OBJECTIVES:

Thrombocytopenia is a common disorder among cirrhotics that has been traditionally explained by splenic platelet pooling and destruction. Thrombopoietin (TPO), the main stimuli for thrombopoiesis is produced primarily in the liver and degraded by circulating platelets, but its role in the thrombocytopenia of liver cirrhosis is not well understood. The main goal of this study is to clarify the role of TPO in the pathogenesis of thrombocytopenia in cirrhosis.

METHODS:

The relation among TPO, platelet count, spleen size, portal hypertension, and liver function was studied in 33 cirrhotic patients before and after either partial splenic embolization or liver transplantation.

RESULTS:

Cirrhotics with thrombocytopenia had lower serum TPO levels than healthy controls (median values (interquartile range: ICR) were 120.7 (42.0–191.6) vs 756.4 (527.0–965.1) pg/mL, respectively; p < 0.001). Among cirrhotics with thrombocytopenia, serum TPO was related to spleen size (=-0.387, p= 0.046), but not to platelet count as occurs physiologically. After partial splenic embolization, TPO and platelet count increased significantly and the physiological relation between TPO and platelet count was restored (=-0.665, p= 0.026). Similar results were observed after liver transplantation.

CONCLUSIONS:

Our results suggest that besides impaired production in the failing liver, an increased TPO degradation by platelets sequestered in the congested spleen may contribute to thrombocytopenia in cirrhotic patients.

INTRODUCTION

Thrombocytopenia is a common finding in patients with liver cirrhosis that usually appears once portal hypertension has been established and goes together with splenomegaly (^1,2,3). Accordingly, the most commonly accepted explanation for thrombocytopenia in cirrhosis is platelet sequestration in the congested spleen (²). Studies using radiolabeled platelets have clearly demonstrated that splenic sequestration occurs in liver cirrhosis (^4,5). And total or partial ablation of the spleen results in an increase in platelet count in most cirrhotic patients (^6,7). Yet, impaired thrombopoiesis in the bone marrow might also contribute to thrombocytopenia among cirrhotics (⁸), but its role has been regarded as negligible.

Recently thrombopoietin (TPO), a ligand for the receptor encoded by protooncogene c-mpl, has been shown to play a key role in the regulation of megakaryocyte development and platelet production (⁹). Since the liver is the main organ producing TPO (¹⁰), a decreased production of TPO in patients with chronic liver disease could result in deficient platelet production by the bone marrow. In fact, the increase in serum TPO observed after liver transplantation (¹¹) suggests that impaired production of TPO may well contribute to thrombocytopenia associated with cirrhosis.

On the other hand, circulating TPO is internalized into platelets after binding to the c-mpl receptor present in their membrane, and it has been proposed that is total platelet mass that directly regulates TPO activity in the bone marrow by sequestering circulating TPO (^12,13). Hence, increased degradation of circulating TPO by platelets sequestered inside the spleen could contribute to the thrombocytopenia observed in cirrhotic patients. The aims of our study were two: first, to determine the relations among thrombocytopenia, serum TPO, splenomegaly, and liver function in a group of cirrhotic patients; second, to observe the effect on these relations of reducing spleen sequestration by means of partial splenic embolization or alleviating portal hypertension and restoring liver function by means of orthotopic liver transplantation.

METHODS

Patients

Medical records of all patients with liver cirrhosis and thrombocytopenia submitted to partial splenic embolization or liver transplantation in our Liver Unit from 1989 to 1999 were reviewed. Both procedures were performed according to protocols approved by our Local Ethics Committee and after patients had signed their informed consent. As previously reported (^7,14), partial splenic embolization was performed by means of repeated injections of gelatin sponge into the splenic artery until a 50–60% reduction of the splenic blood flow was achieved. Immunosuppressive regime after orthotopic liver transplantation consisted of corticosteroids, azathioprine, and either cyclosporin A or tacrolimus (¹⁵). Liver cirrhosis was diagnosed histologically or by the concurrence of clinical, analytical, and imaging findings. Thrombocytopenia was defined as a platelet count below 130 10⁹/L in two consecutive blood tests. Splenectomized patients and those admitted because of active bleeding were excluded from analysis.

The following data were obtained from the records of each patient at baseline and 3 months after the procedure: Child-Pugh's score, platelet count, WBC count, hemoglobin, prothrombin activity and fibrinogen, ALT, alkaline phosphatase, total bilirubin, albumin, and gammaglobulin, and size of the spleen measured by ultrasonography (the maximal longitudinal diameter of the spleen was considered to represent spleen size).

Thrombopoietin Assay

TPO was measured in serum samples with an enzyme-linked immunoabsorbent assay (ELISA; R&D Systems, Abingdon, UK) following manufacturer's specifications. Samples had been stored at -40°C until analyzed. The lower detection limit of the assay was 31.2 pg/mL. The intra-assay coefficients of variation for samples at 117 pg/mL, 408 pg/mL, and 1,190 pg/mL were 4.7%, 3.3%, and 7.5%, respectively, and the interassay coefficients of variation at 88.5 pg/mL, 332 pg/mL, and 1,012 pg/mL were 11.8%, 8.2%, and 8.4%, respectively. Normal TPO levels in human blood were measured using plasma samples from 20 healthy individuals without thrombocytopenia.

Statistical Analysis

Statistical analysis was performed using a commercial statistical software package (SPSS for Windows, version 9.0). The Friedman test was used to compare the differences between TPO kits. When values of TPO were below the lower detection limit of the assay the results were reported in 30 pg/mL. Results are presented as median (interquartile range) unless otherwise expressed. The significance of the differences between means of TPO was assessed by the Mann-Whitney U test. The Wilcoxon signed rank test or Student's t-test were used for the analysis of paired samples. The Spearman or Pearson's rank correlation coefficient was used to examine the association between the parameters. A p-value of <0.05 was considered statistically significant.

RESULTS

Characteristics of the Patients

Twenty-two patients submitted to partial splenic embolization and 11 patients submitted to liver transplantation met the criteria of being cirrhotics with thrombocytopenia having a complete set of data at baseline and stored serum samples at appropriate time intervals to be included in the study. Partial splenic embolization was performed in three situations, namely, to allow full-dose interferon therapy for patients with active viral cirrhosis (five cases), to allow full-dose antineoplastic chemotherapy for patients with hepatocellular carcinoma (six cases), and when severe thrombocytopenia was associated with spontaneous bleeding events (four cases) or patients were facing major surgery (seven cases). Only 11 out of the 22 patients undergoing partial splenic embolization were followed for more than 3 months after the procedure without receiving myelosuppressive therapy (interferon for viral hepatitis or antineoplastic chemotherapy for hepatocellular carcinoma) and are available for analyzing changes after spleen ablation. Finally, a control group of 20 healthy subjects was obtained among individuals consulting for a check-up or workers of our institution.

The 33 cirrhotic patients with thrombocytopenia who form the study group had a median age of 58 yr (ranging from 25 to 67 yr), and 27 (81%) were males and 6 were females. The cause of liver cirrhosis was viral hepatitis in 24 patients, alcohol-induced liver disease in 5, and primary hemochromatosis in 1. The remaining three patients had cryptogenic liver cirrhosis. According to Child-Pugh's classification 17, 9, and 7 patients belonged to class A, B, and C, respectively. Ten patients had hepatocellular carcinoma superimposed on liver cirrhosis. Parameters concerning hypersplenism and liver function at baseline are summarized in Table 1.

Table 1 - Baseline Characteristics of the Study Group.

Full table

As shown in Figure 1, patients with liver cirrhosis and thrombocytopenia had significantly lower serum TPO levels than controls (expressed as median (ICR): 120.7 (42.0–191.6) vs 756.4 (527.0–965.1) pg/mL, respectively; p < 0.001). Contrary to what happens in normal subjects, no association was found between serum TPO and platelet count among patients with liver cirrhosis and thrombocytopenia. However, serum TPO was inversely associated with spleen size (=-0.387, p= 0.046; Fig. 2) so that the larger the spleen, the lower the TPO levels. TPO levels did not show any correlation with leukocyte count or a variety of liver function tests including ALT, albumin or bilirubin. However, there was a negative correlation with prothrombin time (=-0.412, p= 0.019). Also, a negative correlation between splenic size and platelet count was observed (=-0.400, p= 0.008) meaning that increasing splenomegaly was accompanied by increasing thrombocytopenia. Finally, in this population of cirrhotic patients with thrombocytopenia we could not find any statistical correlation between spleen size and other parameters of hypersplenism (hemoglobin or leukocyte count) or liver function (including ALT, total bilirubin, prothrombin activity and serum albumin, and Child-Pugh's score).

Figure 1.

Serum thrombopoietin in cirrhotic patients with thrombocytopenia and controls. Lines show median values.

Full figure and legend (7K)

Figure 2.

Relations between serum thrombopoietin and spleen size and platelet count among patients with liver cirrhosis and thrombocytopenia.

Full figure and legend (17K)

Effect of Partial Splenic Embolization

After partial splenic embolization both serum TPO and platelet count increased significantly at day 90 (p= 0.023 and p= 0.002, respectively). In addition to these changes, a significant increase in leukocyte count was observed at days 7, 30, and 90 (p < 0.001, p= 0.001, and p= 0.009, respectively), which reflects the reduced splenic pooling of white blood cells. As expected, no changes in either liver function tests or spleen size were observed at this time in this group of patients (Table 2).

Table 2 - Analytical Parameters before and 90 Days after Partial Splenic Embolization (n = 11) or Liver Transplantation (n = 11).

Full table

Figure 3 shows serial changes in serum TPO in nine patients at day 7 and 30. In seven out of nine patients the procedure was followed by a sharp, transient increase in TPO levels at day 7 and a more steady increase in platelet count. In the two patients in whom serum TPO levels failed to rise at day 7, platelet count increased.

Figure 3.

Early changes in serum thrombopoietin after partial splenic embolization.

Full figure and legend (21K)

A restitution of the physiological relation between TPO and platelet count was found at day 90 after partial splenic embolization as can be observed in Figure 4.

Figure 4.

Relation between serum thrombopoietin and platelet count observed among cirrhotic patients with thrombocytopenia 90 days after partial splenic embolization.

Full figure and legend (14K)

Effect of Orthotopic Liver Transplantation

TPO levels and platelet count increased significantly at day 90 after liver transplantation (p= 0.026 and p < 0.001, respectively). Leukocyte count also increased significantly (p < 0.001). And contrary to what happened after splenic embolization, bilirubin levels, and splenic size decreased significantly at day 90 after liver transplantation (p= 0.001 and p= 0.004, respectively) as expected (Table 2).

Figure 5 shows that a similar sharp increase in TPO levels was observed in this group of patients 7 days after transplantation. But, in this case TPO levels at day 90 were still significantly higher than before transplantation. Interestingly, platelet count at day 90 did not increase in three out of the four patients in whom TPO levels failed to raise after liver transplantation.

Figure 5.

Early changes in serum thrombopoietin after orthotopic liver transplantation.

Full figure and legend (16K)

Figure 6 shows that a tendency toward restitution of the physiological correlation between platelet count and TPO was observed in this group of transplanted patients, although statistical significance was not reached.

Figure 6.

Relation between serum thrombopoietin and platelet count observed among cirrhotic patients with thrombocytopenia 90 days after orthotopic liver transplantation.

Full figure and legend (10K)

DISCUSSION

Thrombocytopenia is a very common complication of liver cirrhosis (³). Several causes have been postulated for this reduced concentration of platelets including increased pooling in the enlarged spleen (^4,16), inappropriate production in the bone marrow (⁸), and reduced half-life-mediated immunologically (^17,18,19).

We have studied the effect of portal hypertension, spleen sequestration, liver function, and serum levels of the main factor that stimulates platelet production (i.e., TPO) on the intensity of thrombocytopenia in a population of cirrhotic patients. And most interestingly, we have evaluated these effects after two procedures that affect splenic function in cirrhotics through different pathways: first, partial splenic embolization that reduces rapidly and intensely the splenic mass (²⁰) without altering either liver function or portal hypertension (⁷); secondly, liver transplantation that reduces portal hypertension and improves liver function (²¹).

The rapid increase in leukocyte and platelet counts observed after partial splenic embolization (²²), together with the close relation between the intensity of reduction in the three blood series in the cirrhotic patient (^1,23,24) substantiates that thrombocytopenia in cirrhotics is mainly due to the ability of the congested spleen to sequester and destroy platelets. However, this pooling may have important consequences on the central activity of TPO. After binding a receptor on the surface of normal platelets, TPO is internalized and degraded in a process that physiologically regulates the activity of TPO in the bone marrow (^12,25). This phenomenon explains why platelet transfusion reduces TPO levels (¹³) and why a strong inverse correlation between thrombocytopenia and serum TPO can be observed after antineoplastic chemotherapy (²⁶) or interferon therapy for chronic viral hepatitis (²⁷). Among cirrhotic patients, platelet captured in a congested spleen could well result in an increased TPO degradation. Accordingly we have found for the first time a relation between TPO and spleen size, which somehow reflects the ability of the spleen to trap blood cells, but no association between TPO and peripheral platelet count, a finding previously observed (^28,29,30).

The relevance of TPO degradation by those platelets sequestered in the spleen is stressed by the observation that partial splenic embolization, a procedure that reduces spleen sequestration sharply and deeply without altering the ability of the liver to produce TPO, increases serum TPO levels among cirrhotic patients (^31,32), and restores the physiological relation between circulating platelet count and TPO (³³). An increase in serum TPO had also been observed after splenectomy in a small series of patients with cirrhosis, lymphoma, and gastric cancer (³¹). According to our view, the enlarged spleen in cirrhotics has an increased ability not only to entrap and kill platelets, but also to degrade TPO and impair platelet production in the bone marrow (³⁴). However, this rise in serum TPO levels after partial splenic embolization is not a definitive proof of increased degradation of TPO by sequestered platelets in the congested spleen, and only evidence of increased half-life of labeled TPO after partial splenic embolization would prove such issue.

The reduction in serum TPO among cirrhotics has been related to the impaired protein synthesis in the cirrhotic liver. The reasons for this belief include the progressive nature of TPO decline as cirrhosis evolves (^35,36), the direct relation between TPO and serum levels of other proteins synthesized in the liver such as coagulation factors (^28,35), and the restoration of normal TPO levels after liver transplantation (^37,38,39). However, our series is not the first to exclude a relation between TPO and liver function tests including prothrombin activity, albumin, or bilirubin (^29,30). Also, it should be noted that only cirrhotic patients with splenomegaly (⁴⁰) and not those without splenomegaly (^11,39,41,42) have been found to have reduced TPO levels when compared to controls. Portal hypertension and its consequences also develop in parallel with progression of cirrhosis (⁴³).

In summary, our results suggest that thrombocytopenia among cirrhotic patients is not only due to platelet sequestration in the congested spleen, but also to reduced thrombopoiesis because of enhanced degradation of TPO by the platelets entrapped within the spleen and/or reduced synthesis of TPO by the cirrhotic liver.

