Abstract
Objective
Cross-sectional and prospective studies have linked cardiovascular events and traditional risk factors (TRFs) with higher plasma fibrinogen levels. In a young cohort, we sought to determine longitudinal associations between changes in/development of TRFs and fibrinogen levels over 13 years.
Methods
We included 2525 adults from the CARDIA study, aged 25-37 with fibrinogen and TRFs measured at year 7 (study baseline; 1992-1993); and year 20 (follow-up). Multiple linear regressions were used to compare mean changes in fibrinogen to TRFs.
Results
Mean fibrinogen increased by 71mg/dL vs. 70mg/dL (p=NS) in black vs. white men, and 78mg/dL vs. 68mg/dL (p<0.05) in black vs. white women, respectively over 13 years. After multivariable adjustments, fibrinogen generally rose with increasing BMI (p<0.001; all sex/race groups), LDL-cholesterol, log triglycerides and diastolic blood pressure; and fell with increasing HDL-cholesterol and physical activity. 13-year increase in fibrinogen for persons who quit smoking or became non-obese were comparable (p=NS) to that of never-smokers and never-obese persons.
Conclusions
Among young black and white men and women with few baseline cardiovascular risk factors, fibrinogen tracked longitudinally with changes in TRFs over 13 years through middle-age. There was a strong inverse longitudinal relationship between modifiable risk factors (weight loss/smoking cessation) and 13-year change in fibrinogen. Our study helps provide some insight into the role of fibrinogen as a disease marker in the associations between fibrinogen and CVD.
Keywords: Fibrinogen, risk factors, cardiovascular disease prevention, obesity, smoking, sex, race
Introduction
Fibrinogen is the precursor of fibrin - a cofactor for platelet aggregation, and a major determinant of blood viscosity and atherogenesis.1 It is the most abundant component of thrombi, and has been shown in multiple studies to have significant independent positive associations with coronary heart disease (CHD) and cardiovascular disease (CVD).2 Fibrinogen is positively associated with CVD risk factors such as age, smoking history, physical activity, body mass index (BMI), total and low-density lipoprotein (LDL), systolic blood pressure,3 and measures of subclinical atherosclerosis.4, 5 Conversely, fibrinogen levels decline with lifestyle interventions such as smoking cessation, exercise and moderate alcohol consumption.2, 3, 6
There are currently sparse longitudinal data linking changes in CVD risk factors to changes in fibrinogen levels over time. The very few studies that have been able to directly compare longitudinal alterations in established risk factors with fibrinogen levels have been limited by small sample size, examination of the association of changes in fibrinogen with a few/limited number of risk factors, homogeneous populations of participants, and a short follow-up period.7-9
Using data from the Coronary Artery risk Development in Young Adults (CARDIA) cohort – a young, biracial cohort of men and women – we examined long-term associations between changes in traditional cardiovascular risk factors (smoking status, blood pressure and hypertension, cholesterol, triglycerides and dyslipidemia, BMI, physical activity and obesity, as well as glucose and diabetes), and changes in fibrinogen levels over 13 years. Because of the relative youth of the study population with minimal atherosclerosis at baseline, findings from this study would importantly contribute to our understanding of associations between fibrinogen – a novel cardiovascular risk marker – and traditional CVD risk factors. In addition, a longitudinal study of this nature with a relatively long follow-up period in a large biracial cohort might strengthen associations and recognition of fibrinogen as a marker, rather than a causative factor of CVD.
Methods
Study participants
Study participants were from CARDIA, an ongoing multicenter prospective cohort study designed to investigate the evolution of CVD risk factors and subclinical atherosclerosis in young adults. Details of the study design, inclusion/exclusion criteria and baseline characteristics have previously been described.10 Briefly, in 1985-1986 the cohort enrolled 5115 black and white adults (55% women) aged 18-30 years recruited from four U.S. urban areas (Birmingham, Alabama; Oakland, California; Chicago, Illinois and Minneapolis, Minnesota). The study was designed to include roughly balanced numbers of participants by age, sex, race and educational attainment. The institutional review boards at all the study sites approved the study protocol, and written informed consent was obtained from each study participant.
Six follow-up examinations (at years 2, 5, 7, 10, 15 and 20) have been completed since enrollment. Fibrinogen was measured at year 7 (our study baseline [baseline07]), and again at year 20 (termed follow-up in this article). Of the 3844 participants examined at baseline07, 804 persons were lost to follow-up. Our study population included 2971 non-pregnant CARDIA women and men with fibrinogen measurements at both baseline07 and follow-up. We excluded persons with CHD (n=6), persons with non-fasting glucose and missing data for triglyceride and LDL (n=312), and persons with missing covariates of interest (n=105). The final cohort for analysis included 1084 black and 1464 white, making a total of 2548 participants.
