Articles |
From the Department of Internal Medicine II, Cardiology, University of Ulm Medical Center, and the GSFNational Research Center for Environment and Health, MEDIS Institute, Neuherberg (M.S.), Germany.
Correspondence to Dr Margit Fröhlich, Abteilung Innere Medizin II, Universität Ulm, Robert-Koch-Straße 8, D-89081 Ulm, Germany.
| Abstract |
|---|
|
|
|---|
1-glycoprotein (P<.001),
fibrinogen (measured by immunonephelometry; P<.001),
plasminogen activator inhibitor-1
(P=.002), LDL cholesterol (P=.003),
and triglyceride levels (P<.001). HDL
cholesterol (P<.001) and cortisol
(P=.001) showed inverse seasonal patterns, with a maximum
during summertime. No statistically significant seasonal variations
were seen for red blood cell aggregation, complement factor C4, total
cholesterol, ceruloplasmin, haptoglobin, white blood
cell count, and plasminogen. These data do not support the
hypothesis that increased morbidity and mortality from
cardiovascular diseases during winter may be mainly
attributable to increased synthesis of acute-phase proteins due to
infections. The cause for the seasonal variations in rheological and
hemostatic parameters remains unclear and should be studied
in more detail.
Key Words: seasonal variation hemostatic parameters blood rheology acute-phase reactants healthy young adults
| Introduction |
|---|
|
|
|---|
A seasonal variability with peak concentrations during cold months was shown for fibrinogen.15 16 17 Fibrinogen may contribute to atherothrombogenesis by several mechanisms: involvement in early atherosclerotic plaque formation (ie, providing an adsorptive surface for LDL accumulation), involvement in the response to endothelial damage, increased platelet aggregability by interaction with glycoprotein IIb/IIIa receptors on the platelet surface, increased RBC aggregation, and finally, contribution to PV (reviewed in Reference 1818 ).
Several hypotheses have been proposed to explain the rise of plasma fibrinogen levels in winter. Some authors have suggested an increased incidence in winter respiratory infections, which might cause an acute-phase reaction and consecutively lead to an increase in fibrinogen.15 19 20 Previous studies were carried out in cohorts of elderly people, and upper respiratory tract infections had not been excluded.4 5 15 16
We studied seasonal variations of a variety of hemorheological and hemostatic variables in young, healthy subjects and followed them for a 1-year period.
| Methods |
|---|
|
|
|---|
Fasting venous blood samples were obtained in 4-week intervals during identical conditions (in the morning, at room temperature) for a given individual. EDTA blood was immediately centrifuged, and obtained aliquots were stored at -70°C. Blood pressures were measured in the sitting position.
RBC count, WBC count, hemoglobin, and Hk were measured in a
Coulter counter from EDTA blood. Whole BV was determined in EDTA blood
by a controlled shear stress rheometer at different shear stresses
(100, 200, 500, and 2000 Pa) and 37°C, first at native Hk and second
at 45% Hk (Carrimed); PV was measured by a falling ball viscometer at
37°C (Haake) that had been calibrated against a Coulter-Harkness
capillary viscometer (Coulter Electronics). RBC aggregation was
measured by a photooptic method (MA 1 aggregometer, Myrenne) and RBC
deformability by the St George Filtrometer (Carrimed), both in EDTA
blood. Nuclepore filters of the same batch (Nuclepore) were taken
throughout. All viscosity measurements were done within 2 hours after
venipuncture. CRP, ceruloplasmin, and
1-glycoprotein were measured by nephelometry (BNA
nephelometer). Total and HDL cholesterol and
triglycerides were determined by routine enzymatic tests.
LDL cholesterol was calculated by the method of Friedewald
et al.21 Fibrinogen was determined by two different
methods: immunonephelometry (Behringwerke) and by the clotting method
of Clauss.22 In addition, plasminogen,
2-macroglobulin, PAI-1 (Kabi Vitrum/Pharmacia) were measured.
Haptoglobin was determined by immunodiffusion (Behringwerke AG) and
complement factor C4 by immunonephelometry (Behringwerke). Assays were
done for interleukin-1
and -1ß, tumor necrosis factor
(ELISA,
R&D Diagnostics), and cortisol (RIA, Dianovo Immunotech
GmbH). RBC sedimentation rate was determined according to the method of
Westergren.
Routine laboratory tests, including RBC count, WBC count, hemoglobin,
Hk, total cholesterol, HDL cholesterol,
triglycerides, fibrinogen (Clauss method22 ),
cortisol, and viscosity measurements were subject to continuous quality
control. The remaining parameters were determined in
batches. Coefficients of variation for repeated measurements were 3.3%
and 2.1% for BV 2000 and BV 500, 0.7% for PV, 10.3% for RBC
aggregation, 5.9% for RBC deformability, 7% for fibrinogen (measured
by immunonephelometry), 3.8% for plasminogen, 7.6% for
PAI-1, 3.4% for complement factor C4, 4.0% for ceruloplasmin, and
10% for
1-glycoprotein.
Statistical Methods
Seasonal variation was assumed to follow a sinusoidal curve with
a period of 1 year. This curve can be written as
![]() |
![]() |
t/365)
as one independent variable and cos(2
t/365) as the second
independent variable. Random errors were assumed to be normally
distributed. The Shapiro test and other sample statistics were used to
assess departures from normality. However, for no variable were the
departures too serious to necessitate a transformation or any other
action. Since the data of each patient represented repeated
measurements, a correlated error structure had to be considered. After
trying several candidates, the compound symmetry structure appeared to
be satisfactory for all variables. The regression coefficients a,
b, and c were estimated by restricted maximum likelihood. The statistical test for the seasonal difference was performed at a nominal 5% level. All computations were carried out using SAS software, version 6.11 for Windows 3.1, in particular, its procedure MIXED.23
| Results |
|---|
|
|
|---|
|
Peak fitted levels are present during autumn and winter months
(October to February). A strong variation is seen for PV (Fig 1
) and RBC deformability (Fig 2
). PV reaches a maximum fitted level in
January, with a value of 1.23 mPa·s. It decreases during summer
months, and the minimum is reached in July, with a value of 1.16
mPa·s (P<.001 for the seasonal difference). RBC
deformability shows a large seasonal difference, with the peak level in
November (14.5 AU) and the lowest level in April (13.1 AU)
(P<.001). Seasonal differences in whole BV, hemoglobin, Hk,
MCV, and platelet count are not as pronounced but still
statistically significant. However, after standardization on an Hk of
45%, seasonal differences in whole BV decreased and became
statistically insignificant for shear stresses of 100 and 2000 mPa.
There was no statistically significant seasonal variation in RBC
aggregation.
|
|
Acute-Phase Reactants
Seasonal variations for acute-phase reactants are shown in Table 2
. Statistically significant seasonal
variations are seen for fibrinogen (measured immunonephelometrically),
1-glycoprotein, RBC sedimentation rate, cortisol, and
PAI-1. Maximum fitted values are found between February and April.
Exceptions are RBC sedimentation rate and cortisol, which show a
maximum in August. In addition, a weak seasonal difference is found for
plasminogen, with a maximum value in August, although not
statistically significant. Displacement of the slope from zero is large
for
1-glycoprotein. Its fitted maximum level is reached
in February (0.7 g/L) and reveals an impressive decline from May
to August, with a minimum level of 0.4 g/L (P<.001
for the seasonal differences). Seasonal differences for fibrinogen
measured nephelometrically (Fig 3
) and
PAI-1 are less pronounced, although still significant. No statistically
significant seasonal changes are seen for fibrinogen measured by the
Clauss method,22 complement factor C4, ceruloplasmin,
haptoglobin, and WBC count, although absolute differences in some of
these parameters (complement factor C4 and ceruloplasmin)
were sizable.
|
|
Measurements of
2-macroglobulin, CRP, interleukin-1
,
interleukin-1ß, and tumor necrosis factor-
did not provide
conclusive results, probably attributable to the lack of sensitivity in
the methods used.
Lipoproteins
Cholesterol and triglycerides are shown in
Table 3
. HDL and LDL
cholesterol and triglycerides demonstrate a
statistically significant seasonal variation. LDL
cholesterol peaks in January (3.3 mmol/L) and
reaches its lowest levels in July (3.0 mmol/L)
(P=.003). HDL cholesterol reveals an inverse
pattern. The peak value is seen in August (1.55 mmol/L) and
the lowest value is reached in February (1.30 mmol/L)
(P<.001). The result for triglycerides exhibits
a peak in September. The lowest level is reached in April
(P<.001).
|
No gender differences were seen in the seasonal variation of the parameters under study.
| Discussion |
|---|
|
|
|---|
Seasonal variation was confirmed for fibrinogen, measured quantitatively by nephelometry (P<.001). The maximum level was reached in April, and the seasonal difference was 0.32 g/L. This amount of variation may be of considerable clinical relevance, as in the ECAT study, plasma fibrinogen concentrations of patients with coronary events compared with those without differed by the same magnitude (0.28 g/L).9 This was also confirmed by data from the Caerphilly and Speedwell Collaborative Heart Disease Study.12 The Bezafibrate Infarction Prevention Study Group demonstrated that an increase in fibrinogen of 0.75 g/L (equivalent to one SD) increased the risk of CHD by 29% in men.25
Fibrinogen measurement by the clotting ratebased assay of Clauss22 did not reveal a statistically significant seasonal difference. This discrepancy may be explained by interindividual and intraindividual qualitative differences in fibrinogen. In blood, fibrinogen is degraded, which leads to a variety of degradation products with different molecular weights: high-molecular-weight, low-molecular-weight, and verylow-molecular-weight fibrinogen. DeMaat et al26 showed that different fibrinogen assays have distinct sensitivities for the various fibrinogen forms, leading to different values in fibrinogen measurement.
Considerable seasonal differences were demonstrated for
acute-phase reactants like PAI-1 and
1-glycoprotein.
Peak values were seen in February and March, respectively. The seasonal
difference of PAI-1 level was 2.7 ng/mL. By comparison,
Juhan-Vague et al27 demonstrated a difference of 3.4
ng/mL in PAI-1 antigen levels between patients with myocardial
infarction or coronary death compared with the event-free
group.
1-Glycoprotein level rises strongly with the
severity and extent of CHD. Mori et al28 reported a
difference in
1-glycoprotein of 0.12 g/L
comparing patients in the lower quartile of the Gensini's score with
the upper quartile. Our data showed a seasonal variation of 0.29
g/L, which was even larger than the difference reported between
minor and severe CHD. In reference to the markers of the acute phase,
cortisol levels were related inversely, with a maximum level seen in
August. This finding may be explained by a cortisol-induced suppression
of interleukin-6, which is known to be a potent trigger of the
acute-phase reaction.29 Other acute-phase markers like
plasminogen, complement factor C4, and ceruloplasmin did
not show statistically significant seasonal differences. In
interpreting these results, however, it has to be considered that the
study population was relatively small, which limits the statistical
power of the results. This is evident, for example, when looking at the
above-mentioned study of Mori et al,28 in which
ceruloplasmin differed by 30 mg/L between those with minor and
those with severe CHD. In comparison, we observed a seasonal variation
of 14 mg/L, which, was not statistically significant.
The underlying statistical model to fit the data clearly has limitations because it does not seem to be fully appropriate for all variables. For example, monthly means of several variables were not within the 95%CI band.
In contrast to the present study, other studies measured
inflammatory and hemostatic parameters in elderly people,
and subjects with clinical signs of infection had not been
excluded.2 4 5 15 16 30 Significant seasonal variations of
plasma fibrinogen concentrations, with an increase in winter months,
have been reported by Stout and Crawford16 and Woodhouse
et al.15 This was consistent with an increase in
PV.15 The authors attributed these findings to an
increased incidence of upper respiratory tract infections in winter,
during which fibrinogen is acting as an acute-phase protein. In the
study of Woodhouse et al,15 there was a strong association
between fibrinogen and other markers of inflammation, such as
neutrophil count, CRP, and
1-antichymotrypsin, as well as cough and
coryza.
However, several longitudinal studies demonstrated an association between WBC count and myocardial infarction.12 31 32 33 34 35 36 37 38 39 Furthermore, there is experimental evidence that inflammatory processes play an essential role in the pathogenesis of atherosclerosis.40 41 In the ECAT study,9 a positive association has been found between CRP and the incidence of myocardial infarction. CRP levels were higher in patients with unstable angina, compared with a group of patients with stable angina,42 and high CRP levels were associated with a worse prognosis in patients with unstable angina.43 However, the pathomechanism leading to elevated CRP and fibrinogen levels in CHD remains unclear.
The "inflammatory and thrombogenic state" could be due to acute and chronic infections. Some authors have found an association between infectious agents such as Chlamydia pneumoniae, Helicobacter pylori, herpes simplex virus, and cytomegalovirus and CHD.44 45 46 47 48 Chlamydia pneumoniae has been detected in atherosclerotic plaques of coronary arteries.49
In the present study, subjects with acute and chronic infections had been excluded. Furthermore, no seasonal variation in WBC count was seen, which supports the notion that infections were highly unlikely among the subjects during the 1-year period. Despite this, seasonal changes, with elevated levels in winter, were demonstrated for a variety of hemostatic and hemorheological parameters. This has to be attributed to so far unknown factors. Acute and chronic infections may augment basal seasonal changes.
Seasonal changes in LDL concentration, with peak levels in winter months, have been demonstrated by others.4 5 Robinson et al30 showed seasonal variations of total cholesterol levels, with a 3% to 5% increase in winter in a large sample of subjects. Seasonal difference of LDL cholesterol in our study was of the same magnitude. Declining HDL cholesterol levels in winter and an increase in summertime have not been reported yet. Previous studies have shown elevated HDL levels in winter.4 5 In the present study, triglyceride level peaks in September. Gordon et al5 measured the highest triglyceride levels in autumn, although the values were distributed in an irregular pattern, and Woodhouse et al4 measured the highest triglyceride levels in January.
Morbidity and mortality from cardiovascular diseases
are clearly elevated during winter months. Data from England and Wales
suggest a 30% increase in death from these causes during wintertime
among elderly people.1 Data from the MONICA Augsburg
coronary event register50 demonstrate a similar
pattern: an impressive increase in the incidence of acute myocardial
infarction during winter. Since an occlusive thrombus presents the
final pathophysiological hallmark of acute
myocardial infarction, parameters influencing blood
rheology and blood coagulation are most likely to play an important
role. Whole BV is determined by a variety of parameters,
including age, sex, nutrition, temperature, and time of day. Further
determinants are RBC flexibility, RBC aggregation, Hk, and
PV.51 Several high-molecular, asymmetrical proteins,
especially fibrinogen, are the major contributors to PV. In addition,
immunoglobulin M,
2-macroglobulin, haptoglobin, and lipoproteins
play a role.
In the WHO-MONICA project, regional differences in PV between the Glasgow (mean PV 1.327 mPa·s) and Augsburg (mean PV 1.261 mPa·s) populations were consistent with a two (men) to four (women) times higher coronary event rate in the West of Scotland population.52 Resch et al53 found that patients with a second event of stroke or myocardial infarction had higher BV and fibrinogen levels than control subjects, and the Caerphilly and Speedwell Collaborative Heart Disease Studies12 showed that fibrinogen and PV were strong independent risk factors for CHD. Several other studies have confirmed the association of fibrinogen levels and CHD.10 11 12 13 14 It is therefore conceivable that elevated fibrinogen levels in winter, as demonstrated in this study and in others,15 16 correlate with increased PV and thus may contribute to raised mortality from CHD during winter months.
However, the fibrinogen curve did not strictly parallel the graph of PV, where maximum values were shown in January. One possible explanation for this finding is that PV is also markedly influenced by temperature.51 54 This finding may help to explain the relationship between myocardial infarct deaths and decreasing temperatures that had been demonstrated in three recent studies.55 56 57 Classical cardiovascular risk factors seem to be aggravated by cold ambient temperature. Woodhouse et al2 found that a 1°C decrease in living room temperature was associated with a rise of 1.3 mm Hg in systolic blood pressure and a rise of 0.6 mm Hg in diastolic blood pressure. Robinson et al30 demonstrated a 3% to 5% increase of cholesterol levels in winter. Levels were strongly and negatively correlated with mean air temperature. This was confirmed in the large Caerphilly and Speedwell Prospective Heart Disease Study58 in which a fall in temperature of 16°C was associated with higher blood pressures. In addition, fibrinogen and platelet count were significantly inversely associated with air temperature.
In conclusion, the present study shows that a variety of factors influencing blood flow properties are significantly altered in winter and thus may contribute at least in part to the higher incidence of CHD during these months.
| Selected Abbreviations and Acronyms |
|---|
|
Received March 21, 1997; accepted June 25, 1997.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. S. Boekhoorn, J. R. Vingerling, J. C. M. Witteman, A. Hofman, and P. T. V. M. de Jong C-reactive Protein Level and Risk of Aging Macula Disorder: The Rotterdam Study Arch Ophthalmol, October 1, 2007; 125(10): 1396 - 1401. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. R. Rudnicka, A. Rumley, G. D.O. Lowe, and D. P. Strachan Diurnal, Seasonal, and Blood-Processing Patterns in Levels of Circulating Fibrinogen, Fibrin D-Dimer, C-Reactive Protein, Tissue Plasminogen Activator, and von Willebrand Factor in a 45-Year-Old Population Circulation, February 27, 2007; 115(8): 996 - 1003. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. von Kanel and J. E. Dimsdale Hemostatic Alterations in Patients With Obstructive Sleep Apnea and the Implications for Cardiovascular Disease Chest, November 1, 2003; 124(5): 1956 - 1967. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Kamath and G.Y.H. Lip Fibrinogen: biochemistry, epidemiology and determinants QJM, October 1, 2003; 96(10): 711 - 729. [Full Text] [PDF] |
||||
![]() |
W. Koenig, M. Sund, M. Frohlich, H. Lowel, W. L. Hutchinson, and M. B. Pepys Refinement of the Association of Serum C-reactive Protein Concentration and Coronary Heart Disease Risk by Correction for Within-Subject Variation over Time: The MONICA Augsburg Studies, 1984 and 1987 Am. J. Epidemiol., August 15, 2003; 158(4): 357 - 364. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. K. Cheung, G. Yan, T. Greene, J. T. Daugirdas, J. T. Dwyer, N. W. Levin, D. B. Ornt, G. Schulman, and G. Eknoyan Seasonal Variations in Clinical and Laboratory Variables among Chronic Hemodialysis Patients J. Am. Soc. Nephrol., September 1, 2002; 13(9): 2345 - 2352. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Frohlich, M. Sund, B. Thorand, W. L. Hutchinson, M. B. Pepys, and W. Koenig Lack of Seasonal Variation in C-Reactive Protein Clin. Chem., March 1, 2002; 48(3): 575 - 577. [Full Text] [PDF] |
||||
![]() |
M. H. Davidson, K. C. Maki, P. Marx, A. C. Maki, M. S. Cyrowski, N. Nanavati, and J.-C. Arce Effects of Continuous Estrogen and Estrogen-Progestin Replacement Regimens on Cardiovascular Risk Markers in Postmenopausal Women Arch Intern Med, November 27, 2000; 160(21): 3315 - 3325. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Schaadt Oxygenator thrombosis: an international phenomenon Perfusion, December 1, 1999; 14(6): 425 - 435. [Abstract] [PDF] |
||||
![]() |
L J Murray, D P J O'Reilly, G M L Ong, C O'Neill, A E Evans, and K B Bamford Chlamydia pneumoniae antibodies are associated with an atherogenic lipid profile Heart, March 1, 1999; 81(3): 239 - 244. [Abstract] [Full Text] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1997 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |