Abstract

BACKGROUND AND OBJECTIVES: Thalassemic patients have an increased risk for thromboembolic complications. To determine if this might be due to a deficiency in protein C, we investigated the status of the protein C anticoagulant pathway in thalassemia major patients and its relationship to the hypercoagulable state.
PATIENTS AND METHODS: Fifty patients with beta-thalassemia major (30 non-splenectomized and 20 splenectomized) and 20 healthy children as a control group were tested for levels of serum ferritin, liver enzymes, serum albumin, fibrinogen, protein C and protein S, thrombin antithrombin complex (TAT) and D-dimer.
RESULTS: Thalassemic patients had lower levels of protein C and S and higher levels of D-dimer and TAT than the control group. These findings were more obvious in splenectomized patients and in those with infrequent blood transfusion.
CONCLUSIONS: Protein C plays a major role in the hypercoagulable state in thalassemic patients. These findings raise the issue as to whether it would be cost-beneficial to recommend prophylactic antithrombotic therapy in high-risk thalassemic patients. A wider prospective study is necessary to delineate under which circumstances therapy might be needed, and at what level of protein C deficiency to start prophylactic antithrombotic therapy.

Protein C is a vitamin K-dependent serine protease and naturally occurring anticoagulant that plays a role in the regulation of hemostasis by inactivating factors Va and VIIIa in the coagulation cascade. Human protein C circulates as a 2-chain zymogen, but functions at the endothelial and platelet surface following conversion to activated protein C by limited proteolysis with thrombin in complex with the cell surface membrane protein, thrombomodulin.¹
Thalassemias are inherited hemoglobinopathies characterized by a structural hemoglobin defect. These hereditary diseases have significant morbidity and mortality and affect individuals of African-American heritage, as well as those of Mediterranean, Middle Eastern, and Southeast Asian descent.²
 Despite the difficulties associated with treatment, standards of care for thalassemic patients have improved in recent years, resulting in almost a doubling of the average life expectancy. As a consequence, additional previously undescribed complications are now being recognized. In particular, profound hemostatic changes have been observed in patients with beta-thalassemia major, especially splenectomized cases. The presence of a higher than normal incidence of thromboembolic events and the existence of prothrombotic hemostatic anomalies in the majority of the patients, even from a very young age, have led to the recognition of the existence of a chronic hypercoagulable state in thalassemic patients.³
 Several etiologic factors may play a role in the pathogenesis of the hypercoagulable state in thalassemia such as platelet activation and increased circulating aggregates,^4,5 shortened platelet survival,⁶ increased urinary excretion of thromboxane A2 and prostacyclin metabolites,⁷ decreased levels of naturally occurring anticoagulants such as protein C and protein S, and an elevated plasma level of thrombin-antithrombin complex have also been reported.^8,9 A mechanism involving the loss of membrane phospholipid asymmetry in thalassemic red or erythroid cells, with the exposure of procoagulant phosphatidylserine in the outer leaflet of the red cell membrane has also been proposed.¹⁰
 The lower protein C levels in thalassemia are either due to decreased production or increased consumption. Therefore, a coagulation imbalance exists in patients with thalassemia, which in turn may be responsible for the adverse clinical effects observed in those patients.² We investigated the status of the protein C anticoagulant pathway in thalassemic patients and its relationship to the hypercoagulable state in thalassemia major.

Patients and Methods

In this case-control study we included 50 patients with beta-thalassemia major (29 males and 21 females) and 20 healthy children matched by sex and age, who were selected as a control group. Subjects were classified into group NS (30 patients with thalassemia major and non-splenectomized), group S (20 splenectomized thalassemia major patients), and group C (20 healthy subjects as a control group). Subjects were excluded for liver enzymes more than 5 times normal range, hepatitis C and/or B positivity, co-morbid conditions known to alter hemostasis, recent infection, or recent administration of aspirin or drugs that affect haemostatic function. None of our patients had a history of thrombosis including thrombophlebitis due to peripheral angiocatheter insertion. For every subject, we took a complete history, performed a thorough clinical examination, and performed routine investigations as well as an estimation of serum ferritin, liver enzymes, serum albumin, fibrinogen level, protein C and protein S levels, thrombin antithrombin complex and D-dimer.

Blood samples from groups NS and S were collected immediately before blood transfusion. Protein C, protein S, TAT, fibrinogen and D-dimer assay, citrated plasma was prepared as follows: blood samples were collected in tubes containing 3.2% sodium citrate as anticoagulant (9:1 ratio) and centrifuged at 2000g for 15 minutes to collect the plasma and the samples were stored at -80ºC until the time of assay. Protein C assay was measured using REAADS protein C antigen test kit (Corginex Inc, Colorado, USA). Protein S assay was measured using REAADS protein S antigen test kit (Corginex Inc. Colorado, USA). Both assays were performed using Sandwich ELISA technique (Behring ELISA processor P instrument). Plasma TAT levels were measured using Enzygnost TAT micro ELISA kits (Dade Behring Inc. New York, USA). The assay was also performed using the Sandwich ELISA technique by Behring ELISA processor P instrument. D-dimer assay was measured using VIDAS D-dimer exclusion kits (bioMérieux Inc. France), for the immune enzymatic determination of fibrin degradation products (FbDP) in plasma using ELFA technique (Vidas pc instrument, bioMérieux Inc. France).

STATA version 6.0 was used for data analysis (StataCorp, College Station, Texas, USA).Where applicable, the mean (standard deviation) of the results are shown. One-way analysis of variance (ANOVA) was used to assess differences among means of the three groups. Assumptions of one-way ANOVA, i.e. normal distribution and homoscedasticity of the data, were checked, and the raw data were transformed to an appropriate scale, such as log scale, to meet these assumptions. If the one-way ANOVA showed statistically significant differences among groups, each group was paired with the others and differences between the groups assessed by the Scheffe test. A statistically significant difference was defined as P<.05. Stepwise multiple regression analysis was used to assess the influence of multiple variables on a single variable, entering those variables that correlated significantly with the single variable by univariate analysis.¹¹

Results

The 30 non-splenectomized thalassemic patients had a median age of 11.5 years; there were 16 males and 14 females; 6 patients with regular blood transfusions (12 times/year) and 24 patients with infrequent blood transfusions (<12 times /year). The 20 splenectomized thalassemic patients had a median age of 12 years; there were 13 males and 7 females; 3 patients had regular blood transfusions and 17 had infrequent blood transfusions. The 20 healthy children had a median age of 12 years; there were 13 males and 7 females.

Serum ferritin, AST, and ALT were significantly higher in groups NS and S versus group C (Table 1). Plasma levels of protein C and protein S were significantly lower in NS and S groups than in the C group. Plasma levels of TAT and D-dimer were significantly higher in NS and S groups than in NC group. There was a significant decrease in protein C and protein S in the S group compared to the NS group (Table 2), while there was significant increase in TAT and D-dimer levels in the S group compared to the NS group.

Thalassemic patients with regular blood transfusions, whether NS or S, had higher levels of protein C than their peers with infrequent blood transfusion (Table 3). This finding was also present with protein S, while D-dimer levels were lower in frequently transfused patients than infrequently transfused ones in both groups. Although TAT levels were lower in frequently transfused patients than infrequently transfused ones in the NS group, the difference was not statistically significant.

Forty percent of patients had pulmonary hypertension detected by Doppler echocardiogram (Table 4). Most of these patients (80%) belonged to the splenectomized and infrequently transfused group. Also, there was a highly significant difference in the platelet count between S and NS groups (P<.001) with the highest platelet counts in the infrequently transfused S group. Platelet counts were normal in the NS group. There was significant negative correlation between protein C level in the whole population and both serum ferritin and TAT levels (P<.001), while there was no significant correlation between protein C and age, protein C and serum albumin, or protein C and AST (Figure 1).

Discussion

Current therapeutic approaches have substantially prolonged life expectancy in patients with thalassemia.¹² A consequence of this is the appearance of new complications. Venous thromboembolic events, such as pulmonary embolism, deep venous thrombosis and portal vein thrombosis, have been observed.^13-15 Several alterations that indicate a state of activation of the hemostatic mechanisms have been described in thalassemia major.