References

1. Toghill, PJ, Green, S. Splenic influences on the blood in chronic liver disease. Q J Med 1979;48: 613–625.
2. Toghill, P. The syndromes of splenic dysfunction. In: AJ Bowdler (ed.) The spleen, structure, function and clinical significance. Chapman and Hall: London, 1990: 209–232.
3. Bashour, FN, Teran, JC, Mullen, KD. Prevalence of peripheral blood cytopenias (hypersplenism) in patients with nonalcoholic chronic liver disease. Am J Gastroenterol 2000;95: 2936–2939.
4. Penny, R, Rozenberg, MC, Firkin, BG. The splenic platelet pool. Blood 1966;27: 1–16.
5. Aster, RH. Pooling of platelets in the spleen: Role in the pathogenesis of "hypersplenic" thrombocytopenia. J Clin Invest 1966;45: 645–657.
6. Ries, CA, Price, DC. (51Cr)platelet kinetics in thrombocytopenia. Correlation between splenic sequestration of platelets and response to splenectomy. Ann Intern Med 1974;80: 702–707.
7. Sangro, B, Bilbao, I, Herrero, I, et al. Partial splenic embolization for the treatment of hypersplenism in cirrhosis. Hepatology 1993;18: 309–314. | Article | PubMed | ChemPort |
8. Peltz, S. Severe thrombocytopenia secondary to alcohol use. Postgrad Med 1991;89: 75–76, 85.
9. Kuter, DJ, Beeler, DL, Rosemberg, RD. The purification of megapoietin: A physiological regulator of megakaryocyte growth and platelet production. Proc Natl Acad Sci USA 1994;91: 11104–11108. | PubMed | ChemPort |
10. Stoffel, R, Wiestner, A, Skoda, RC. Thrombopoietin in thrombocytopenic mice: Evidence against regulation at the mRNA level and for a direct regulatory role of platelets. Blood 1996;87: 567–573. | PubMed | ISI | ChemPort |
11. Martin, TG, Somberg, KA, Meng, YG, et al. Thrombopoietin levels in patients with cirrhosis before and after orthotopic liver transplantation. Ann Intern Med 1997;127: 285–288.
12. Fielder, PJ, Hass, P, Nagel, M, et al. Human platelets as a model for the binding and degradation of thrombopoietin. Blood 1997;89: 2782–2788. | PubMed | ChemPort |
13. Folman, CC, de Jong, SM, de Hass, M, et al. Analysis of the kinetics of TPO uptake during platelet transfusion. Transfusion 2001;41: 517–521.
14. Herrero, JI, Sangro, B, Quiroga, J, et al. Partial splenic embolization in the treatment of thrombocytopenia after liver transplantation. Transplantation 1997;63: 482–484.
15. Gomez-Manero, N, Herrero, JI, Quiroga, J, et al. Prognostic model for early acute rejection after liver transplantation. Liver Transpl 2001;7: 246–254.
16. Schaffner, A, Augustiny, N, Otto, RC, et al. The hypersplenic spleen. A contractile reservoir of granulocytes and platelets. Arch Intern Med 1985;145: 651–654.
17. Feistauer, SM, Penner, E, Mayr, WR, et al. Target platelet antigens of autoantibodies in patients with primary biliary cirrhosis. Hepatology 1997;25: 1343–1345.
18. Nagamine, T, Ohtuka, T, Takehara, K, et al. Thrombocytopenia associated with hepatitis C viral infection. J Hepatol 1996;24: 135–140.
19. Kajiwara, E, Akagi, K, Azuma, K, et al. Evidence for an immunological pathogenesis of thrombocytopenia in chronic liver disease. Am J Gastroenterol 1995;90: 962–966.
20. Bader Meunier, B, Husseini, K, Nouyrigat, V, et al. Partial splenic embolization in lymphangiomatosis. J Pediatr 2001;138: 613–614.
21. Mor, E, Shaharabani, E, Ben Ari, Z, et al. Experience with 100 liver transplant recipients at the Rabin Medical Center and Schneider Children's Medical Center. Transplant Proc 2001;33: 2943–2944.
22. Han, MJ, Zhao, HG, Ren, K, et al. Partial splenic embolization for hypersplenism concomitant with or after arterial embolization of hepatocellular carcinoma in 30 patients. Cardiovasc Intervent Radiol 1997;20: 125–127.
23. Alwmark, A, Bengmark, S, Gullstrand, P, et al. Evaluation of splenic embolization in patients with portal hypertension and hypersplenism. Ann Surg 1982;196: 518–524.
24. Brubaker, LH, Johnson, CA. Correlation of splenomegaly and abnormal neutrophil pooling (margination). J Lab Clin Med 1978;92: 508–515.
25. Möller, P, Andersson, PO, Andersson, AC, et al. Plasma thrombopoietin concentrations in response to platelet transfusion therapy. Acta Haematol 1999;102: 131–134.
26. Meng, YG, Martin, TG, Peterson, ML, et al. Circulating thrombopoietin concentrations in thrombocytopenic patients, including cancer patients following chemotherapy, with or without peripheral blood progenitor cell transplantation. Br J Haematol 1996;95: 535–541.
27. Shiota, G, Okubo, M, Kawasaki, H, et al. Interferon increases serum thrombopoietin in patients with chronic hepatitis C. Br J Haematol 1997;97: 340–342.
28. Okubo, M, Shiota, G, Kawasaki, H. Thrombopoietin levels in serum and liver tissue in patients with chronic viral hepatitis and hepatocellular carcinoma. Clin Sci (Colch) 2000;99: 207–214.
29. Espanol, I, Gallego, A, Enriquez, J, et al. Thrombocytopenia associated with liver cirrhosis and hepatitis C viral infection: role of thrombopoietin. Hepatogastroenterol 2000;47: 1404–1406.
30. Shimodaira, S, Ishida, F, Ichikawa, N, et al. Serum thrombopoietin (c-Mpl ligand) levels in patients with liver cirrhosis. Thromb Haemost 1996;76: 545–548.
31. Ichikawa, N, Kitano, K, Shimodaira, S, et al. Changes in serum thrombopoietin levels after splenectomy. Acta Haematol 1998;100: 137–141.
32. Hidaka, H, Kokubu, S, Saigenji, K, et al. Restoration of thrombopoietin production after partial splenic embolization leads to resolution of thrombocytopenia in liver cirrhosis. Hepatol Res 2002;23: 265.
33. Hobisch Hagen, P, Jelkmann, W, Mayr, A, et al. Low platelet count and elevated serum thrombopoietin after severe trauma. Eur J Haematol 2000;64: 157–163. | Article | PubMed | ChemPort |
34. George, JN. Platelets. Lancet 2000;355: 1531–1539. | Article | PubMed | ISI | ChemPort |
35. Kawasaki, T, Takeshita, A, Souda, K, et al. Serum thrombopoietin levels in patients with chronic hepatitis and liver cirrhosis. Am J Gastroenterol 1999;94: 1918–1922. | PubMed |
36. Adinolfi, LE, Giordano, MG, Andreana, A, et al. Hepatic fibrosis plays a central role in the pathogenesis of thrombocytopenia in patients with chronic viral hepatitis. Br J Haematol 2001;113: 590–595.
37. Peck Radosavljevic, M, Wichlas, M, Zacherl, J, et al. Thrombopoietin induces rapid resolution of thrombocytopenia after orthotopic liver transplantation through increased platelet production. Blood 2000;95: 795–801.
38. Goulis, J, Chau, TN, Jordan, S, et al. Thrombopoietin concentrations are low in patients with cirrhosis and thrombocytopenia and are restored after orthotopic liver transplantation. Gut 1999;44: 754–758.
39. Stiegler, G, Stohlawetz, P, Peck Radosavljevic, M, et al. Direct evidence for an increase in thrombopoiesis after liver transplantation. Eur J Clin Invest 1998;28: 755–759.
40. Espanol, I, Hernandez, A, Muniz Diaz, E, et al. Usefulness of thrombopoietin in the diagnosis of peripheral thrombocytopenias. Haematologica 1999;84: 608–613.
41. Peck Radosavljevic, M, Zacherl, J, Meng, YG, et al. Is inadequate thrombopoietin production a major cause of thrombocytopenia in cirrhosis of the liver? J Hepatol 1997;27: 127–131.
42. Stockelberg, D, Andersson, P, Bjornsson, E, et al. Plasma thrombopoietin levels in liver cirrhosis and kidney failure. J Intern Med 1999;246: 471–475. | Article | PubMed |
43. Sezai, S, Kamisaka, K, Ikegami, F, et al. Regulation of hepatic thrombopoietin production by portal hemodynamics in liver cirrhosis. Am J Gastroenterol 1998;93: 80–82.

Acknowledgements

This work was supported in part by a grant from Instituto de Salud Carlos III (C03/02).