Immunosuppressant-related hip pain can greatly affect a patient’s mobility and increase the number of total hip arthroplasty. We therefore investigate the risk factors and causes of hip pain after orthotopic liver transplantation (OLT).

Methods: The medical records of 175 adult OLT patients, who were followed-up for more than 2 years, were retrospectively reviewed. Data collected from the records included primary disease, medications, biochemical results, Child-Turcotte-Pugh (CTP) score, death, rejection, and complications related to liver transplantation.

Results: A total of 11 (6.3%) patients complained of hip pain, which was diagnosed as calcineurin inhibitor induced pain syndrome (CIPS) in 4 patients (2.3%), osteonecrosis of the femoral head (ONFH) in 3 patients (1.7%), osteoporosis in 2 patients (1.1%). The incidence of CIPS was related to the dose of FK506 (p<0.05) but independent of methylprednisolone usage. The occurrence of ONFH was independent of the dose and early withdrawal of methylprednisolone (both P>0.05). Patients with MP withdrawal within 6 months had significantly longer survival than those using MP for more than 6 months (50±15 vs. 41±18 months, P=0.007).

Conclusions: CIPS and ONFH are main causes of hip pain in adult OLT patients.  ONFH was not common but the incidence of hip pain due to CIPS was relatively high in OLT patients. Early withdrawal of MP could benefit the patients’ survival.

Keywords:

calcineurin inhibitor induced pain syndrome;  hip pain;  immunosuppressant;  osteonecrosis of femoral head;  osteoporosis

J.Orthopaedics 2012;9(2)e4

Introduction :

The use of immunosuppressant is essential after solid organ transplantation. Over the past few decades, while glucocorticoids have remained an important part of initial and maintenance immunosuppression, other immunosuppressants have been developed and especially those acting on immunophilins such as cyclosporine, tacrolimus, and sirolimus have greatly reduced the incidence of rejection. While immunosuppressants prevent the rejection of transplanted organs and tissues, they also cause various side effects. Immunosuppressant-related hip pain can greatly affect patient’s mobility and increase the chances of hip arthroplasty. Hip pain is most often a manifestation of osteoporosis and osteonecrosis. The prevalence of osteoporosis varies from 17.5% to 30% 1, 2, which is less in femoral neck than in other parts of body 3. It is multifactorial in etiology, which may result from existing liver diseases or immunosuppression therapy. The prevalence of osteonecrosis of the femoral head (ONFH) in patients who have undergone solid organ transplantation varies from 1.5 to 24%, depending on primary disease, immunosuppression therapy, age, gender, etc. Marston et al. reported the incidence of ONFH as 20% one year after renal transplantation 4. In other studies, however, the incidence of ONFH was as low as 1.5 - 4.6% in liver, kidney and heart transplant patients 5-8. In addition, some rare conditions should be also considered, including calcineurin inhibitor induced pain syndrome (CIPS).

The relationship between calcineurin inhibitors and bone pain was first reported by Bouteiller et al. in 1989 9 and it was named as calcineurin inhibitor induced pain syndrome (CIPS) by Grotz and colleagues in 2001 10. The prevalence of CIPS is 1-17%, mostly occurring in renal transplant recipients 10-15. The mechanism of CIPS is unknown because it is a rare syndrome and there is a lack of systematic study about it. Over the past decade, the number of liver transplantations performed worldwide has increased dramatically. The development of other immunosuppressants allows the early withdrawal of glucocorticoids, which may affect the outcome of immunosuppression after liver transplantation, including the manifestation of side effects. Although the techniques of transplantation and post-transplant treatment in China have been developed in line with Western countries, the immunosuppressant-related hip pain after liver transplantation has never been investigated in Chinese adult patients. We therefore conducted a retrospective study to investigate the risk factors and causes of hip pain after orthotopic liver transplantation in order to provide useful information for our clinical practice.