Protein C and protein S work together to inhibit the coagulation cascade. In our study, the levels of protein C and protein S in the two patient groups were statistically lower than those in the control group regardless of age. Similar results were obtained by Singer et al in Oakland, California,¹⁶ in studies of beta-thalassemia intermedia patients in Italy,¹⁷ and patients with beta-thalassemia major in Thailand and Turkey.^18,19

Explanations for decreased levels of protein C and protein S in thalassemic patients include vitamin K deficiency, liver dysfunction due to hemosiderosis, and the increased turnover rate of protein C and protein S.²⁰ In our study, no correlation was found between protein C levels in beta-thalassemia major patients and the levels of serum transaminases or albumin, eliminating hepatic dysfunction as a cause of these alterations. Shirahata et al19 found that liver damage was not the only cause of the reduction in these anticoagulant proteins. One alternative explanation for the significant reduction in protein C may be that this type of protein binds to phosphatidylserine, or other negatively charged phospholipids, abnormally present in the external membrane of the thalassemic erythrocytes.²¹

The reduction in protein C and S levels was significantly higher in splenectomized thalassemic patients rather than non-splenectomized ones. Consistent with this, Musumeci et al⁸ reported that the lowest values of protein C were found in older splenectomized patients. Also, these results are consistent with Shirahata et al¹⁹ in his study on 48 patients with beta-thalassemia/hemoglobin E disease, beta-thalassemia major and hemoglobin E disease. In contrast to our results, Shebl et al²² found that there was no significant difference in protein C and protein S levels between splenectomized and non-splenectomized patients with thalassemia major. The same was also reported by Cappellini et al¹⁷ in patients with thalassemia intermedia. The greater reduction in protein C and S levels in splenectomized thalassemic patients might be due to procoagulants on the surface of RBCs and abnormal platelets that are not removed from circulation in the case of splenectomy,⁵ with the resultant consumption of protein C and protein S in an attempt to control the hypercoagulable state.

Our results showed that serum ferritin levels in the patients were high and negatively correlated with protein C levels. Similar results were obtained in Italy by Musumeci et al⁸ on 74 thalassemic patients and in India by Naithani et al²³ on 54 thalassemic patients. This can be explained by the fact that iron overload (reflected in high serum ferritin levels) leads to a state of chronic oxidative stress in patients with beta-thalassemia as evidenced by higher reactive oxygen species and lower reduced glutathione levels in platelets.²⁴ This leads to platelet activation, protein C and protein S consumption and susceptibility to the thromboembolic consequences.²⁵

As a marker of coagulation activation, TAT levels were significantly increased in our patients. However, there was a marked increase in splenectomized rather than non-splenectomized patients. These findings indicate a low-grade hypercoagulable state. Similar results were obtained by Eldor et al,⁹ who found an elevated level of TAT complexes in 50% of their beta-thalassemia major patients. In contrast, Cappellini et al¹⁷ reported that levels of coagulation activation markers were normal in a population of patients with thalassemia major.

As a manifestation of enhanced fibinolysis, our patients had high plasma D-dimer levels, particularly those who had been splenectomized. In studies conducted by Kemahli et al26 and Tripatara et al,27 D-dimer levels were increased, being highest in the splenectomized group while Naithani et al23 and Cappellini et al¹⁷ reported that D-dimers were similar in beta-thalassemic patients and controls. Increased plasma levels of TAT and D-dimer are suggestive of continuous thrombin generation and enhanced fibinolysis, leading to a subclinical process of thrombosis which could contribute to the pulmonary hypertension, low lung capacity, hypoxemia, and diffusion defects associated with right heart failure (cor pulmonale)²⁸ and to the high frequency of ischemic brain lesions associated with asymptomatic brain damage as detected by MRI.²⁹

We also studied the impact of regularity of blood transfusion on the hypercoagulable state in our patients and found that levels of protein C and protein S were lower in infrequently transfused patients than in frequently transfused ones, while D-dimer levels were higher in infrequently transfused patients than frequently transfused ones. The same relationships were reported by Eldor and Rachmilewitz.³ Also, Cappellini et al17 reported that there was a larger prevalence of venous thromboembolic events in transfusion-independent patients with thalassemia intermedia (29%) than in regularly transfused patients with thalassemia major (2%). This was explained by the higher plasma levels of D-dimer and lower levels of naturally occurring anticoagulant proteins in transfusion-independent patients with thalassemia intermedia than in regularly transfused patients with thalassemia major. As an additional or alternative mechanism of hypercoagulability in infrequently transfused patients, the role of thalassemic red cells was considered because protein C is adsorbed to the abnormal red blood cell membrane of thalassemics, it is plausibly accepted to be lower in those patients who are not on a transfusion program because most of their RBCs are of thalassemic nature capable of adsorbing protein C compared to those frequently transfused patients whose blood contains less of thalassemic red cells and more on normal RBCs.

