Relationship of Fibrinogen Levels and Hemostatic Abnormalities With Organ Damage in Hypertension
- From the Hypertension Unit, Chair of Internal Medicine, Department of Experimental and Clinical Pathology and Medicine, University of Udine School of Medicine, Udine, Italy.
- Correspondence to Leonardo A. Sechi, MD, Hypertension Unit, Internal Medicine, University of Udine, School of Medicine, Ospedale Civile, Padiglione Medicine, 33100 Udine, Italy. E-mail Sechi{at}uniud.it
Abstract
Abstract—Elevated plasma levels of fibrinogen and activated coagulation pathways are risk factors of cardiovascular disease in the general population. In a cross-sectional study of a case series, we investigated the relationship between fibrinogen and hemostatic markers with target-organ damage (TOD) in patients with arterial hypertension. Prothrombin time, partial thromboplastin time, fibrinogen, fibrin D-dimer, prothrombin fragment 1+2 (F1+2), and antithrombin III were measured in 352 untreated patients with mild to moderate essential hypertension and 92 normotensive controls. Staging of TOD was assessed according to W.H.O. guidelines by clinical evaluation and laboratory tests including measurements of creatinine clearance, proteinuria, ophthalmoscopy, electrocardiography, echocardiography, and ultrasound examination of major arteries. F1+2 concentrations were significantly greater in hypertensive patients than normotensive controls and were positively correlated with blood pressure. Age, blood pressure levels, duration of hypertension, smoking, HDL-cholesterol, triglycerides, and plasma fibrinogen, fibrin D-dimer, and F1+2 levels were significantly related to the presence and severity of TOD in univariate analysis. Plasma fibrinogen and D-dimer levels were related to organ damage independent of age, blood pressure, duration of hypertension, and smoking status. Separate analysis indicated significant association of fibrinogen and D-dimer levels with cardiac, cerebrovascular, peripheral vascular, and renal damage. In conclusion, elevated plasma levels of fibrinogen and a prothrombotic state are associated with the presence and severity of TOD in patients with essential hypertension and may contribute to the development of atherosclerotic disease in these patients.
- antithrombin III
- coagulation
- fibrin D-dimer
- fibrinogen
- hypertension, essential
- prothrombin fragment 1+2
Introduction
Frequency of cardiovascular events is associated in a continuum manner with the level of blood pressure.1 Blood pressure, however, is not the only determinant of cardiovascular damage, and the propensity of hypertensive patients to develop target-organ damage (TOD) is markedly influenced by coexisting risk factors such as age, gender, smoking, obesity, diabetes, and dyslipidemia. In addition to these factors, the hemostatic system is directly involved in the atherosclerotic process2 and may affect the impact of hypertension on the development of TOD and thereby on cardiovascular morbidity.3 4 Because arterial disease is the major underlying factor leading to most clinically relevant cardiovascular events and these events are usually due to formation of a thrombus at the site of an atherosclerotic plaque, research has concentrated on the state of the coagulation pathways. Fibrinogen is both a coagulation factor and an acute-phase reactant that has been identified as a major independent risk factor for coronary artery,5 6 cerebrovascular,6 and peripheral vascular7 disease in the general population. Moreover, the results of the Leigh study, in which hypertensive patients with plasma fibrinogen above 3.5 g/L had a 12-fold greater coronary risk than those with fibrinogen below 2.9 g/L, suggest that fibrinogen levels may affect prognosis in hypertension.8 Similar to fibrinogen, elevations of D-dimer, the principle breakdown fragment of fibrin and a reliable indicator of the overall state of activation of the coagulation pathways,9 are associated with increased risk of future myocardial infarction,10 stroke,11 and peripheral vascular disease.12
These and other clinical and laboratory observations have led to the hypothesis that hypertension per se may confer a hypercoagulable state13 14 that might be related to development of TOD and long-term prognosis.15 To investigate the possible role of fibrinogen and the coagulation system as risk factors for the development of TOD in hypertension, we have evaluated hemostatic markers in a large group of untreated patients with essential hypertension and different degrees of target organ involvement.
Methods
Patients and Study Design
Three-hundred-fifty-two patients (age: 54±12 years; 186 men and 166 women) with mild to moderate essential hypertension who were referred to our hypertension clinic from January 1995 to June 1998 were included in a cross-sectional study. Blood pressure was measured by a mercury sphygmomanometer in standard conditions, and diagnosis of hypertension was established by standard criteria.16 The patients seen at our clinic are white, include individuals with all grades of hypertension living in northeast Italy, and are representative of hypertensive patients in this geographic area.17 Exclusion criteria were: age >75 years; body mass index >32 kg/m2; diabetes mellitus; pregnancy; renal failure with creatinine clearance of <30 mL · min−1 · 1.73 m2−1 body surface area; urinary protein excretion >3.0 g/d; or presence of other diseases or treatments that might interfere with the coagulation system. Patients with severe obesity,18 diabetes,19 and advanced renal failure20 were excluded because these conditions are known to affect hemostatic variables. We also excluded hypertensive subjects with acute illness and recent (<6 months) myocardial infarction, unstable angina, or stroke. The number of patients was planned to have a power of more than 95% in detecting a 20% difference (55% of standard deviation) between patients with and without TOD with an α-value of 0.05. Secondary causes of hypertension were excluded on the basis of exhaustive laboratory testing.21 Staging of hypertension was performed, according to the W.H.O./I.S.H. guidelines, by definition of the extent of TOD (stage I: no signs of organic changes; stage II: initial involvement in at least one target organ; stage III: advanced involvement in at least one target organ).16 TOD was assessed by history, physical examination, and laboratory tests including measurements of creatinine clearance, 24-hour proteinuria, ophthalmoscopy, electrocardiography, echocardiography, and ultrasound examination of aorta, carotid, and iliac arteries as performed in previous studies to which readers are referred for details.22 Echocardiographic, vascular ultrasonographic, and ophthalmoscopic examination were performed by independent investigators who, at the time of examination, were unaware of the patients’ other clinical characteristics.
Essential hypertensive patients were compared with 92 normotensive subjects (age: 52±18 years; 48 men and 44 women). These subjects were selected from the general population of the same geographic area as the hypertensive patients by frequency matching after specification of inclusion criteria to avoid age and gender as potential confounding variables. Controls were not taking any regular medications and did not have any concomitant disease. Two-hundred-twenty-five (64%) of the hypertensive patients included in the study were treated with antihypertensive drugs (monotherapy, 44%; multiple-drug therapy 56%; angiotensin converting enzyme inhibitors, 43%; calcium antagonists, 40%; diuretics, 37%; β-blockers, 27%; angiotensin II receptor antagonists, 16%; α-blockers, 8%) and were withdrawn from treatment a minimum of 1 week before measurement of blood parameters. β-blockers, lipophilic calcium antagonists, angiotensin converting enzyme inhibitors, and angiotensin I receptor antagonists were withdrawn for 3 weeks. At the time of the study, patients were allowed to maintain their usual unrestricted diet. The study was approved by the Ethical Committee of the University of Udine.
Biochemical Analysis
Subjects fasted for 12 to 14 hours, after which blood was drawn between 8:00 and 9:00 am into silicone-treated glass tubes where it was mixed with 10% of its volume of 0.1 mol/L trisodium citrate. The blood was immediately centrifuged at 1700g for 20 minutes at 4°C and the plasma was frozen at −80°C until assayed. Plasma fibrinogen was measured by a functional assay23 in an automatic coagulometer autoanalyzer (Instrumentation Laboratory,; normal values: 2.4 to 3.5 g/L). Prothrombin fragment 1+2 (F1+2) plasma levels were evaluated by ELISA according to the method of Peltzer et al24 (Boehringwerke; normal values: 0.44 to 1.11 nmol · L−1 · L−1). D-dimer was measured immunoenzymatically according to the method of Rylatt et al25 (Boehringer; normal values: 4.0 to 43.0 ng/mL). Antithrombin III was determined by a functional chromogenic assay (Instrumentation Laboratory; normal values: 80% to 120%). Inter- and intra-assay coefficients of variation for fibrinogen, F1+2, D-dimer, and antithrombin III were below 8%. Total, LDL, and HDL cholesterol and triglycerides were assayed enzymatically by an automated method as reported previously.22
Statistical Analysis
Normally distributed variables are presented as means (SD). Medians and interquartile ranges have been used for variables with skewed distribution. The Student’s t test was used for comparisons between 2 groups and ANOVA was used for comparisons of >2 groups. The Mann-Whitney test was used for comparisons of variables with skewed distribution between 2 groups and the Kruskal-Wallis test was used for comparisons of >2 groups. The Tukey-Kramer post-hoc analysis was performed to test pairwise comparisons. Hypertensive patients with different degree of TOD were then compared with the stepwise multiple-discriminant analysis (GB Stat 6.5, Dynamic Microsystems Inc). The relationship between different variables was examined by linear regression analysis, and the correlation was expressed by the Pearson’s correlation coefficient. In this analysis, logarithmic transformation was performed for variables with skewed distribution. Probability value <0.05 were considered to indicate statistical significance.
Results
The clinical, metabolic, and coagulation measurements of control and hypertensive subjects are shown in Table 1⇓. Hypertensive patients had greater body mass index and triglyceride levels and lower HDL cholesterol than controls. Plasma F1+2 level was significantly greater in hypertensive patients than normotensive controls (Figure 1⇓), whereas no significant difference was observed in platelets number, prothrombin time, partial thromboplastin time, and plasma fibrinogen, fibrin D-dimer, and antithrombin III levels.
Scatter plots of fibrinogen, D-dimer, F1+2, and antithrombin III in hypertensive patients and normotensive controls.
Clinical, Metabolic, and Coagulation Measurements of the Study Subjects
In both stage II and stage III hypertensive patients, the signs and symptoms of TOD were variously combined (Table 2⇓). Table 3⇓ shows the measurements of hypertensive patients subdivided according to the stage of the disease. Platelet counts, partial thromboplastin time, and prothrombin time were comparable in hypertensive patients with different degree of TOD. Fibrinogen and D-dimer levels were significantly greater among hypertensive patients with TOD (stage II and stage III combined: 185/307/360/421/936 g/L and 2.8/15.7/31.8/58.0/402.3 ng/mL, respectively) than among those without TOD (stage I: 97/277/326/380/815 g/L, P<0.001; 1.9/12.5/22.1/40.7/180.5 ng/mL, P=0.002). Antithrombin III activity was significantly less in hypertensive patients with TOD (62/88/96/105/126%) than in those without TOD (72/92/101/109/144%, P=0.007). Fibrinogen and D-dimer levels were also significantly greater (both P<0.001) in patients with advanced TOD (stage III) than in patients with initial TOD (stage II), which indicates that higher concentrations of these parameters were associated not only with the presence of TOD but also with its severity. The relationship between fibrinogen, D-dimer and target-organ damage was independent of the type of anti-hypertensive drug the patients were taking.
Distribution of TOD in Hypertensive Patients
Clinical, Metabolic, and Coagulation Measurements of the Hypertensive Patients
Univariate comparison indicated 7 more variables that, in addition to fibrinogen, D-dimer, F1+2, and antithrombin III, differed significantly among the hypertensive subgroups: age, systolic and diastolic blood pressure, duration of hypertension, smoking status, HDL-cholesterol, and triglycerides (Table 3⇑). Stepwise multiple-discriminant analysis was performed to evaluate the independence of these factors in predicting TOD among patients with hypertension (Table 4⇓), which indicates that systolic blood pressure, plasma fibrinogen, duration of hypertension, and the smoking status were independently associated with the presence of TOD. Plasma fibrinogen and D-dimer levels were also independently associated with the presence of advanced TOD, as demonstrated by stepwise multiple-discriminant analysis performed comparing stage III hypertensive subjects to stage I and stage II subjects (Table 4⇓). Analysis of coagulation parameters in patients with different types of organ damage indicated that elevated levels of fibrinogen and D-dimer were independently associated with the presence of coronary artery, cerebrovascular, peripheral vascular, and renal disease (Table 5⇓).
Stepwise Multiple-Discriminant Analysis of Risk Factors for TOD in Hypertensive Patients (n=352)
Coagulation Measurements of the Hypertensive Patients with Different Types of TOD
Log fibrinogen, log D-dimer, and log F1+2 concentrations were positively correlated (Figure 2⇓). Age was directly correlated with log fibrinogen (r=0.341; P<0.001), log D-dimer (r=0.242; P<0.001), and log F1+2 (r=0.141; P=0.011) and inversely correlated with log antithrombin III (r=−0.206; P<0.001). Log D-dimer (r=0.178; P=0.002) and log F1+2 (r=0.125; P=0.028) were also directly correlated with duration of hypertension. Systolic blood pressure was directly correlated with log F1+2 (r=0.164; P=0.002), but not with log fibrinogen and log D-dimer. No correlations were found between hemostatic variables, plasma glucose, and plasma lipids.
Linear regression analysis between log fibrinogen, log D-dimer, and log F1+2 in hypertensive patients (n=352). The 3 parameters were directly and significantly correlated.
No difference in fibrinogen, D-dimer, and F1+2 levels was observed between hypertensive patients who, at inclusion, were taking antihypertensive drugs and untreated patients. Similarly, comparison of fibrinogen, D-dimer, and F1+2 levels in hypertensive patients treated with antihypertensive drugs before and after withdrawal of such treatment did not reveal any significant difference.
Discussion
This study shows that high plasma concentrations of fibrinogen, fibrin D-dimer, and F1+2 are associated with TOD in patients with essential hypertension. In addition to elevated fibrinogen, D-dimer, and F1+2 levels, the hypertensive patients with TOD have higher blood pressure levels and longer duration of hypertension, are older, smoke more frequently, and have lower concentrations of HDL-cholesterol and higher concentrations of triglycerides than those without TOD. Multivariate analysis demonstrates that fibrinogen and D-dimer are independent predictors of the severity of TOD, which indicates that the occurrence of TOD in hypertension is under the direct influence of the coagulation system.
Several prospective investigations have demonstrated a direct relationship between plasma fibrinogen concentrations and the risk of atherosclerotic cardiovascular disease3 4 5 6 7 8 11 and many recent studies26 27 28 29 30 have compared fibrinogen levels in hypertensive and normotensive subjects. These studies have yielded conflicting results, probably related to the limitations imposed by small numbers of patients, difference in blood pressure values of patients included in these studies, interference of antihypertensive treatments, and inclusion of patients with diabetes and severe obesity. Although we did not observe significant differences in fibrinogen levels between hypertensive patients and normotensive controls, our study clearly demonstrates that there is a strong and independent association between fibrinogen and the presence and severity of hypertension-related damage in different target organs. Similar evidence was presented by Lip et al who demonstrated that fibrinogen levels are related to the left ventricular mass in hypertensive patients.28 These data suggest a pathogenic role of fibrinogen in the development of TOD in hypertensive patients that could be related to the atherogenic31 and thrombotic32 actions of this protein. It is unlikely that the association between plasma fibrinogen and TOD in these patients is merely due to chance, because the strength of the association, the dose-response relationship, and the independent association demonstrated by multivariate analysis suggest causality and reduce the likelihood of confounding variables. Because fibrinogen is an acute phase reactant, it is also possible that vascular lesions with inflammatory components could be responsible for the association between fibrinogen and TOD. Against this possibility is the observation of comparable values of erythrocyte sedimentation rate and reactive C-protein in hypertensive patients with different degrees of TOD.
Fibrin D-dimer is the principle breakdown fragment of fibrin9 and a good biochemical marker of an existing thrombophilia. Prospective epidemiological studies have demonstrated that D-dimer is independently associated with risk of myocardial infarction,10 cerebrovascular accidents,11 and peripheral arterial disease.12 Two studies28 33 have reported elevated D-dimer levels in hypertensive patients, but this observation was not confirmed in a subsequent study.34 As for fibrinogen, small sample size, different blood pressure levels, interference of antihypertensive treatments, and inclusion of unselected patients are possible explanations for inconsistent results. The present study was performed in a large group of untreated hypertensive patients who were practically devoid of other factors known to affect coagulation, without evidence of significant differences in plasma D-dimer concentrations in comparison to normotensive controls. Nonetheless, D-dimer concentrations were significantly related to the severity of TOD in hypertensive patients. This relationship was independent of confounders such as age, duration of hypertension, and smoking status and suggests a role of the hemostatic system in the occurrence of organ damage in these patients. Significant association between plasma F1+2 levels and TOD, which indicates significant activation of the coagulation pathways,35 36 provides further support for this possibility. In the present study, F1+2 was the only hemostatic parameter directly correlated with blood pressure, which suggests that elevation in F1+2 might occur as the result of a hypertension-induced blood vessel damage.
Several studies have been conducted to assess the function of the hemostatic system in hypertensive patients. Increased platelet activation (reviewed in 37 ), increased activity of the coagulation system (reviewed in 15 ), and decreased function of the fibrinolytic system29 have been consistently reported in comparison to normotensive subjects, which suggests the existence of a prothrombotic state related to hypertension.13 This possibility is confirmed by our findings on F1+2. On the other hand, other factors related to coagulation, such as elevated fibrinogen and elevated D-dimer, may act as risk factors of TOD in hypertensive patients. Antihypertensive drugs with benefits in decreasing these factors would be likely to decrease the occurrence of cardiovascular events in these patients.
Some potential limitations of this study should be discussed. First, the cross-sectional design does not permit the establishment of a clear evidence of a causal relationship between fibrinogen, coagulation abnormalities, and TOD in hypertensive patients. Second, the strength of some of the correlations found in this study, such as those between hemostatic parameters, age, and duration of hypertension, was modest and therefore should be considered with caution. Third, the association between hemostatic parameters and TOD shown in this study does not rule out the possibility that abnormalities of coagulation are the consequence and not the cause of TOD. Fourth, the use of a clinic sample might limit the generalizability of the conclusions of this study to the general population because of a bias in the referral of patients to the source of care.
The findings of this study have some important implications for the identification of organ damage in patients with hypertension and for the management and prognosis of these patients. The strength of the relationship between TOD and plasma fibrinogen and D-dimer levels implies that these parameters might be useful in the diagnostic work-up of hypertensive patients to identify those who have likely developed or will develop TOD. Although our study demonstrates only an association between fibrinogen, D-dimer, and TOD, it is reasonable to assume that the coagulation system plays a primary role in the occurrence of organ damage in hypertensive subjects. Elevated plasma fibrinogen and D-dimer levels in hypertensive patients may be useful in guiding the physician toward more exhaustive diagnostic procedures to detect and follow-up TOD and toward a better control of blood pressure and other risk factors to prevent cardiovascular morbidity. Correction of the prothrombotic state by use of drugs that decrease the activity of the hemostatic system could reduce the incidence of atherosclerotic events in these patients.
Acknowledgments
This research was supported by CNR (Consiglio Nazionale delle Ricerche) Grants (92.01096.CT04; 94.084231.CT 04; 94.02438.CT04), a Fulbright Senior Fellowship, and a Ferrero Foundation Grant to LAS.
- Received April 20, 2000.
- Revision received May 11, 2000.
- Accepted June 10, 2000.