Patients and Methods

The medical records of patients, who underwent orthotopic liver transplantation at the Third Affiliated Hospital of Sun Yat-sen University between January 2004 and December 2007, were retrospectively studied in January 2010. This time period was selected to allow a minimum 2-year postoperative follow-up. Patient inclusion criteria included over 18 years of age; first-time liver transplantation; more than 2 years of follow-up; without pre-transplant osteoarticular disease, corticosteroid exposure and excessive alcohol intake. The study protocol was approved by the University Ethics Committee and has been performed in accordance with the ethical standards laid down in an appropriate version of the 2000 Declaration of Helsinki as well as the Declaration of Istanbul 2008. The use of the transplanted organs was approved by the ethics Committee of the Third Affiliated Hospital of Sun Yat-sen University, all organ donations were voluntary, and all donors and/or their next of kin provided consent for organ donation. Data collected from the records included primary disease, medications (preoperative, perioperative, and postoperative use of glucocorticoids and other immunosuppressants), laboratory examination results such as serum alkaline phosphatase, creatinine, urea nitrogen, cholesterol, triglycerides, calcium, total bilirubin and direct bilirubin levels, clinical outcome including death and rejection, Child-Turcotte-Pugh (CTP) score, and complications related to liver transplantation. When patients returned for routine follow-up, they were specifically asked whether they had pain or other problems with hips or other joints. If there was obvious deep and throbbing pain, clinical examination, conventional X-rays and magnetic resonance imaging (Signa Excite1.5 T Excite II, GE) were performed.

Osteoporosis and osteonecrosis were diagnosed according to the clinical guidelines of American College of Rheumatology (www.rheumatology.org ). ONFH was graded according to Ficat and Arlet classification (grades I to IV) 16. The diagnosis of CIPS was based on the following findings: a progressive pain on joints or muscles occurs after transplantation, related to calcineurin inhibitors; no abnormalities of joints revealed by conventional X-rays; MRI may show bone marrow oedema in the femur condylus; pain due to obvious clinical diseases such as osteoporosis, osteonecrosis, etc. has been excluded; the pain is only relieved by withdrawal of calcineurin inhibitors, not anti-inflammatory drugs.  The post-transplant immunosuppression regimen consisted of methylprednisolone (MP) and a calcineurin inhibitor such as FK506 and CsA. The dose of FK506 was adjusted according to the desired serum level, targeted to a concentration of 10-15 ng/ml during the first month, and then slowly reduced to attain peak levels of 10-12 ng/ml at 6 months and 5-10 ng/ml at 1 year and thereafter. In a few cases, monoclonal antibodies such as daclizumab were used for induction. In patients with renal failure, toxicities associated with calcineurin inhibitors or CIPS, the dose of FK506 was reduced first. If the condition could not be improved, FK506 was replaced by the combination of mycophenolate mofetil and sirolimus. The cumulative dose of FK506 was calculated at 12 months and 24 months.

Glucocorticoids were still considered as first line armamentarium in the post-transplant immunosuppression. All patients were given preoperative MP 1000 mg/day or 500 mg/day plus induction intravenously, which was rapidly tapered to 40 mg/day within 7 days following the transplantation. Beginning on postoperative day 7, a 48-mg dose of MP was given orally and then gradually tapered to 4-8 mg/day over 12 days. This dose was maintained for 3 to 12 months according to the primary diseases and patient’s physical condition. The cumulative dose of MP was calculated at 6 months and 12 months.

SPSS Statistics software version 17.0 was used for statistical analysis. Quantitative data were expressed as mean ± standard deviation (SD). The difference between two groups was analyzed by Welch t test or Mann–Whitney test. Qualitative data were expressed as percentile and subjected to Fisher exact test. Pearson's correlation coefficient analysis or Kruskal–Wallis test was performed to determine the correlation between hip pain and the use of corticosteroid and other immunosuppressive agents. Multivariate analysis was performed to determine the risk of hip pain.P<0.05 was considered statistically significant.

Results

During the study period, 175 liver transplant recipients were followed up in our hospital for more than 2 years, including 159 men and 16 women. The mean age was 47 ± 11 years (range, 19 - 72 years) and the mean survival time was 46 ± 16 months (range, 24 - 72 months). The primary diseases for liver transplantation were hepatitis B-related liver cancer, cirrhosis, and liver failure in 150 patients (85.7%) and other diseases including hepatocellular carcinoma, primary biliary cirrhosis, Wilson disease, and primary sclerosing cholangitis in 25 patients (14.3%). All patients used MP as a fixed part of initial and maintenance immunosuppression, in which 61 patients (34.9%) withdrew MP within 6 months and 114 patients (65.1%) used MP up to 12 months. A total of 11 patients complained of hip pain, which was diagnosed as CIPS in 4 patients (2.3%), ONFH in 3 patients (1.7%), osteoporosis in 2 patients (1.1%), simple muscular pain in one patient (0.6%), and metastatic cancer in one patient (0.6%). All of them were treated with MP and FK506. In 4 patients with CIPS, severe pain occurred in hip and other parts of lower limbs within 3 months after transplantation, which could not be relieved by anti-inflammatory drugs or reducing dose of FK506. X-ray examination did not reveal any abnormality in lower limb joints. MRI scans revealed unilateral bone marrow oedema in the femoral head in 2 patients (Fig. 1). After FK506 was withdrawn, the pain was finally eased. Compared with those without CIPS, the cumulative dose of FK506 in CIPS patients was significantly reduced but no difference was noted in terms of age, CTP score, biochemical characteristics and survival time between two groups (all P>0.05). (Table 1)  Furthermore, a positive correlation was found between hip pain and the cumulative dose of FK506 in CIPS patients (12 months: r=0.300, P=0.001; 24 months: r=0.280, P=0.002).

ONFH was ultimately diagnosed in 3 patients (1.7%) an average 14 ± 6 months (range, 10 - 21 months) after transplantation. As shown in Table 2, all of them had bilateral ONFH. The immunosuppression regimen consisted of MP and FK506 in all 3 patients and concurrent CsA or daclizumab in 2 patients. MP was withdrawn within 6 months in two patients. No difference was found in terms of cumulative doses of MP (6-month: 3553 ± 501 mg vs. 3147 ± 884 mg, P=0.126; 12-month: 4067 ± 572 mg vs. 3588 ± 1287 mg, P=0.123) between ONFH patients and non-ONFH patients. There was no correlation between the incidence of ONFH and the dose and early withdrawal of MP (both P>0.05). We further compared patients who withdrew MP within 6 months and those who had MP more than 6 months (total cumulative dose of MP: 2951 ± 864 mg vs. 4485 ± 928 mg, P<0.0001). The incidence of CIPS, osteoporosis and ONFH was not significantly different between the two groups (all P>0.05), suggesting that the early withdrawal of MP did not reduce the occurrence of CIPS, osteoporosis and ONFH. However, the patient withdrawing MP within 6 months had significantly longer survival time than those using MP for more than 6 months (50±15 vs. 41±18 months, P=0.007) but the mortality rate was not significant different (9.8% vs. 15.8%, P=0.356).

Discussion

This is the first study to investigate the incidence of hip pain and related risk factors in adult patients after orthotopic liver transplantation. We found that the main causes of hip pain were CIPS, ONFH, and osteoporosis in this cohort of patients without pre-existing osteoarticular disease, corticosteroid exposure and excessive alcohol intake. The mechanism of CIPS is poorly understood. Lucas et al. 11 studied 26 renal transplant patients who had isolated musculoskeletal pain due to cyclosporine A usage and suggested that the interaction between cyclosporine A and glucocorticoids, and the toxicity of cyclosporine metabolites may contribute to CIPS. Grotz and colleagues 10 hypothesized that calcineurin inhibitor-induced vascular disturbance might lead to an increased permeability of bone marrow vessels with consecutive bone marrow oedema. However, bone marrow oedema is a nonspecific change of CIPS because it can also occur in osteonecrosis, a joint disorder caused by wear and tear, osteoporosis, ischemia, and tumors, etc. Nevertheless, specific changes of CIPS are lacking. The diagnosis of CIPS is therefore mainly based on the exclusion of pain due to other causes and its correlation to the use of calcineurin inhibitor. In the present study, patients with pre-existing osteoarticular disease, corticosteroid exposure and excessive alcohol intake have been excluded.

Although long standing cholestasis due to liver diseases may remain, the diagnosis of osteoporosis and osteonecrosis has been also excluded by x-ray and MRI scans in four patients suspected with CIPS. Their hip pain was closely related to the use of FK506 and not responding to anti-inflammatory drugs. As a result, the diagnosis of CIPS can be established. Although CIPS is characterized by symmetrical pain mainly involving ankles, feet and knees because they are subjected to greater venous blood pressure in standing position 10, 17, it can also occur in femur as reported by Nishikawa et al 18. Previously, CIPS cases were rarely reported in liver transplant patients. In the present study, the incidence of CIPS due to FK506 treatment is 2.3%, which is relatively high. It is not clear whether it is particularly related to FK506 or liver dysfunction. Jam and colleagues reported that patients with hepatic dysfunction had longer half-life of FK506 (correspondingly lower clearance rate and higher trough plasma concentration)19. And the presence of CIPS is correlated to a high blood concentration of calcineurin inhibitor 12, 20. This may explain why the incidence of CIPS is high in the present study involving OLT patients; hence, a specific attention should be paid to the application of FK506 in patients with liver dysfunction in clinical practice.

In the present study, the incidence of ONFH was only 1.7% and the diagnosis was made about one year after transplantation. It is not related to the accumulative dose of methylprednisolone and FK506 or the early withdrawal of methylprednisolone. Our finding is consistent with the reports by others 4, 8, 21-24, suggesting that symptomatic ONFH is rather uncommon in this cohort of liver transplant patients with current individual immunosuppressive protocols. The underlying mechanism of osteonecrosis in liver transplant recipients is complex, which may be related to corticosteroid use, hypercoagulable state or hypofibrinolysis 5. Many patients with liver failure may have multi-factorial metabolic bone disease already prior to the transplantation. Pre-existing liver disease and alcohol excess may precipitate the occurrence of ONFH. We have excluded patients with alcoholic liver disease; therefore, the prevalence of ONFH is no doubt lower in our study than others.

The relationship between osteonecrosis and corticosteroid dose is controversial. Some researchers found a significantly decreased incidence of ONFH in transplant patients who were treated with reduced dose of corticosteroids compared to those treated with traditional protocols 25. Others however demonstrated that the induction of osteonecrosis due to corticosteroid use was dose-independent 5 26. In the present study, we also found that the induction of ONFH was not correlated to the accumulative dose and early withdrawal of methylprednisolone, confirmed a dose-independent effect of methylprednisolone on the occurrence of ONFH in Chinese adult patients after liver transplantation. A number of factors contribute to osteoporosis after liver transplantation, which include factors associated with chronic liver disease per se and those related to the immunosuppression such as corticosteroids.

Glucocorticoids can reduce bone formation and increase bone resorption, thereby leading to osteoporosis 27, 28.  The mechanisms by which these effects are achieved are complex, including inhibition of osteoblast differentiation and activity, the latter effect being mediated by stimulation of apoptosis 29. Corticosteroids also reduce osteoclast apoptosis, thus increasing bone resorption and increase osteocyte apoptosis, an effect that is implicated in the development of avascular necrosis 30. Although corticosteroid dosage has not been shown to be related to fracture risk in liver transplant recipients, several studies have shown corticosteroid dose-dependent effect on bone loss in patients undergoing renal, cardiac and lung transplantation 31-33. Bone loss may occur soon after corticosteroid initiation and is greatest in the first year 34. In recent years, the development of new immunosuppression agents has allowed the early steroid withdrawal after organ transplantation, which may reduce the incidence of osteoporosis and osteonecrosis 24, 35. In the present study, tacrolimus has been used in most patients, which allowed nearly half of patients to withdraw methylprednisolone within 6 months after transplantation, thereby dramatically reducing the total dose of methylprednisolone.

However, we only have two cases of osteoporosis. It is not powerful enough to perform any statistical analysis, thus it is hard to tell whether such low incidence of osteoporosis in our study is due to early withdrawal of corticosteroid usage. Studies in animal model showed that other immunosuppressants such as cyclosporine A and tacrolimus could reduce bone mass 36, but such effect is difficult to identify in humans because these immunosuppressive agents are always combined with corticosteroids.  In our cohort of patients, male sex is predominant. It may reflect the present condition of liver transplantation in China because male patients usually have stronger financial support than female patients in Chinese society. In addition, we found that the survival time in patients withdrawing methylprednisolone within six months was longer than those using methylprednisolone for more than six months but the rejection rate was similar between two groups of patients, suggesting that early withdrawal of methylprednisolone won’t affect the rejection. It can certainly prolong patient’s life, probably via reducing the side effects of corticosteroid treatment. The present study has certain limitations due to a relatively small number of patients with ONFH. Also, we did not measure bone mass, thereby underestimating the effect of methylprednisolone on this cohort of patients. Nevertheless, this is the first study about immunosuppression-related hip pain in Chinese OLT patients. It still provides important information about the symptomatic ONFH and CIPS with current individual treatment, which is helpful for our clinical practice.

Conclusion

CIPS and ONFH are main causes of hip pain in adult OLT patients. ONFH was not common, but the incidence of hip pain due to CIPS was relatively high in OLT patients compared with other study. Early withdrawal of MP could benefit the patients’ survival.

Acknowledgments

This work was supported by grants from the Major State Basic Research Development Program (973 Program) of China (2009CB522404), the National Natural Science Foundation of China (No. 30772135), and the Science and Technology Planning Project of Guangdong Province, China (2009B030801078).

Reference: