Thrombopoietin in postoperative thrombocytopenia following living donor hepatectomy



Thrombocytopenia is a frequent finding following living donor hepatectomy. It appears more pronounced in right graft donors than in left graft donors. This study analyzed postoperative thrombocytopenia in 20 living liver donors and examined the change of endogenous thrombopoietin (TPO) in its recovery. Platelet count, TPO level, fibrinogen degradation product (FDP), and D-Dimer were measured before surgery and on postoperative days (PODs) 1, 2, 3, 5, 7, and 14. Concurrently, liver and spleen volumes were calculated by computed tomography. Platelet count on POD 3 was significantly lower in right graft donors than in left graft donors (13.0 ± 3.7 × 104/μL vs. 16.8 ± 4.0 × 104/μL, P = 0.039) but recovered by POD 7 in all donors. Postoperative elevations of FDP and D-Dimer were significantly higher in right graft donors than in left graft donors. TPO level rose immediately after surgery, peaked on POD 5 in left graft donors and on POD 7 in right graft donors, and fell nearly to preoperative levels by POD 14. Postoperative TPO level per liver volume was significantly higher in right graft donors than in left graft donors. In conclusion, thrombocytopenia following living donor hepatectomy resolved within the first week regardless of graft type and was mainly associated with increasing consumption of circulating platelets, possibly due to intrahepatic and splenic congestion. With a reduced number of circulating platelets, TPO level rapidly increases. Also, with reduced consumption of platelets related to recovery from surgery, thrombocytopenia should resolve. As a consequence, TPO level would be expected to fall. Liver Transpl 12:435–439, 2006. © 2006 AASLD.

Thrombocytopenia is common after living donor hepatectomy but usually improves soon after surgery.1 The intensity of thrombocytopenia is correlated with the magnitude of the hepatic resection.2 With right lobe grafts now commonly used for adult-to-adult living donor liver transplantation, our understanding of the factors involved in postoperative thrombocytopenia after massive liver resection is particularly important.

Thrombopoietin (TPO) is a hematopoietic cytokine that promotes megakaryocytic development and platelet production. It is synthesized mainly by hepatocytes. In thrombocytopenic states, serum levels of TPO increase.3 This study explores postoperative thrombocytopenia in living donors of various-size hepatic grafts for transplantation, as well as transition of endogenous TPO in its recovery.


TPO, thrombopoietin; FDP, fibrinogen degradation product; POD, postoperative day.



Data on 20 consecutive living donors who underwent uncomplicated hepatectomy between September 2000 and December 2001 were analyzed. Donors were divided into 2 groups: left donors (left lobectomy, n = 5; left lateral segmentectomy, n = 5) and right donors (right lobectomy, n = 10). In donors who underwent right lobe hepatectomy, the middle hepatic vein trunk was retained. The clinicopathological data for donors and grafts were compared between groups. No donor received any blood products perioperatively.

Postoperative Change of Platelet Count and Fibrinolytic Profile

Platelet count, fibrin/fibrinogen degradation products (FDP, μg/mL) and D-Dimer (μg/mL) were measured preoperatively and on postoperative days (PODs) 1, 2, 3, 5, 7, and 14 and were compared between groups.

Serum TPO Assay

Serum samples were collected immediately prior to the donor operation and on PODs 1, 2, 3, 5, 7, and 14. Samples were rapidly frozen and preserved at −80°C until analysis. Serum TPO concentrations were measured by a sensitive sandwich enzyme-linked immunosorbent assay method, as previously described by Tahara et al.4

Volumetric Study of Liver and Spleen

Liver and spleen volumes were measured by helical computed tomography using the volumetric software 3D Virtuoso (Siemens-Asahi Medical Technologies, Tokyo, Japan) preoperatively and on PODs 7 and 14. On the day of the transplant, the liver volume was calculated by subtracting the computed tomography graft volume from the computed tomography whole-liver volume in order to exclude any error due to blood volume included within the graft.

Statistical Analysis

Values are expressed as mean ± standard deviation. Pearson's correlation test was used for analysis of continuous variables. For statistical comparison, Student's t-test was used to assess the group difference in continuous variables and chi-squared test for discrete variables. P < 0.05 was considered statistically significant.


Clinicopathological Donor and Graft Data

Characteristics of the left and right graft groups are summarized in Table 1.

Table 1. Comparison of Clinical Data between Left and Right Graft Donors
ParameterLeft graft donors (n = 10)Right graft donors (n = 10)P value
  1. NOTE: Values represents mean (range) except gender.

Age (yr)34.7 (20–52)34.1 (23–53)0.90
Gender (% female)
Liver resection ratio (%)28.2 (20.4–38.1)62.6 (56.8–70.6)<0.001
Operation time (minutes)517.2 (435–681)569.6 (476–630)0.14
Blood loss (mL)318.1 (150–565)407.0 (50–670)0.31
Intraoperative volume balance+2989.9 (1900–4280)+2484.0 (1510–4180)0.18
Hospital stay (day)16.6 (13–20)18.5 (16–25)0.10

The liver resection ratio (ratio of graft volume to whole-liver volume) was higher in the right graft group than in the left graft group (62.6 ± 5.0% vs. 28.2 ± 5.4%, P < 0.001). There were no significant differences between groups in age, gender, operation time, intraoperative blood loss, or total length of hospital stay.

Postoperative Changes in Platelet Count and Serum TPO Concentration

Postoperative changes in platelet count and serum TPO concentration are shown in Figure 1. In both groups, platelet count fell markedly immediately after surgery, but then rose again to exceed pretransplant levels by POD 14. The median nadir in platelet count occurred on POD 2 in left donors and on POD 3 in right donors. The nadir among right donors was significantly lower than in left donors (13.0 ± 3.7 × 104/μL vs. 16.8 ± 4.0 × 104/μL, P = 0.039).

Figure 1.

Serial change in postoperative (A) platelet count and (B) serum TPO concentration after living donor hepatectomy. Values are presented as mean ± standard deviation. *P < 0.001; **P < 0.05 vs. preoperative level.

Serum TPO concentration also rose immediately after surgery in both groups, peaking on POD 5 in left donors and on POD 7 in right donors. On each of the first 3 PODs, serum TPO was significantly higher in left donors vs. right donors (POD 1, P = 0.017; POD 2, P = 0.004; POD 3, P = 0.007). In both groups, serum TPO concentrations returned nearly to preoperative levels by POD 14.

Spearman's ranked correlation coefficient for platelet count and serum TPO concentration was −0.36 in left donors (P = 0.003) and −0.27 in right donors (P = 0.027), indicating that a significant inverse correlation was present between them in both groups.

Postoperative Fibrinolytic Profile Changes

Serial changes in postoperative fibrinolytic profiles are shown in Figure 2. FDP and D-Dimer values were considerably higher in right donors. Between groups, there were significant differences on POD 2 (FDP, P = 0.027; D-Dimer, P = 0.036), POD 3 (FDP, P = 0.019; D-Dimer, P = 0.037), and POD 5 (D-Dimer, P = 0.012).

Figure 2.

Serial change in postoperative (A) FDP and (B) D-Dimer following living donor hepatectomy. Values are presented as mean ± standard deviation. *P < 0.001; **P < 0.05 vs. preoperative level.

Serial Increases of Remnant Liver Volume and Spleen Volume

Serial changes in remnant liver and spleen volumes are shown in Figure 3. Remnant liver volume was 1077.8 ± 170.9 cm3 on POD 7 and 1006.9 ± 190.3 cm3 on POD 14 in left donors, and 675.5 ± 138.4 cm3 on POD 7 and 762.7 ± 131.5 cm3 on POD 14 in right donors.

Figure 3.

Transition of postoperative (A) remnant liver volume and (B) spleen volume measured by a helical computed tomography image. Values are presented as mean ± standard deviation. *P < 0.001; **P < 0.05 vs. preoperative level.

In both groups, spleen volumes were enlarged significantly after surgery. There were, however, no statistically significant differences in spleen volume between left and right donors on either POD 7 (left, 170.9 ± 45.4 cm3 vs. right, 197.9 ± 77.1 cm3) or POD 14 (left, 153.0 ± 49.6 cm3 vs. right, 184.7 ± 58.0 cm3).

Change in Postoperative Serum TPO Concentration per Liver Volume

Postoperative changes in serum TPO concentration per liver volume were significantly higher immediately after operation and on PODs 7 and 14 in right donors than in left donors (Fig. 4).

Figure 4.

Serial change in postoperative serum TPO concentration per liver volume. Values are presented as mean ± standard deviation. *P < 0.001; **P < 0.05 vs. preoperative level.


When the indications for living donor liver transplantation were successfully expanded from pediatric to adult patients, the most common graft type shifted from the left liver to the right liver. Extensive liver resection may result in transient metabolic impairment and coagulation derangement,5 but in testament to the enormous regenerative capacity of the normal liver, functional recovery after living donor hepatectomy occurs earlier than morphologic restoration.6 Postoperative thrombocytopenia is known to be more severe after massive liver resection. Obviously, therefore, it is likely to be more severe in right graft donors than in left graft donors. If serious platelet depletion occurs, replacement therapy should be considered, but thrombocytopenia usually improves soon after surgery.

In our patients, postoperative thrombocytopenia resolved within the first week regardless of graft type. Rudow et al.,7 however, recently reported that in 5 of 22 (23%) right graft living liver donors followed for at least 1 yr, asymptomatic thrombocytopenia persisted beyond the 90-day perioperative period. The significance of their finding remains uncertain.

Several factors are thought to play a role in the development of postoperative thrombocytopenia following living donor hepatectomy, including sequestration of platelets in the remnant liver, impaired production of TPO due to reduction of hepatocytes, increased platelet consumption, hypersplenism, hemodilution, or a combination of these.8 In this study, postoperative elevation of FDP and D-Dimer indicates the exacerbation of secondary hyperfibrinolysis early after hepatectomy.

In a study of platelet kinetics after liver resection in rats, Siemensma et al.2 reported that four-fifths hepatectomy produced an immediate cessation of entry of platelets into peripheral blood. Recovery of platelet number began on POD 5. Platelets were found to accumulate in the hepatic sinusoids of the remnant liver after hepatectomy, but these were not believed to contribute significantly to the thrombocytopenia. Instead, the authors concluded, the fact that massive hepatectomy had a major effect on entry of platelets into the peripheral circulation suggests that in addition to accumulation of platelets in the sinusoids, thrombocytopenia must also result from a second mechanism involving the final step of platelet production.

Presumably, however, sequestration of platelets in the remnant liver and increasing consumption of circulating platelets may be important factors in postoperative thrombocytopenia in living donors. Splenic sequestration, pooling, and destruction of platelets may also have roles in the postoperative thrombocytopenia seen in these patients, given their significant splenic enlargement after surgery. On the other hand, reduced platelet production cannot entirely account for the postoperative thrombocytopenia, because serum TPO concentration rose immediately after surgery. Similarly, the median nadir day in the circulating platelet count did not occur immediately after the operation, which may not support the theory that hemodilution refers to the postoperative thrombocytopenia.

Postoperative thrombocytopenia in liver transplant recipients has been considered in detail. McCaughan et al.9 suggested that allograft dysfunction was the most consistent independent predictor of the nadir platelet count post liver transplantation. Chatzipetrou et al.8 have reported that thrombocytopenia is related to poor function of the liver allograft and patient survival. They concluded that a rise in the mean platelet count after the second postoperative week reflects proper graft function. These reports suggest that satisfactory metabolic allograft function is essential for the resolution of postoperative thrombocytopenia. On the other hand, Chang et al.10 postulated that persistent thrombocytopenia portended a poor outcome in liver transplant recipients and was not related to low TPO levels. They emphasized major infections as the cause of thrombocytopenia. Indeed, impaired production of TPO and the presence of conditions that exacerbate platelet consumption, such as major infection, are both important factors that can result in persistent postoperative thrombocytopenia in transplant recipients.

TPO is presumed to be the main regulator of platelet count. Circulating levels of TPO are inversely related to platelet mass. Platelets contain an avoid TPO receptor (c-Mpl) that efficiently binds and removes TPO from circulation.11 Because TPO is synthesized mainly by hepatocytes, inadequate production in patients with chronic liver disease may contribute to thrombocytopenia that is rapidly reversed by the living donor allograft.12 In spite of greater decrease of platelet after right-side hepatectomy, a lower level of TPO than in left graft donors reflects decreased hepatocytes mass. This is supported by the reverse of this when the TPO levels are expressed by liver volume (Fig. 4). In this study, donor tissue could not be collected, so TPO production by the remnant hepatocytes has not been confirmed. Although the precise role of bone marrow, spleen, kidney, or muscle in TPO production during thrombocytopenia remains unknown, further studies should elucidate the role of TPO mRNA expression in hepatocytes in the process of ameliorating postoperative thrombocytopenia following living donor hepatectomy.

In summary, postoperative thrombocytopenia in living liver donors is attributable mainly to increased consumption of circulating platelets possibly due to intrahepatic and splenic congestion. With a reduced number of circulating platelets, and hence a reduced number of TPO receptors, serum concentrations of free TPO rapidly increase. Also, with reduced consumption related to recovery from the surgery, thrombocytopenia resolved within 1 week of hepatectomy in both left and right graft living donors. As a consequence, serum TPO levels would be expected to fall.