Risk Factor Measurements
Age, race, socioeconomic measures, diabetes history, cigarette smoking status, family history and medication use were obtained by interviewer-administered questionnaire.10 BMI was obtained by using a balanced beam scale and a vertical ruler for obtaining height and weight respectively; then calculated by dividing weight in kilograms by the square of the height in meters. Resting blood pressure was measured three times at each examination, and the average of the last two measurements was used. Current smoking was defined as at least 5 cigarettes per week almost every week for at least 3 months. Physical activity was assessed by a standard instrument, and energy expenditure for all moderate and vigorous activities was calculated in exercise units. 11 Venous blood samples were obtained from participants after an 8-hour fast. Plasma triglycerides, LDL and high-density lipoprotein (HDL) cholesterol levels were determined using an enzymatic assay by Northwest Lipids Research Laboratory (Seattle, Washington). Glucose was measured by the hexokinase method.
Risk factor progression was characterized by changes over 13 years from baseline07 to follow-up (year 20 – year 7). Diabetes was defined as fasting glucose ≥126 mg/dL or taking diabetic medication; hypertension was defined as systolic blood pressure ≥140mmHg, diastolic blood pressure (DBP) ≥90mmHg, or taking anti-hypertensive medication; dyslipidemia was defined as low HDL cholesterol (<40mg/dL [men] or <50mg/dL [women]) and/or high LDL cholesterol (>130mg/dL) or taking lipid-lowering medication, hypertriglyceridemia was defined as triglycerides >200mg/dL; obesity was defined as BMI ≥30kg/m2. We categorized changes in risk factors as never had risk factor, new-onset of risk factor, and continued with risk factor.
Fibrinogen measurement
Blood samples were drawn from participants between 7a.m. and 10a.m. after an 8-hour fast. The samples were then mixed by repeated inversion and within 10 minutes of collection, were spun in a refrigerated centrifuge at 4°C for 20 minutes. The samples were stored within 90 minutes at -70°C freezer. Samples stored since year 7 (1992-93) were assayed in 2003 using automated nephelometry as previously described.12 Fibrinogen was also measured at year 20 using this method. There is no evidence of sample degradation when samples are stored at -70°C.13 The intra- and inter-assay coefficients of variation were 2.7% and 2.6% at year 7, and 3.1% and 4.2% at year 20.
Statistical analysis
All analyses were done by race/sex strata. Baseline07 and follow-up characteristics and 13-year changes in major cardiovascular risk factors were computed and pairwise differences in covariates by race/sex groups were estimated using t-tests, chi-square tests, and Fisher's exact tests, as appropriate. Linear regression models were used to examine baseline07 or 13-year change in risk factors and 13-year change in fibrinogen as continuous variables while adjusting for age, baseline07 fibrinogen level, family history of heart disease, education, baseline07 smoking and alcohol use, baseline07 medications use, and all other risk factors [including DBP, glucose, HDL, LDL, triglycerides, BMI and physical activity] being examined in the model simultaneously. ANCOVA models were used to relate changes in adjusted mean fibrinogen over the 13 years to categories of changes in risk factors, with adjustments for covariates listed in the model above. In this case, the risk factors for the model were categorized as smoking, obesity, hypertension, diabetes, hypertriglyceridemia and dyslipidemia. To assess the influence of participant exclusions that resulted due to missing data or loss to follow-up, a multiple imputation analysis was performed based on the procedure described by Raghunathan et al.14 Imputations were performed 5 times and the results were pooled using the SAS MIANALYZE procedure to estimate valid confidence intervals and standard errors. Analyses were conducted with SAS statistical software version 9.2 (SAS Institute Inc, Cary, NC). P<0.05 was considered statistically significant.
Results
Baseline characteristics
Our study included 2548 participants (55% women, 43% black) with a mean age of 32.2 years (range 25-37 years) at baseline07. Tables 1 and 2 present summary statistics for key variables at baseline07 and follow-up; as well as 13-year changes in these variables.
Table 1.
Baseline Characteristics by Sex- Race: the CARDIA Study, 1992-2006¶
| Variables | Men | Women | ||
|---|---|---|---|---|
| Black (N=428) | Whites (N=726) | Black (N=656) | Whites (N=738) | |
| Age, y | 31.6 (3.7) | 32.7 (3.3)‡ | 31.6 (3.8) | 32.7 (3.3)‡ |
| Highest education attained, y | 14.2 (2.3) | 16.1 (2.6)‡ | 14.6 (2.1) | 16.2 (2.4)‡ |
| Family history of CVD, % | 39.7 | 32.0† | 43.3 | 33.3‡ |
| Diastolic BP, mmHg | 72.3 (9.8) | 70.3 (9.0)‡ | 69.9 (10.3) | 64.8 (8.0)‡ |
| LDL cholesterol, mg/dL | 111.8 (35.6) | 113.7 (32.4) | 104.5 (29.3) | 101.9 (27.6) |
| HDL cholesterol, mg/dL | 50.8 (14.2) | 45.7 (10.9)‡ | 55.1 (13.7) | 56.5 (12.5)* |
| Glucose, ug/dL | 94.1 (22.7) | 93.3 (10.8) | 89.1 (15.3) | 88.4 (10.2) |
| Triglycerides, mg/dL | 81.6 (51.4) | 97.2 (59.3)‡ | 67.0 (37.8) | 69.7 (39.1) |
| Physical activity, exercise unit | 477.6 (345.1) | 414.2 (263.5)‡ | 231.5 (215.7) | 309.5 (229.2)‡ |
| BMI, kg/m2 | 27.1 (5.2) | 25.8 (4.0)‡ | 28.7 (7.3) | 24.4 (5.1)‡ |
| Fibrinogen, mg/dL | 328.3 (72.7) | 305.1 (56.8)‡ | 371.0 (80.9) | 324.5 (67.0)‡ |
| Antihypertensive medication, % | 2.1 | 1.1 | 2.9 | 0.4‡ |
| Lipid-lowering medication, % | 0.2 | 0.0 | 0.3 | 0.5 |
| Current smoking, % | 31.5 | 18.2‡ | 25.6 | 15.7‡ |
| Current alcohol use, % | 61.4 | 68.9† | 36.3 | 54.2‡ |
| Obesity, % | 24.1 | 12.8‡ | 36.9 | 12.7‡ |
| Hypertension, % | 6.1 | 4.1 | 6.7 | 0.8‡ |
| Diabetes, % | 1.6 | 0.7 | 1.1 | 0.5 |
| Hypertriglyceridemia, % | 4.0 | 6.5 | 0.9 | 1.4 |
| Dyslipidemia, % | 39.9 | 46.6* | 46.8 | 39.6† |
P<0.05
P<0.0l
P<0.001 compared with blacks between the value of the characteristic within gender.
§ Data are given as means (SD) unless otherwise specified.
Abbreviations: CVD, cardiovascular disease (included heart attack and stroke); BP, blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein; BMI, body mass index
Table 2.
Selected Year 20 Characteristics and 13-Year Changes in Cardiovascular Risk Factors by Sex- Race: the CARDIA Study, 1992-2006¶
| Variables | Men | Women | ||
|---|---|---|---|---|
| Black (N=428) | Whites (N=726) | Black (N=656) | Whites (N=738) | |
| Year 20 | ||||
| Antihypertensive medication, % | 19.2 | 11.7‡ | 28.5 | 6.4‡ |
| Lipid-lowering medication, % | 7.9 | 13.4† | 5.5 | 4.2 |
| Current smoking, % | 27.1 | 13.9‡ | 20.0 | 11.1‡ |
| Current alcohol use, % | 54.9 | 69.7‡ | 34.9 | 60.1‡ |
| Obesity, % | 43.9 | 26.3‡ | 57.8 | 24.9‡ |
| Hypertension, % | 25.0 | 14.9‡ | 34.9 | 8.3‡ |
| Diabetes, % | 10.5 | 4.4‡ | 10.1 | 2.6‡ |
| Hypertriglyceridemia, % | 6.8 | 16.8‡ | 2.9 | 5.0* |
| Dyslipidemia, % | 52.8 | 58.5 | 48.3 | 37.3‡ |
| 13-y Difference (year 20 – year 7) | ||||
| Diastolic BP, mmHg | 3.0 (11.7) | 0.8 (9.4)‡ | 5.6 (12.3) | 2.4 (9.1)‡ |
| LDL cholesterol, mg/dL | 0.1 (33.1) | 1.1 (32.2) | 3.3 (25.9) | 4.3 (24.7) |
| HDL cholesterol, mg/dL | -1.1 (11.3) | 1.0 (9.8)‡ | 2.1 (12.0) | 5.1 (12.3)‡ |
| Glucose, ug/dL | 9.6 (29.5) | 7.6 (16.9) | 11.3 (26.6) | 5.1 (13.7)‡ |
| Triglycerides, mg/dL | 24.1 (57.4) | 29.9 (63.3) | 21.8 (42.7) | 24.9 (48.0) |
| Physical activity, exercise unit | -51.5 (345.2) | -1.3 (260.1)† | -9.9 (248.4) | 27.1 (235.9)† |
| BMI, kg/m2 | 2.9 (3.6) | 2.5 (4.9) | 3.8 (4.5) | 2.7 (4.1)‡ |
| Fibrinogen, mg/dL | 71.0 (71.2) | 69.5 (67.8) | 78.1 (86.2) | 67.8 (77.1)* |
P<0.05
P<0.0l
P<0.001 compared with blacks between the value of the characteristic within gender.
§ Data are given as means (SD) unless otherwise specified.
Abbreviations: CVD, cardiovascular disease (included heart attack and stroke); BP, blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein; BMI, body mass index
Baseline07 mean fibrinogen level was higher in black men and women compared with white men and women, respectively (p<0.001, Table 1). Whites were on average one year older and attained higher education compared with blacks (p<0.001). Compared with black men, white men at baseline07 had significantly lower mean DBP and BMI and a lower prevalence of family history of CVD, obesity and smoking history; but had higher mean triglycerides, more prevalent dyslipidemia, lower HDL cholesterol, and less robust physical activity profile. On the other hand, the overall baseline07 cardiovascular risk factor profile shown in table 1 was better in white than black women.
Over 13 years, mean fibrinogen increased the most in black women, and least in white women (p <0.05; Table 2); with similar increases in black and white men. The Pearson's correlation coefficients between baseline07 and follow-up were 0.60, 0.47, 0.54 and 0.53 for black men, white men, black women and white women, respectively (all p<0.001). With the exception of hypertriglyceridemia, which increased more in white men and women than black men and women (p<0.05), other risk factors overall became more prevalent in blacks.
Multivariable-adjusted 13-year changes in fibrinogen level in relation to baseline07 risk factors
For both sexes and races, an unfavorable baseline07 risk factor profile was generally associated with increases in mean fibrinogen levels over 13 years of follow-up (Table 3). Baseline07 BMI showed the highest correlation with 13-year increase in fibrinogen in all persons regardless of sex/race (all p<0.05).
Table 3.
Adjusted Regression Coefficients for Mean Changes in Fibrinogen over 13 Years in relation to Baseline Risk Factor by Sex and Race: the CARDIA Study, 1992-2006¶
| Change in fibrinogen (mg/dL) | ||||
|---|---|---|---|---|
| Men | Women | |||
| Baseline07 risk/lifestyle factors (per 1 SD increase) | Blacks (N=428), β (SE) | Whites (N=726), β (SE) | Blacks (N=656), β (SE) | Whites (N=738), β (SE) |
| Diastolic BP (per 10 mmHg) | 12.4 (3.4)§ | 1.8 (2.8)† | 6.4 (3.2)¶ | 7.9 (3.4)¶ |
| Glucose (per 15 ug/dL) | -2.2 (3.7) | -5.5 (4.8) | -17.6 (5.9)∥ | -0.9 (4.0)* |
| LDL cholesterol (per 31 mg/dL) | 5.2 (3.1) | 2.4 (2.4) | -0.8 (3.5) | 5.0 (3.4)* |
| HDL cholesterol (per 13 mg/dL) | -1.6 (3.7) | 3.9 (3.6)* | -0.4 (3.5) | 1.4 (3.4) |
| Log triglycerides (per 0.5 log mg/dL) | -1.9 (3.8) | 0.6 (2.9) | -7.1 (3.9)¶ | -6.5 (3.5)¶ |
| Body mass index (per 6 kg/m2) | 7.7 (4.1)¶ | 11.4 (4.1)∥ | 11.9 (3.1)§ | 21.5 (3.8)§‡ |
| Physical activity (per 274 activity units) | -2.5 (2.6) | -3.5 (2.5) | -5.6 (4.1) | -1.2 (3.3) |
P<0.05
P<0.01
P<0.001 compared to Blacks within gender.
P <0.001
P<0.01
P<0.05
# Each risk factor represents a separate multiple linear regression model.
** Model adjusted for (baseline07) age and fibrinogen level, family history of heart disease, education, baseline07 medications (antihypertensive, lipid-lowering, diabetes), baseline07 (smoking, alcohol use), and all other baseline07 risk factors shown in table simultaneously.
†† The regression coefficient (β) represents the 13-year changes (year 20 – year 7) in fibrinogen per 1 SD (in parentheses) increase in baseline07 risk/lifestyle factor, adjusting for all other covariates in the model.
‡‡ Abbreviations: SE denotes standard error; SD, standard deviation; BP, blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein.
A significant inverse relationship was observed between change in mean fibrinogen per 1SD change in glucose in black (p<0.01) but not white women, and baseline07 log triglycerides in both black and white women (both p<0.05).
Multivariable-adjusted changes in fibrinogen level in relation to changes in risk factors over 13 years
For both sexes and races, mean fibrinogen increased with worsening risk factor profiles during 13 years of follow-up (Table 4). Increasing BMI exhibited the strongest influence on increased fibrinogen (p<0.01 for both sexes and races).
Table 4.
Adjusted Regression Coefficients for Mean Changes in Fibrinogen in relation to Changes in Risk Factor over 13 Years by Sex- Race: the CARDIA Study, 1992-2006¶
| Change in fibrinogen (mg/dL) | ||||
|---|---|---|---|---|
| Men | Women | |||
| Baseline07 risk/lifestyle factors (per 1 SD increase) | Blacks (N=428), β (SE) | Whites (N=726), β (SE) | Blacks (N=656), β (SE) | Whites (N=738), β (SE) |
| Δ Diastolic BP (per 10 mmHg) | -2.6 (3.5) | 5.0 (3.0)† | 4.5 (3.3) | 7.9 (3.4)¶ |
| Δ Glucose (per 21 ug/dL) | -2.1 (2.9) | -4.5 (3.2) | 0.6 (2.6) | 6.2 (4.7) |
| Δ LDL cholesterol (per 29 mg/dL) | 4.9 (3.5) | 4.6 (2.5)¶ | 17.7 (3.9)§ | 19.8 (3.4)§ |
| Δ HDL cholesterol (per 11 mg/dL) | -5.8 (3.4) | -13.3 (2.9)§* | -14.6 (3.1)§ | -17.9 (2.5)§ |
| Δ Log triglycerides (per 0.5 log mg/dL) | -1.0 (3.7) | 3.6 (2.7) | -0.1 (3.9) | 12.8 (3.1)§* |
| Δ Body mass index (per 4 kg/m2) | 13.6 (4.0)§ | 6.7 (2.1)§ | 30.2 (3.0)§ | 26.1 (2.7)§ |
| Δ Physical activity (per 268 activity units) | -4.4 (2.9) | -9.6 (2.8)§ | -13.9 (4.1)§ | -9.4 (3.4)∥ |
P<0.05
P<0.01
P<0.001 compared to Blacks within gender.
P <0.001
P<0.01
P<0.05
# Each risk factor represents a separate multiple linear regression model.
** Model adjusted for (baseline07) age and fibrinogen level, family history of heart disease, education, baseline07 medications (antihypertensive, lipid-lowering, diabetes), baseline07 (smoking, alcohol use), and all other baseline07 risk factors shown in table simultaneously.
†† The regression coefficient (β) represents the 13-year changes (year 20 – year 7) in fibrinogen per 1 SD (in parentheses) increase in baseline07 risk/lifestyle factor, adjusting for all other covariates in the model.
‡‡ Abbreviations: SE denotes standard error; SD, standard deviation; BP, blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein.
The longitudinal association between HDL cholesterol and fibrinogen was strongest in white compared with black men (p<0.05), while that for log triglycerides was most robust in white compared with black women (p<0.05).
Multivariable-adjusted changes in fibrinogen level by category of risk factor changes over 13 years
When risk factors were examined as categorical variables, participants with continued positive risk factor status over 13 years as well as those who acquired the risk factors during follow-up, had significantly larger increase in mean fibrinogen levels compared to individuals who never had the risk factor (Table 5). Apart from a non-significant increase in fibrinogen levels observed in black men who became obese and black women who stayed dyslipidemic during the 13-year interim, black and white men and women who became or stayed obese and those who became or stayed dyslipidemic exhibited significantly larger increases in fibrinogen compared with never-obese and never-dyslipidemic persons (p<0.05). The relationship between obesity and fibrinogen was strongest in women (p<0.001). Among persons who became non-obese (persons who lost weight), mean increase in fibrinogen levels was about the same as for never-obese persons at follow-up.
Table 5.
Adjusted Mean Changes in Fibrinogen in relation to Categorical Changes in Risk Factors over 13 Years by Sex- Race: the CARDIA Study, 1992-2006¶
| Change in fibrinogen (mg/dL) | ||||||||
|---|---|---|---|---|---|---|---|---|
| Men | Women | |||||||
| N | Blacks, Mean Δ (SE) | N | Whites, Mean Δ (SE) | N | Blacks, Mean Δ (SE) | N | Whites, Mean Δ (SE) | |
| Smoking | ||||||||
| Continued non-smoker (ref) | 277 | 69.8 (4.7) | 575 | 65.7 (3.2) | 471 | 77.2 (4.6) | 595 | 64.9 (3.4) |
| Became smoker | 16 | 121.5(16.8)∥ | 19 | 89.9 (14.9) | 17 | 55.0 (19.9) | 27 | 60.7 (14.2) |
| Stayed smoker | 100 | 70.2 (8.6) | 82 | 90.5 (11.8)∥ | 114 | 83.5 (11.8) | 55 | 102.4(13.2)∥† |
| Quit smoking | 35 | 59.7 (12.4) | 50 | 70.7 (11.3) | 54 | 81.5 (12.3) | 61 | 67.7 (12.6) |
| Obesity | ||||||||
| Continued non-obese (ref) | 231 | 63.7 (4.5) | 525 | 64.2 (2.8) | 264 | 52.3 (5.1) | 542 | 53.5 (3.1) |
| Became obese | 94 | 68.9 (6.9) | 108 | 88.5 (6.1)§ | 150 | 99.1 (6.4)§ | 102 | 114.2 (7.0)§ |
| Stayed obese | 94 | 90.6 (7.3)∥ | 83 | 81.1 (7.5)¶ | 229 | 97.8 (5.6)§ | 82 | 112.6 (8.8)§ |
| Became non-obese | 9 | 74.4 (23.0) | 10 | 42.7 (20.4) | 13 | 9.8 (22.0) | 12 | 20.6 (20.3) |
| Hypertension | ||||||||
| Continued non-hypertensive (ref) | 314 | 67.8 (3.8) | 611 | 67.2 (2.6) | 424 | 73.4 (4.0) | 676 | 66.5 (2.8) |
| Became hypertensive | 88 | 80.6 (7.3) | 85 | 77.9 (7.2) | 188 | 85.5 (6.0) | 56 | 89.7 (10.0)¶ |
| Stayed hypertensive | 26 | 77.4 (13.5) | 30 | 92.4 (12.0)¶ | 44 | 91.2 (12.4) | 6 | 11.9 (30.0) |
| Dyslipidemia | ||||||||
| Continued non-dyslipidemia (ref) | 166 | 56.1 (5.3) | 245 | 62.8 (4.2) | 251 | 71.6 (5.2) | 366 | 59.5 (4.0) |
| Became dyslipidemia | 91 | 81.6 (7.0)∥ | 143 | 75.8 (5.4)¶ | 98 | 91.1 (8.2)¶ | 80 | 84.7 (8.2)∥ |
| Stayed dyslipidemia | 171 | 79.8 (5.3)∥ | 338 | 71.6 (3.6)* | 307 | 79.2 (4.8) | 292 | 73.5 (4.6)¶ |
| Hyper TG | ||||||||
| Continued non-hyper TG (ref) | 386 | 72.6 (3.4) | 582 | 69.5 (2.7) | 631 | 79.5 (3.2) | 696 | 67.4 (2.8) |
| Became hyper TG | 25 | 62.2 (13.5) | 97 | 66.4 (6.6) | 19 | 46.3 (18.7) | 32 | 76.4 (13.1)* |
| Stayed hyper TG | 17 | 46.7 (17.0) | 47 | 74.9 (10.0) | 6 | 31.4 (33.7) | 10 | 63.9 (23.6) |
| Diabetes | ||||||||
| Continued non-diabetic (ref) | 382 | 70.0 (3.4) | 694 | 69.9 (2.4) | 590 | 80.1 (3.3) | 719 | 67.0 (2.7) |
| Became diabetic | 39 | 89.5 (11.0) | 27 | 58.1 (12.9)* | 59 | 62.1 (11.1) | 15 | 112.0(19.4)¶† |
| Stayed diabetic | 7 | 23.5 (26.1) | 5 | 65.0 (29.5) | 7 | 35.4 (31.3) | 4 | 43.7 (36.9) |
P<0.05
P<0.0l
P<0.001 compared to Blacks within gender.
P <0.001
P<0.01
P<0.05 compared with the referent category of risk factor within sex/race.
# Each risk factor represents a separate ANCOVA model. Ref=referent. SE=standard error.
** Model adjusted for (baseline07) age and fibrinogen level, family history of heart disease, education, baseline07 (physical activity, number of cigarettes/d, and all other risk factors shown in table simultaneously).
†† Changes in fibrinogen or risk factors were defined by changes over 13 years from baseline07 to year 20 (year 20 – year 7).
‡‡ Abbreviations: ref, reference; TG, triglyceridemia
Hypertension and smoking statuses followed the same general pattern as dyslipidemia and obesity. Quitting smoking was associated with small increase in follow-up fibrinogen levels, similar to persons who never smoked.
While becoming diabetic was associated with higher mean fibrinogen among black men (p=NS) and white women (p<0.001), the association was in the opposite direction (p=NS) for white men and black women (p<0.05 for white vs. black men; p<0.01 for white vs. black women).
Data were similar with initial adjustments for baseline07 age and fibrinogen levels only. Additionally, all study findings were unchanged when use of oral contraceptives and hormone therapy were included in the fully-adjusted models for women (data not shown).
Discussion
Major findings
To our knowledge, this is the first study that has examined long-term associations between fibrinogen and changes in traditional cardiovascular risk factors in a baseline07 young non-homogeneous population of black and white men and women. We found strong independent correlations between baseline07 levels/changes in many traditional cardiovascular risk factors and changes in mean fibrinogen levels over a 13-year follow-up period. Notably, smoking cessation and weight loss were associated with a smaller follow-up increase in fibrinogen levels that were the same as for individuals who remained non-smokers and non-obese through the 13 years. Of all the traditional risk factors studied, fibrinogen levels tracked the most with change in BMI and obesity status over 13 years, and correlated strongly with baseline07 BMI – even after full adjustment in both races and sexes.
Potential Clinical Implications
Many of the studies that have evaluated the link between fibrinogen and CVD have been either cross-sectional, or prospective. The few longitudinal studies examining this relationship have been limited in that they have been small, have studied few risk factors, have a short follow-up period, or comprised of homogeneous populations of participants.7-9 Here, we examined the long-term relationship between traditional CVD risk factors and fibrinogen levels (at baseline07 and follow-up). Our study shows that in these mostly CVD risk factor-free young adults with baseline07 mean age of 32 years, fibrinogen levels at middle age were affected by baseline07 and/or changes in risk factors acquired at a younger age.
A vital component of our study is noted by the lower levels of fibrinogen observed among individuals with improved risk factor profiles (i.e. those who quit smoking and/or became non-obese) at follow-up. This has significant implications in that it likely supports the notion of fibrinogen as a disease marker, rather than causative factor for disease (fibrinogen levels likely did not incite participants to lose weight or quit smoking, but rather the other way around). It is also possible that fibrinogen could be part of the mechanism whereby other risk factors exert their effects. Even so, our findings suggest that changes in risk factor profile led to eventual changes in fibrinogen levels, and not vice versa. This provides some insight into mechanisms of associations between fibrinogen – as a probable risk marker, rather than causative factor – and CVD in general.
The clinical implications of our study findings are that it calls for risk factor prevention (primordial prevention) and/or modification at younger ages, consequently leading to disease prevention at middle/older ages.
Obesity and Smoking
Similar to cross-sectional and prospective studies,3, 6, 15 we showed obesity status and BMI to have the strongest longitudinal associations with fibrinogen, while physical activity reversed this trend. Consequently, weight loss was associated with significantly smaller increase in fibrinogen relative to those who became or stayed obese over 13 years. Obesity is the chief component of the metabolic syndrome, and influences other components such as insulin resistance, elevated triglycerides, low HDL and elevated blood pressure. As such, all these other risk factors tracked well with fibrinogen in our study. It is proposed that obesity leads to low grade inflammation with increased proinflammatory cytokines such as leptin, interleukin-6 and tumor necrosis factor-α leading to platelet activation.3, 16 Our study also suggests fibrinogen as one of the inflammatory/prothrombotic agents whose levels are increased by obesity and weight gain. As in cross-sectional studies,3 the association between obesity/BMI and fibrinogen was strongest in women, particularly black women identified as the most obese at baseline07 in our univariate data.
Similar to fibrinogen levels for individuals who lost weight, the increase in fibrinogen for those who quit smoking was the same as never-smokers, and was significantly lower than that of persons who became or stayed smokers through the 13-year interim. Other studies have shown a rapidly declining CHD and CVD mortality with smoking cessation.17, 18 Fibrinogen is strongly associated with CHD and CVD mortality, and the decreased fibrinogen level observed in these individuals who quit smoking likely represents a CHD/CVD risk similar to those who never smoked.
Other Findings
Contrary to our expectations, there was a weak inverse association between blood glucose/diabetes status and changing fibrinogen levels in some of our participants. This is likely due to the fact that our diabetic participants took medications, which have been shown to modify fibrinogen levels.19
Study Strengths and Limitations
Our study represents a unique assessment of changes in fibrinogen relative to changes in CHD/CVD risk factors because it is a relatively large study carried out in a baseline07 young population with minimal risk factors and atherosclerosis. As such, there were few residual confounding factors (including medication use) at baseline07. In addition, our study included a standardized examination of risk factors and their progression with 13 years of follow-up. This allowed study participants to advance to middle-age by the end of the study. Furthermore, the demographics of our study population – balanced on age, sex, race, and education – allowed for comparative analysis by sex and racial makeup. It is noteworthy that we adjusted for potential bias caused by dropouts by conducting a sensitivity analysis using a multiple imputation technique (data not shown). The two sets of estimates (imputed and not-imputed) for relationship of fibrinogen to risk factors were consistent in direction and magnitude, thus providing assurance that our results were likely robust despite some loss of participants due to dropouts or missing data.
Conclusions
We found that in this young cohort with relatively low baseline prevalence of CVD, fibrinogen levels tracked longitudinally with many traditional cardiovascular risk factors (particularly BMI and obesity) through middle age, independent of covariates included in our models. Furthermore, we observed a strong inverse long-term relationship between weight loss/smoking cessation (major modifiable cardiovascular risk factors) and change in 13-year fibrinogen levels by middle age. Findings from our study help answer the question of fibrinogen as a disease risk marker (rather than a causative agent), thus providing some important insight into the mechanisms of disease and associations between fibrinogen and cardiovascular disease. Our study also has important public health implications relating to CVD reduction through risk factor modification. Future studies tracking longitudinal changes in fibrinogen levels with incident CHD/CVD as well as CHD/CVD mortality would be of great interest.
Highlights.
- We determined 13-year associations between CVD risk factors and fibrinogen levels
- Fibrinogen increased with development/worsening status of most CVD risk factors
- Increase in fibrinogen for persons who quit smoking was comparable to never-smokers
- Fibrinogen increase for those who lost weight was comparable to never-obese persons
- Our study provide insight into the role of fibrinogen as a CVD risk marker
Acknowledgements
This study was supported by grant HL-43758 and contracts NO1-HC-48049 and NO1-HC-95095 from the National Heart, Lung, and Blood Institute (NHLBI) and grant AG032136 from the National Institute on Aging, National Institutes of Health.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Disclosures: None
Bibliography
- 1.Herrick S, Blanc-Brude O, Gray A, Laurent G. Fibrinogen. Int J Biochem Cell Biol. 1999;31:741–746. doi: 10.1016/s1357-2725(99)00032-1. [DOI] [PubMed] [Google Scholar]
- 2.Fibrinogen Studies C, Danesh J, Lewington S, Thompson SG, et al. Plasma fibrinogen level and the risk of major cardiovascular diseases and nonvascular mortality: An individual participant meta-analysis. JAMA. 2005;294:1799–1809. doi: 10.1001/jama.294.14.1799. [erratum appears in JAMA. 2005 dec 14;294(22):2848]
- 3.Fibrinogen Studies C, Kaptoge S, White IR, Thompson SG, et al. Associations of plasma fibrinogen levels with established cardiovascular disease risk factors, inflammatory markers, and other characteristics: Individual participant meta-analysis of 154,211 adults in 31 prospective studies: The fibrinogen studies collaboration. American Journal of Epidemiology. 2007;166:867–879. doi: 10.1093/aje/kwm191. [DOI] [PubMed] [Google Scholar]
- 4.Green D, Foiles N, Chan C, Schreiner PJ, Liu K. Elevated fibrinogen levels and subsequent subclinical atherosclerosis: The CARDIA study. Atherosclerosis. 2009;202:623–631. doi: 10.1016/j.atherosclerosis.2008.05.039. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Green D, Chan C, Kang J, et al. Longitudinal assessment of fibrinogen in relation to subclinical cardiovascular disease: The CARDIA study. J Thromb Haemost. 2009;8:489–495. doi: 10.1111/j.1538-7836.2009.03727.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Chainani-Wu N, Weidner G, Purnell DM, et al. Changes in emerging cardiac biomarkers after an intensive lifestyle intervention. Am J Cardiol. 2011;108:498–507. doi: 10.1016/j.amjcard.2011.03.077. [DOI] [PubMed] [Google Scholar]
- 7.Folsom AR, Wu KK, Rasmussen M, Chambless LE, Aleksic N, Nieto FJ. Determinants of population changes in fibrinogen and factor vii over 6 years: The atherosclerosis risk in communities (ARIC) study. Arterioscler Thromb Vasc Biol. 2000;20:601–606. doi: 10.1161/01.atv.20.2.601. [DOI] [PubMed] [Google Scholar]
- 8.Meade TW, Imeson J, Stirling Y. Effects of changes in smoking and other characteristics on clotting factors and the risk of ischaemic heart disease. Lancet. 1987;2:986–988. doi: 10.1016/s0140-6736(87)92556-6. [DOI] [PubMed] [Google Scholar]
- 9.Ditschuneit HH, Flechtner-Mors M, Adler G. Fibrinogen in obesity before and after weight reduction. Obes Res. 1995;3:43–48. doi: 10.1002/j.1550-8528.1995.tb00119.x. [DOI] [PubMed] [Google Scholar]
- 10.Friedman GD, Cutter GR, Donahue RP, et al. CARDIA: Study design, recruitment, and some characteristics of the examined subjects. J Clin Epidemiol. 1988;41:1105–1116. doi: 10.1016/0895-4356(88)90080-7. [DOI] [PubMed] [Google Scholar]
- 11.Schmitz KH, Jacobs DR, Jr., Leon AS, Schreiner PJ, Sternfeld B. Physical activity and body weight: Associations over ten years in the CARDIA study. Coronary artery risk development in young adults. Int J Obes Relat Metab Disord. 2000;24:1475–1487. doi: 10.1038/sj.ijo.0801415. [DOI] [PubMed] [Google Scholar]
- 12.Reiner AP, Carty CL, Carlson CS, et al. Association between patterns of nucleotide variation across the three fibrinogen genes and plasma fibrinogen levels: The coronary artery risk development in young adults (CARDIA) study. Journal of thrombosis and haemostasis : JTH. 2006;4:1279–1287. doi: 10.1111/j.1538-7836.2006.01907.x. [DOI] [PubMed] [Google Scholar]
- 13.Lewis MR, Callas PW, Jenny NS, Tracy RP. Longitudinal stability of coagulation, fibrinolysis, and inflammation factors in stored plasma samples. Thrombosis and haemostasis. 2001;86:1495–1500. [PubMed] [Google Scholar]
- 14.Raghunathan TE, Lepkowski JM, Van Hoewyk J, Solenberger P. A multivariate technique for multiply imputing missing values using a sequence of regression models. Survey Methodology. 2001;27:85–95. [Google Scholar]
- 15.Rana JS, Arsenault BJ, Despres JP, et al. Inflammatory biomarkers, physical activity, waist circumference, and risk of future coronary heart disease in healthy men and women. European Heart Journal. 2011;32:336–344. doi: 10.1093/eurheartj/ehp010. [DOI] [PubMed] [Google Scholar]
- 16.Bhatt DL. What makes platelets angry: Diabetes, fibrinogen, obesity, and impaired response to antiplatelet therapy? J Am Coll Cardiol. 2008;52:1060–1061. doi: 10.1016/j.jacc.2008.06.027. [DOI] [PubMed] [Google Scholar]
- 17.Bertoni AG, Burke GL, Owusu JA, et al. Inflammation and the incidence of type 2 diabetes: The multi-ethnic study of atherosclerosis (MESA). Diabetes Care. 2010;33:804–810. doi: 10.2337/dc09-1679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Kenfield SA, Stampfer MJ, Rosner BA, Colditz GA. Smoking and smoking cessation in relation to mortality in women. JAMA : the journal of the American Medical Association. 2008;299:2037–2047. doi: 10.1001/jama.299.17.2037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Yilmaz H, Gursoy A, Sahin M, Guvener Demirag N. Comparison of insulin monotherapy and combination therapy with insulin and metformin or insulin and rosiglitazone or insulin and acarbose in type 2 diabetes. Acta diabetologica. 2007;44:187–192. doi: 10.1007/s00592-007-0004-9. [DOI] [PubMed] [Google Scholar]