| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Laboratory Medicine (M.K.H., K.G.R., P.K.M., A.C., R.C., G.C.), Pharmacy Department (K.K.S., F.P.), and Biostatistics and Clinical Epidemiology Service (R.W.), Warren G. Magnuson Clinical Center; the National Institute of Diabetes and Digestive and Kidney Diseases (M.C.S., T.E.); Office of the Director (A.P.); and the National Cancer Institute (D.M.B.), The National Institutes of Health, Bethesda, Maryland 20892
Address all correspondence and requests for reprints to: McDonald Horne, M.D., Department of Laboratory Medicine, National Institutes of Health, Building 10, Room 2C-306, Bethesda, Maryland 20892. E-mail: MHorne{at}mail.cc.nih.gov.
| Abstract |
|---|
|
|
|---|
We obtained blood samples from 14 thyroid cancer patients while on THST and after they had become hypothyroid for radioiodine whole-body scanning and therapy. Prothrombin fragment 1 + 2, fibrinogen, factor VIII, antithrombin, tissue plasminogen activator antigen (tPA), plasminogen activator inhibitor 1 (PAI-1), PAI-1/tPA, and C-reactive protein were significantly (P < 0.05) higher in the hyper- than in the hypothyroid state, whereas protein C and plasmin-antiplasmin complexes were significantly lower during the hyperthyroid period. When the 10 female patients were hyperthyroid, their levels of prothrombin fragment 1 + 2, fibrinogen, protein S, antithrombin, tPA, PAI-1, and PAI-1/tPA were significantly higher (P
0.05) than in healthy female controls, whereas when the female patients were hypothyroid, their antithrombin and plasmin-antiplasmin were lower and their protein S was higher than in controls. Factor II, plasminogen, and D-dimer were not significantly affected by the thyroid status in either assessment.
In conclusion, we found evidence that the majority of patients treated with THST have a prothrombotic profile.
| Introduction |
|---|
|
|
|---|
|
These apparent inconsistencies and others could be explained if the observations were related to factors other than the direct effects of thyroid hormone. Because all of the reported patients had various forms of thyroiditis, it is possible that their prothrombotic state was actually related to inflammation or autoimmunity rather than hormonal mechanisms. It has even been suggested that hypercoagulability associated with thyroiditis is caused by tissue factor released from the inflamed gland (12, 16).
Recently, we had an opportunity to tease out the effect of thyroid hormone per se in a group of patients with thyroid cancer whose THST was temporarily discontinued, in order to reevaluate them with radioiodine scans. We obtained blood samples from these patients while they were chronically hyperthyroid on L-T4 and again after their THST had been stopped and they had become chemically hypothyroid. We assayed these samples for procoagulant, anticoagulant, and fibrinolytic factors.
| Patients and Methods |
|---|
|
|
|---|
Patients were eligible if they were at least 18 yr old and had had a total thyroidectomy for thyroid cancer followed by 131I ablative therapy and THST. They were excluded if they had been acutely ill within 2 wk, had laboratory evidence of renal or hepatic dysfunction, were taking any medication known to affect the planned blood testing (e.g. anticoagulants, nonsteroidal antiinflammatory agents, estrogens), or had a history of thrombosis or abnormal bleeding. The study was approved by the institutional review boards of the National Institute of Diabetes and Digestive and Kidney Diseases and the National Institute of Arthritis and Musculoskeletal and Skin Diseases. All patients gave written informed consent to participate.
Under a separate protocol approved by the institutional review board of the National Institute of Allergy and Infectious Diseases, blood was obtained by convenience sampling from normal control subjects in the same age range (2665 yr old) as the thyroid cancer patients. Control subjects also gave written informed consent to be donors.
Blood sampling
Venous blood samples were obtained in the morning after a 12-h fast. Patients were instructed to avoid alcohol and strenuous exercise for 48 h before blood collection. While the patients were receiving THST, blood was drawn for thyroid function tests and for the assays described below. Only patients who were biochemically hyperthyroid (defined as a TSH < 0.4 mU/liter) at that time were included in subsequent analysis. Study participants subsequently followed the standard National Institute of Diabetes and Digestive and Kidney Diseases protocol to prepare them for radioiodine scanning. L-T4 was discontinued and replaced with T3 (5075 µg/d) for 4 wk, after which all thyroid hormone was stopped. Over the next 2 wk, the patients followed an iodine-restricted diet and became progressively hypothyroid. Venous blood samples were obtained before the tracer dose of radioiodine for thyroid function tests and for the assays described below. After this, the patients underwent thyroid scanning.
Laboratory assays
Plasma anticoagulated with 0.11 mol/liter sodium citrate was assayed for fibrinogen, fVIII, factor II, antithrombin, protein C, protein S, and plasminogen using a STA COAG instrument manufactured by Diagnostica Stago (Parsippany, NJ). Diagnostica Stago was also the source for ELISAs to measure PAI-1 antigen and D-dimer. ELISAs for prothrombin fragment 1 + 2 (F1 + 2) and plasmin-antiplasmin (PAP) complexes were obtained from Dade Behring (Marburg, Germany). Tissue plasminogen activator antigen (tPA) was assayed with an ELISA from Biopool (Bray, Ireland). C-reactive protein (CRP) was measured with a high-sensitivity assay, Immulite 2000 High Sensitivity CRP (Diagnostic Products Corp., Los Angeles, CA). FT4 and TSH were measured with an Access analyzer (Sanofi Diagnostics Pasteur, Chaska, MN).
Statistical analyses
Differences between each patients hyper- and hypothyroid periods were analyzed using the Wilcoxon nonparametric signed rank test for paired samples. For all parameters, data from 14 patients were analyzed. Data from the control groups for the 12 different hemostatic parameters (CRP data from normal controls were not available) and from the patients during their hyper- and hypothyroid periods were initially compared separately using the Wilcoxon test (control vs. hypothyroid and control vs. hyperthyroid). To account for the inherent multiple comparisons in these analyses, we used the Bonferroni adjustment of doubling each nominal P value (because there are two related comparisons for each parameter). These are the primary P values reported for these control comparisons. In addition, we note what these P values are calculated using Holms method of adjustment (17) whenever it is necessary also to take into account the fact that we analyzed 12 separate but somewhat related parameters. All P values are two-sided with
set at 0.05.
| Results |
|---|
|
|
|---|
Nineteen patients were enrolled in the study. Five were excluded because they were not chemically hyperthyroid on THST. The remaining 14 included 10 women (ages 3776 yr, median age of 48 yr) and four men (30, 34, 61, and 63 yr old). Six of these 14 (five women and one man) had no evidence of carcinoma: no detectable thyroglobulin and negative radioiodine scans. Four (three women and one man) had minimal residual cancer defined as a thyroglobulin less than 10 mg/liter and a negative radioiodine scan or minimal uptake in the thyroid bed. Three subjects (one woman and two men) had thyroglobulins of 7, 16, and 29 mg/liter with small radioiodine uptake in the mediastinum or lung. One woman had advanced carcinoma in the thyroid bed with a thyroglobulin of 380 mg/liter.
The median L-T4 dose of the 14 subjects was 2.4 µg/kg·d (range, 1.64.2). During THST, the median TSH was 0.07 mU/liter (range, 0.020.2 mU/liter) with normal reference limits of 0.434.6 mU/liter. The median FT4 level was 2.0 ng/liter [SI, 25.8 pmol/liter; range, 1.63.3 (SI, 20.642.6)] with normal reference limits of 0.91.6 ng/dl (SI, 11.620.6 pmol/liter). After discontinuing THST, the patients median TSH rose to 78 mU/ml (range, 45165), and their FT4 fell to less than or equal to 0.1 ng/dl [SI, 1.3 pmol/liter; range,
0.10.3 (SI range,
1.33.9 ng/dl)].
Data from the hyper- and hypothyroid periods of the patients
Test results from all 14 patients and P values for the differences between each patients hyper- and hypothyroid periods are shown in Fig. 2
. By nonparametric analysis, F1 + 2, fibrinogen, fVIII, antithrombin, tPA, PAI-1, PAI-1/tPA, and CRP were significantly (P < 0.05) higher during THST than later when the patients were hypothyroid. In contrast, protein C and PAP were significantly lower during the hyperthyroid period. The differences in prothrombin (fII), protein S, plasminogen, and D-dimer were not significant. In all of the tests, there was extensive overlap of the results observed in patients without evidence of disease with the results from patients with metastases or residual tumor in the thyroid bed.
|
Comparison of female patient data with data from female controls
Table 1
includes the results of only the 10 female patients and the results from healthy control women. The four male patients were omitted from this analysis because our control group of men was not well age-matched for the male patients, two of whom were over 60 yr old.
|
| Discussion |
|---|
|
|
|---|
The mechanisms responsible for this shift are suggested by the higher fVIII and PAI-1 and lower protein C during the hyperthyroid period. FVIII promotes thrombin production, whereas protein C inhibits it (Fig. 1
). Increased fVIII and low protein C are both associated with hypercoagulability (18, 19), and so is elevated fibrinogen, which was higher when our patients were hyperthyroid (20). PAI-1 suppresses fibrinolysis (Fig. 2
) by inhibiting tPA. Although our patients tPA levels were also higher when they were hyperthyroid, their PAI-1 levels were disproportionately greater, making the ratio of PAI-1 to tPA significantly higher during their hyperthyroid period (21). The net effect, therefore, was a shift toward reduced plasminogen activation during hyperthyroidism, consistent with the lower PAP observed (Fig. 2
). However, D-dimer levels, which reflect plasmin activity, were not significantly lower during the hyperthyroid period.
Although CRP was measured to rule out subclinical inflammation, we found that the CRP levels of the patients tended to be above normal, especially during THST, suggesting that low-grade inflammation might in fact be contributing to the other changes we observed (Fig. 2
). However, the CRP levels did not correlate with the concentrations of factors known to be acute phase reactants, i.e. PAI-1, fibrinogen, fVIII, fII, and plasminogen (22, 23). Therefore, it appears that the effects of THST on the parameters of coagulation and fibrinolysis were independent of its effect on CRP and not mediated by an inflammatory response.
In patients with thyrotoxicosis, the most commonly reported thromboembolic events are emboli related to atrial fibrillation (1, 2, 3). Our data suggest that their thrombotic predisposition may not only result from their arrhythmias but may also be augmented by a prothrombotic biochemical milieu stimulated by their high levels of thyroid hormone. This would explain the impression that emboli are more commonly associated with thyrotoxic atrial fibrillation than with nonthyrotoxic arrhythmias (3).
In contrast to patients with pathological hyperthyroidism, patients on THST are maintained in a less toxic state but potentially for a much longer time. Therefore, the prothrombotic effects of their hyperthyroidism might chronically interact with other coincident diseases, such as coronary atherosclerosis, which leads to myocardial infarction when an acute thrombosis develops at a ruptured cholesterol plaque (24). Most of our patients on THST not only had elevations of PAI-1, which is associated with a risk of coronary occlusion, but also elevations of CRP 1.0 mg/dl or more (median, 2.9 mg/dl; range, 1.012.8), which are associated with a moderate to high risk for a future cardiovascular event (25, 26). These data at least support an aggressive use of primary prophylaxis for coronary disease in patients taking THST (27).
Elevated CRP and PAI-1 may also be associated with an increased risk for deep venous thrombosis, but this is not as clear an association as with coronary artery thrombosis (28, 29). However, the statistically significant elevation of F1 + 2 when the patients were taking THST reflects heightened thrombin production and implies that these patients would be predisposed to deep venous thrombosis whenever they encounter circumstances that carry additional risk, such as prolonged immobility or surgery. Clinical and epidemiological studies are needed, however, to assess the benefit of antithrombotic prophylaxis for patients taking long-term thyroid suppression therapy (30).
| Footnotes |
|---|
Received March 23, 2004.
Accepted May 21, 2004.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Squizzato, E. Romualdi, H. R. Buller, and V. E. A. Gerdes Thyroid Dysfunction and Effects on Coagulation and Fibrinolysis: A Systematic Review J. Clin. Endocrinol. Metab., July 1, 2007; 92(7): 2415 - 2420. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Pacini, M. Schlumberger, H. Dralle, R. Elisei, J. W A Smit, W. Wiersinga, and the European Thyroid Cancer Taskforce European consensus for the management of patients with differentiated thyroid carcinoma of the follicular epithelium. Eur. J. Endocrinol., June 1, 2006; 154(6): 787 - 803. [Full Text] [PDF] |
||||
![]() |
M. Dorr, D. M. Robinson, H. Wallaschofski, C. Schwahn, U. John, S. B. Felix, and H. Volzke Low Serum Thyrotropin Is Associated with High Plasma Fibrinogen J. Clin. Endocrinol. Metab., February 1, 2006; 91(2): 530 - 534. [Abstract] [Full Text] [PDF] |
||||
![]() |
B Jarzab, D Handkiewicz-Junak, and J Wloch Juvenile differentiated thyroid carcinoma and the role of radioiodine in its treatment: a qualitative review Endocr. Relat. Cancer, December 1, 2005; 12(4): 773 - 803. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |