5-hydroxytryptamine) was also assayed directly in serum prepared from whole blood using thrombin-coated vacuum tubes. Serotonin recovered from serum amounted to only 465±53 nmol/L compared with 901±113 nmol/L in PRP, and the correlation coefficient was only 0.586. Because serotonin binds to proteins during the coagulation process and reduces the amount of measurable serotonin in serum,25 we used platelet rich plasma throughout the experiment.
Cholesterol was determined in the hospital routine laboratory by a Technicon Chem 1 (Technicon Instruments) according to the instrument manual. 5-hydroxytryptamine (serotonin)-creatinine sulfate complex was purchased from Sigma Chemical Co (St. Louis, Mo). All other reagents were of the highest purity commercially available.
Angiography
Left ventricular angiography and selective
coronary angiography from the groin were performed according to
standard techniques in our laboratory.26 The angiograms
were assessed by 2 cardiologists who had no knowledge of the
serotonin values. Coronary artery stenosis
was confirmed in orthogonal views and a luminal reduction of 50% in
any of the main coronary arteries or major side branches was
considered significant. A few patients with minimal coronary
artery disease were included in the group with normal angiograms, but
in general the angiographic distinction between the 2 groups was
definite and without disagreement between the 2 observers. The extent
of cardiac disease was determined according to the number of main
vessels or major side branches with stenoses.
Long-Term Follow-Up
End points were cardiac events, defined as reoccurrence of
nonfatal myocardial infarction (ICD9, Code 410), cardiac death and
readmission for unstable angina (ICD9, Code 411), and based on a
systematic and detailed review of the hospital charts on deaths (via
the National Death Registry) and readmissions. All the patients were
living in the catchment area of our hospital, and follow-up was
complete for the whole study population.
Statistical Analysis
Data are presented as mean±SEM. Student's t
test was used when comparing means, and Yates corrected
2 or Fisher's exact test for
analysis of 2x2 tables. Univariate and
multivariate logistic regression analysis (SPSS
version 6.1 for Windows) were performed to assess the influence of
various variables on coronary artery disease
(normal/abnormal). Change in estimate method was used to assess
possible confounders. If OR of serotonin for CAD was
reduced by <10%, the variable was not included as a confounder.
However, age, smoking, hypertension, history of CAD,
hypercholesterolemia, and current use of
acetylsalicylic acid were always included. A
multivariate analysis with the Cox proportional
hazards model (SPSS version 6.1 for Windows) was performed to assess
the association of serotonin with cardiac events during
follow-up period. Inclusion of possible confounders was performed
according to the same principles as in the logistic regression
analysis. Kaplan-Meier method was used to evaluate differences
in event-free survival between patients with high/low values around the
mean (cut-point 1000 nmol/l) of serotonin. Differences were
considered significant with 2-sided values of P<0.05.
| Results |
|---|
|
|
|---|
|
Serotonin and Platelets
Total serotonin levels and the number of platelets
were significantly higher among patients who had coronary
artery disease compared with those with normal angiograms (Table 2
). The subgroup with previous myocardial
infarction did not have higher serotonin levels or
platelet counts than patients with coronary
stenosis alone. The univariate relation between
total serotonin and platelet count is shown in Figure 1
, r=0.55,
P<0.001.
|
|
Figure 2
shows mean values and 95% CI
for serotonin in patients with and without coronary
artery disease according to 3 age groups. The difference in
serotonin level was most striking <60 years of age
(P=0.001); in the groups 60 to 70 and >70 years the
difference was steadily reduced and became nonsignificant.
|
In the lowest to highest age groups of patients with coronary artery disease, the platelet counts (109/L PRP) were 351±18 (P=0.008 versus nonCAD group), 296±11, and 273±18, respectively. In nonCAD patients, there was no significant difference in total serotonin and platelet count among the age groups.
Serotonin and Conventional Risk Factors
Table 3
shows the unadjusted and
adjusted OR for CAD comparing individuals with serotonin
1000 nmol/L PRP with individuals with lower serotonin
values.
|
Serotonin remained significantly associated with CAD after adjustment for the conventional risk factors age, hypertension, diabetes, hypercholesterolemia, smoking, and family history of cardiac disease. When platelet count (cut-point 325 109 /L PRP) was included in the multivariate analysis, high serotonin was associated with a 3.7-fold (1.0 to 15.1, P=0.054) risk for CAD. The univariate OR of platelet count was 3.0 (1.0 to 9.5), P=0.059, and adjusted OR 1.0 (0.02 to 4.2), P=1.00, increasing to 2.0 (0.6 to 7.0, P=0.296) when serotonin was excluded from the multivariate analysis.