While pulmonary hypertension is increasingly recognized as part of the clinical spectrum for beta-thalassemia, little is understood about the mechanisms and risk factors for its development. None of our patients had clinically manifest pulmonary hypertension, but Doppler echocardiograms revealed pulmonary hypertension, defined as pulmonary artery systolic pressure more than 25 mm Hg in 20 patients (40%), and most of them (80%) belonged to the splenectomized and infrequently transfused group.

Our results are nearly identical with those reported by Singer et al16 where they found that 68% of their thalassemic patients had pulmonary hypertension by Doppler echocardiograms and most (94%) were splenectomized. Their higher percentages can be explained by the older age of their patients (26 [10] versus 12.1 [2.4] years in our study) and the higher percentage of splenectomized patients (72% versus 40% in our study).

The lowest values of protein C and S were observed in the splenectomized and infrequently transfused group, which explains the higher percentage of pulmonary hypertension in these patients. Also, we investigated the role of postsplenectomy thrombocytosis and found that there was a significant difference in the platelet count between splenectomized and non-splenectomized groups (P<.001) with the highest platelet counts observed in the splenectomized and infrequently transfused group. Similar results were obtained by Singer et al16 who also found that postsplenectomy thrombocytosis was correlated with evidence of pulmonary hypertension in their patients.

In conclusion, protein C levels were significantly reduced in thalassemic patients, especially among splenectomized and infrequently transfused ones, indicating a major role for protein C in the hypercoagulable state in those patients. These findings raise the issue as to whether it would be cost-beneficial to recommend prophylactic antithrombotic therapy in high-risk thalassemic patients. A wider prospective study will be necessary to delineate under which circumstances this might be implicated, and at what level of protein C deficiency to start prophylactic antithrombotic therapy.

References

1. El-Hamzi MA, Warsy AS, Bahakim H. Blood Proteins C and S in Sickle Cell Disease. Acta Haematol. 1993; 90(3): 114-119.
2. Giordano P, Del Vecchio GC, Altomare M, Coppola B, Schettini F, Iolascon A, et al. Resistance to activated protein C in thalassemic patients: An underlying cause of thrombosis. Eur J Haematol. 1998; 61(2):123-127.
3. Eldor A, Rachmilewitz EA.The hypercoagulable state in thalassemia. Blood. 2002; 99(1):36-43.
4. Eldor A. Abnormal platelet function in _-thalassemia. Scand J Haematol. 1978; 20(5): 447-452.
5. Ruf A, Pick M, Deutsch V, Patscheke H, Goldfarb A, Rachmilewitz EA, et al. In vivo platelet activation correlates with red cell anionic phospholipid exposure in patients with _-thalassemia major. Br J Haematol.1997; 98(1): 51-56.
6. Eldor A, Krausz Y, Atlan H, Snyder D, Goldfarb A, Hy-Am E, et al. Platelet survival in patients with _-thalassemia. Am J Hematol.1989; 32(2): 94-99.
7. Eldor A, Lellouche F, Goldfarb A, Rachmilewitz EA, Maclouf J. In vivo platelet activation in _-thalassemia major reflected by increased platelet-thromboxane urinary metabolites. Blood.1991; 77(8): 1749-1753.
8. Musumeci S, Leonardi S, Di Dio R, Fischer A, Di Costa G. Protein C and antithrombin III in polytransfused thalassemic patients. Acta Haematol. 1987; 77(1):30-33.
9. Eldor A, Durst R, Hy-Am E, Goldfarb A, Gillis S, Rachmilewitz EA, et al. A chronic hypercoagulable state in patients with beta-thalassemia major is already present in childhood. Br J Hematol. 1999; 107(4):739-746.
10. Kuypers FA, Yuan J, Lewis RA, Snyder LM, Kiefer CR, Bunyaratvej A, et al. Membrane phospholipid asymmetry in human thalassemia. Blood.1998; 91(8): 3044-3051
11. Nourusis MJ. Statistical package for social science (SPSS) base 8.0 for windows. Users guide. Chicago: SPSS Inc., 1997.
12. Olivieri NF, Brittenham GM. Iron chelating therapy and the treatment of thalassemia. Blood.1997; 89(3): 739-761.
13. Michaeli J, Mittelman M, Grisaru D, Rachmilewitz EA. Thromboembolic complications in beta thalassemia major. Acta Haematol. 1992; 87(1-2): 71-74.
14. Sumiyoshi A, Thakerngpol K, Sonakul D. Pulmonary microthromboemboli in thalassemic cases. Southeast Asian J Trop Med Public Health.1992; 23(2): 29-31.
15. Gillis S, Cappellini MD, Goldfarb A, Ciceri L, Fiorelli G, Rachmilewitz E.A. Pulmonary thromboembolism in thalassemia intermedia patients. Haematologica. 1999; 84(10): 959-960.
16. Singer ST, Kuypers FA, Styles L, Vichinsky EP, Foote D, Rosenfeld H. Pulmonary Hypertension in Thalassemia: Association with Platelet Activation and Hypercoagulable State. Am J Haematol. 2006; 81(9):670-675.
17. Cappellini MD, Robbiolo L, Bottasso BM, Coppola R, Fiorelli G, Mannucci AP.Venous thromboembolism and hypercoagulability in splenectomized patients with thalassemia intermedia. Br J Haematol. 2000; 111(2):467-473.
18. Visudhiphan S, Ketsa-Ard K, Tumliang S, Piankijagum A. Significance of blood coagulation and platelet profiles in relation to pulmonary thrombosis in beta-thalassemia/Hb E. Southeast Asian J Trop Med Public Health. 1994; 25(3):449-456.
19. Shirahata A, Funahara Y, Opartkiattikul N, Fucharoen S, Laosombat V, Yamada K. Protein C and protein S deficiency in thalassemic patients. Southeast Asian J Trop Med Public Health. 1992; 23(2):65-73.
20. Romcai D, Tositarat T, Kulapongs P. Activities of liver cell-producing coagulation factors in thalassemic children. Birth Defects Orig Artic Ser.1988; 23(5B):237-243.
21. Wilairat P, Kittikalayawong A, Chaicharoen S. The thalassaemia red cell membrane. Southeast Asian J Trop Med Public Health. 1992, 23(2):74-78.
22. Shebl SS, El-Sharkawy HM, El-Fadaly NH. Haemostatic disorders in nonsplenectomized and splenectomized thalassemic children. East Mediterr Health J. 1999; 5(6): 1171-1177.
23. Naithani R, Chandra J, Narayan S, Sharma S, Singh V. Thalassemia major-- on the verge of bleeding or thrombosis? Hematology. 2006; 11(1):57-61.
24. Amer J, Fibach E. Chronic oxidative stress reduces the respiratory burst response of neutrophils from beta-thalassaemia patients. Br J Haematol. 2005; 129(3): 435-441.
25. Rachmilewitz EA, Weizer-Stern O, Adamsky K, Amariglio N, Rechavi G, Breda L, et al. Role of iron in inducing oxidative stress in thalassemia: Can it be prevented by inhibition of absorption and by antioxidants? Ann N Y Acad Sci. 2005; 1054:118-123.
26. Kemahli S, Gürman C, Egin Y, Yildirmak Y, Sipahi T, Uysal Z, et al. Hypercoagulability in children with thalassaemia major. Clin Appl Thromb Hemost. 1997; 3(2):129-132
27. Tripatara A, Jetsrisuparb A, Teeratakulpisarn J, Kuaha K. Hemostatic alterations in splenectomized and non-splenectomized patients with _-thalassemia/ hemoglobin E disease. Thromb Res. 2007; 120(6): 805-810.
28. Grisaru D, Rachmilewitz EA, Mosseri IM, Gotsman M, Lafair JS, Okon E, et al. Cardiopulmonary assessment in beta thalassemia major. Chest. 1990; 98(5):1138-1142
29. Manfre L, Giarratano E, Maggio A, Banco A, Vaccaro G, Lagalla R.MR imaging of the brain: findings in asymptomatic patients with thalassemia intermedia and sickle cell-thalassemia disease. Am J Roentgenol. 1999; 173(6): 1477-1480.