In patients <70 years (n=82), the association between
serotonin and coronary artery disease was
strengthened (Table 4
). If platelet
count was included in the multivariate
analysis, OR of serotonin for CAD rose from 15.3
(1.6 to 147.6) to 20.3 (1.6 to 249.8). There was no association between
platelet count and coronary artery disease in this
subgroup.
|
Serotonin and Extent of Coronary Artery
Disease
In Table 5
the CAD group is divided
according to single (26%) or multivessel disease (74%). Among the
patients in the lower quartile of serotonin values, the
highest percentage of patients were found in the group with no
coronary artery disease. In the same quartile, there was a
steady decreasing percentage of patients with single vessel disease
compared with patients with multivessel disease.
|
Long-Term Follow-Up
During the mean follow-up period of 44±15 (SD) months, 11 cardiac
events occurred and all events were in patients with CAD. Four patients
suffered a nonfatal acute myocardial infarction, 2 suffered cardiac
deaths, and 6 were admitted for unstable angina. Mean time for an event
to happen was 24 months, range 1 to 45 months.
In a multivariate Cox regression analysis, adjusting for conventional risk factors and use of acetylsalicylic acid, hazard ratio of serotonin (high/low values, cut-point 1000 nmol/L) was 2.6 (0.7 to 9.6), P=0.14. Age was significantly associated with occurrence of cardiac events (P=0.04), and hazard ratio of current smoker was 5.7 (1.3 to 26.2), P=0.02. In patients <70 years (n=82), hazard ratio of serotonin increased to 12.3 (1.2 to 122.4), P=0.03.
Figure 3
shows a Kaplan-Meier plot with
event-free curves for CAD-patients (< 70 years, n=82), suggesting a
better prognosis with regard to cardiac events for patients with low
serotonin (log rank test, P=0.046) during a mean
of 3.7 years of follow-up.
|
| Discussion |
|---|
|
|
|---|
Clinical studies have shown that blood platelet count and function are related to total and cardiovascular death during long-term follow-up.27 In the present study, there was a significant correlation between blood platelets and total serotonin levels. When blood platelets were adjusted for in our statistical analysis, the relationship between serotonin and coronary artery disease was somewhat attenuated. In contrast, the relationship between platelets and coronary artery stenosis demonstrated in the univariate analysis disappeared completely after adjustment with serotonin in the multivariate analysis. Current use of acetylsalicylic acid, which has important effects on platelet function and serotonin secretion,28 was also adjusted for but did not weaken the association of serotonin with coronary disease.
The important role that serotonin plays in initiating thrombosis and vasoconstriction has been extensively studied in an animal model.6 7 13 29 In the early experiments, it was shown that the specific serotonin inhibitor ketanserin protected against the cyclic flow variations at the site of coronary stenosis. The serotonin concentration at the stenosis site is elevated and causes platelet aggregation and vasospasm which are the main mediators of the reduced flow.7 9 Thromboxane A2 has been shown to act synergistically with serotonin to cause vasoconstriction, platelet aggregation, and reduction in blood flow.8 9 30 In the animal models, endothelial injury was created to induce cyclic flow variations. However, serotonin itself may induce endothelial injury17 and thereby lay the groundwork for thrombosis.
The effects of serotonin as a mitogen could be of importance for the initiation or progression of vascular disease. Serotonin has growth-promoting effects both on vascular smooth muscle cells and endothelial cells,16 the effects depends both on its concentration and duration of action. This may contribute both to the restenotic lesions seen after angioplasty procedures and to the proliferation that is found in the advancing native atherosclerotic lesions.15 16 Furthermore, in an experimental model the development of arteriosclerosis was preceded by an augmented vasoconstrictor response to serotonin.31
Clinically, there is firm evidence for important actions of serotonin, although cardiovascular complications have not yet been convincingly reduced by ketanserin. Serotonin release during coronary angioplasty and infusion of serotonin at coronary angiography both have contributed to clinically relevant vasoconstriction in patients with coronary artery disease.18 19 The effects were seen with plasma levels of serotonin equivalent to those resulting from local platelet activation, and were blocked by ketanserin. In nonrandomized studies, administration of ketanserin has been reported to decrease mortality among patients with angiographically demonstrated coronary stenosis,20 and ketanserin also reduced mortality in patients with intermittent claudication and no medication of potassium loosing diuretics.21 In a small placebo-controlled study, ketanserin has been shown to prevent early restenosis,22 but a larger study failed to show effect on minimal lumen loss 6 months after angioplasty and did not significantly improve clinical outcome.23 However, there was a trend toward fewer events in the ketanserin group. There is also experimental evidence suggesting that a combination of serotonin receptor antagonist and a thromboxane A2 inhibitor may be useful for attenuation of restenosis.14
Platelet count and platelet hyperreactivity have been associated with acute myocardial infarction and prognosis,27 32 33 and increased platelet serotonin has been detected in patients with clinically suspected coronary artery disease or myocardial infarction compared with controls.34 An experimental animal model also suggests that serotonin could be important for acute clinical syndromes showing that serotonin and thromboxane A2 receptor blockade protected against epinephrine-induced cyclic flow variation.35 In our study, serotonin adjusted for conventional risk factors was associated with acute cardiac events known to be precipitated by coronary thrombosis. Although the design of our cross-sectional study would neither observe transient rises in epinephrine and serotonin during acute chest pain,12 21 36 37 38 nor does it prove a causal relationship with coronary artery disease, the absolute levels during stable phases of the disease seem to be important for subsequent clinical exacerbation. Serotonin levels were not higher in patients with previous infarctions. There is no obvious explanation for this, but factors such as therapeutic or lifestyle changes after the infarction could play a role. Consumption of large platelets during the thrombotic process with subsequent decrease of serotonin levels39 is less likely to have occurred in our patients who had long-term symptomatic stability before blood sampling.
Age had an expected relationship with cardiac endpoints during follow-up. Whereas risk factors, such as cholesterol and homocysteine, normally have increasing blood levels with age,26 serotonin seems to be inversely related to age, having higher levels in younger age groups in which it also had a significant relationship with acute cardiac events. We have no clear explanation for this observation or for the reduction in serotonin with advancing age (also has been shown previously).40 However, because morbidity and mortality increase with age, and the finding that serotonin is most important in the younger age groups, the decline in serotonin with age may support the concept that it is not only a passive marker of coronary artery disease and cardiac events.
The stability of serotonin levels is an issue for the interpretation of our results. Only traces of serotonin are normally found in plasma where it is bound to proteins. The measurements of total serotonin in platelet rich plasma reflect both serotonin in platelets and the protein bound contents in the circulation.25 Serotonin levels vary within a relatively wide range during the day41 and may show a modest reduction with age.40 There is also a gender difference with slightly higher levels in women than in men.40 42 In our male study population, blood was sampled at the same time in the day in all subjects and thus some possible confounders were avoided. It has been shown that when sampled at the same time of the day, serotonin levels are stable over long periods of time.40
Factors associated with coronary artery disease and acute events are probably not exclusively either thrombogenic or atherogenic; there is evidence to support both effects for serotonin. High serotonin levels may also augment the effects of other platelet activation agonists39 43 and may thereby take part in the spontaneous aggregation and platelet hyperreactivity observed in patients with atherosclerotic disease.28 33 44 45 Thus we cannot exclude confounding by other substances that promote platelet activation like thromboxane A2, oxygen-derived free radicals, activated thrombin, ADP, tissue factor, fibroblast growth factor, fibrinogen, ß-thromboglobulin, platelet factor 4, and platelet derived growth factor.46 The concentration of platelet aggregation inhibitors (prostacyclin, endothelium-derived relaxing factor, tissue plasminogen activator) has also been shown to be reduced at sites of vascular endothelium injury46 and may further contribute to increased platelet aggregation, thrombus formation, and vasospasm.
In conclusion, we found that total blood serotonin is associated with the presence of angiographically confirmed coronary artery disease and subsequent cardiac events, particularly in younger age groups and gives further evidence that serotonin is related to clinical disease.
| Acknowledgments |
|---|
Received March 24, 1999; revision received April 6, 1999; accepted April 12, 1999.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
Y. Shinohara, K. Nishimaru, T. Sawada, A. Terashi, S. Handa, S. Hirai, K. Hayashi, H. Tohgi, Y. Fukuuchi, S. Uchiyama, et al. Sarpogrelate-Aspirin Comparative Clinical Study for Efficacy and Safety in Secondary Prevention of Cerebral Infarction (S-ACCESS): A Randomized, Double-Blind, Aspirin-Controlled Trial Stroke, June 1, 2008; 39(6): 1827 - 1833. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. J. Lorenowicz, M. Fernandez-Borja, and P. L. Hordijk cAMP Signaling in Leukocyte Transendothelial Migration Arterioscler. Thromb. Vasc. Biol., May 1, 2007; 27(5): 1014 - 1022. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. J. Lorenowicz, J. van Gils, M. de Boer, P. L. Hordijk, and M. Fernandez-Borja Epac1-Rap1 signaling regulates monocyte adhesion and chemotaxis J. Leukoc. Biol., December 1, 2006; 80(6): 1542 - 1552. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Bianchi, O. Kunduzova, E. Masini, C. Cambon, D. Bani, L. Raimondi, M.-H. Seguelas, S. Nistri, W. Colucci, N. Leducq, et al. Oxidative Stress by Monoamine Oxidase Mediates Receptor-Independent Cardiomyocyte Apoptosis by Serotonin and Postischemic Myocardial Injury Circulation, November 22, 2005; 112(21): 3297 - 3305. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Schins, A. Honig, H. Crijns, L. Baur, and K. Hamulyak Increased Coronary Events in Depressed Cardiovascular Patients: 5-HT2A Receptor as Missing Link? Psychosom Med, September 1, 2003; 65(5): 729 - 737. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. L. Serebruany, A. H. Glassman, A. I. Malinin, C. B. Nemeroff, D. L. Musselman, L. T. van Zyl, M. S. Finkel, K. R. R. Krishnan, M. Gaffney, W. Harrison, et al. Platelet/Endothelial Biomarkers in Depressed Patients Treated With the Selective Serotonin Reuptake Inhibitor Sertraline After Acute Coronary Events: The Sertraline AntiDepressant Heart Attack Randomized Trial (SADHART) Platelet Substudy Circulation, August 26, 2003; 108(8): 939 - 944. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Fumeron, D. Betoulle, V. Nicaud, A. Evans, F. Kee, J.-B. Ruidavets, D. Arveiler, G. Luc, and F. Cambien Serotonin Transporter Gene Polymorphism and Myocardial Infarction: Etude Cas-Temoins de l'Infarctus du Myocarde (ECTIM) Circulation, June 25, 2002; 105(25): 2943 - 2945. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Linz and R. Veelken Serotonin 5-HT3 receptors on mechanosensitive neurons with cardiac afferents Am J Physiol Heart Circ Physiol, May 1, 2002; 282(5): H1828 - H1835. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Brasil, R. M. Temsah, K. Kumar, H. Kumamoto, N. Takeda, and N. S. Dhalla Blockade of 5-HT2A Receptors by Sarpogrelate Protects the Heart Against Myocardial Infarction in Rats Journal of Cardiovascular Pharmacology and Therapeutics, March 1, 2002; 7(1): 53 - 59. [Abstract] [PDF] |
||||
![]() |
A. K.L. Banes and S. W. Watts Enhanced Contraction to 5-Hydroxytryptamine Is Not Due to "Unmasking" of 5-Hydroxytryptamine1B Receptors in the Mesenteric Artery of the Deoxycorticosterone Acetate-Salt Rat Hypertension, October 1, 2001; 38(4): 891 - 895. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Kawano, H. Tsuji, H. Nishimura, S. Kimura, S. Yano, N. Ukimura, Y. Kunieda, M. Yoshizumi, T. Sugano, K. Nakagawa, et al. Serotonin induces the expression of tissue factor and plasminogen activator inhibitor-1 in cultured rat aortic endothelial cells Blood, March 15, 2001; 97(6): 1697 - 1702. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Watanabe, R. Pakala, S. Koba, T. Katagiri, and C. R. Benedict Lysophosphatidylcholine and Reactive Oxygen Species Mediate the Synergistic Effect of Mildly Oxidized LDL With Serotonin on Vascular Smooth Muscle Cell Proliferation Circulation, March 13, 2001; 103(10): 1440 - 1445. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Ito, U. Ikeda, M. Shimpo, K. Yamamoto, and K. Shimada Serotonin Increases Interleukin-6 Synthesis in Human Vascular Smooth Muscle Cells Circulation, November 14, 2000; 102(20): 2522 - 2527. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1999 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |