In vitro data suggest that calcium plays an important role in normal and disordered erythropoiesis. The purpose of this study is to determine whether there is an association between serum calcium, various hormone levels, and the development of post transplant erythrocytosis (PTE). Data were collected on 283 patients who underwent renal transplantation between 1994 and 1998. The relationship between serum calcium and PTE development was tested using the chi-square test. Univariate and multivariable adjusted models were employed to determine predictors of maximum hematocrit. Selected patients underwent measurement of intact parathyroid hormone (PTH), 1,25-dihydroxy vitamin D, and erythropoietin (EPO). Seventy-three patients (26%) developed PTE. Post transplant erythrocytosis was more common in patients with hypercalcemia compared with patients with normal serum calcium (34% vs. 18%, p = 0.002). In multivariable analyses, serum calcium was a strong independent predictor of maximum hematocrit post transplant, even after adjustment for renal function. A serum calcium of ≥10.2 mg/dL was associated with greater than two-fold increased odds of PTE. There were no differences in hormone levels between subjects with hypercalcemia and PTE, subjects with PTE alone, and subjects with hypercalcemia alone. Hypercalcemia is associated with the development of PTE in renal transplant recipients.
Post transplant erythrocytosis (PTE) is a common complication following renal transplantation, and is estimated to occur in 10–20% of renal transplant recipients (1–3). Despite recognition of this common complication, the etiology of PTE remains unknown. Early studies implicated erythropoietin (EPO) in the pathogenesis of PTE. However more recent studies have produced conflicting data regarding the role of EPO, leading to speculation that factors other than EPO may play a role in the development of PTE.
With the observation that angiotensin-converting enzyme (ACE) inhibitors and, more recently, angiotensin II receptor blockers (ARBs) are effective in treating PTE, attention has now shifted to the role of the renin-angiotensin axis (4–6). Angiotensin Type I (AT1) receptors have been demonstrated on erythroid precursors, supporting a role for angiotensin II in erythropoiesis (7). Emerging evidence suggests that calcium may play an important role in erythropoiesis and, more specifically, angiotensin I-stimulated erythropoiesis. To examine the relationship between serum calcium and the development of PTE, we reviewed a large cohort of renal transplant recipients to identify risk factors for PTE. We also measured additional hormone levels in selected patients to further clarify the relationship between calcium and the development of PTE.
Materials and Methods
Data were collected retrospectively on consecutive patients undergoing renal transplantation from August 1994 to January 1998 at Duke University Medical Center. All patients were treated with standard triple immunosuppression therapy consisting of cyclosporine, prednisone, and either azathioprine or mycophenolate mofetil. The patients' medical records were reviewed for demographic and laboratory data including cause of end-stage renal disease, date of transplantation, ACE inhibitor use, and hematocrit, serum calcium, phosphorous, and creatinine levels.
Measurement of intact PTH, 1,25 Vitamin D, and erythropoietin levels
Selected patients were screened for additional laboratory measurements during transplant clinic visits from January 1999 to January 2002. Patients were eligible if they were at least 18 years of age, signed a written informed consent form, and had PTE, hypercalcemia (HC), or both on the basis of current laboratory studies as defined later. Three patients analyzed in the retrospective cohort were included in the prospective phase of the study, two with both HC and PTE and one with PTE alone. In addition to routine laboratory studies, serum levels of intact parathyroid hormone (PTH), 1,25-dihydroxy vitamin D, and EPO were measured at study entry. All laboratory studies were performed in the clinical labs of Duke University Medical Center with the exception of PTH, 1,25-dihydroxy vitamin D, and EPO levels, which were performed at the Mayo Medical Laboratories (Rochester, MN).
For both parts of the study, PTE was defined as a single hematocrit measurement of 50% in the absence of clinical pulmonary disease, or bone marrow disorder. Hypercalcemia was defined as a single serum calcium measurement of ≥10.2 mg/dL. Both definitions of PTE and HC are consistent with the published literature. In the prospective study, the serum calcium was normalized to a serum albumin of 4.0 mg/dL, assuming that the serum calcium concentration changed by 0.8 mg/dL for every 1 g/dL change in the serum albumin concentration. Glomerular filtration rate (GFR) was estimated using the modified MDRD equation (8). All analyses were performed using SAS version 8.2 (Cary, NC).
Retrospective study. A two-by-two table was created for the association of PTE with HC. The association was tested for significance with the Pearson chi-square test. After excluding patients who had graft loss within the first 6 months, the relationship was further explored by fitting a general linear model with the maximum hematocrit value as the dependent variable and age, race, gender, estimated GFR, and maximum calcium value as potential explanatory variables. A logistic regression model was also created using PTE as the outcome variable and age, race, gender, estimated GFR, and maximum calcium value as regressors.
Prospective study. Patients were classified into three groups, as detailed earlier: HC, PTE, and those with both PTE and HC. Demographic variables, the numbers of months since transplant, and values for EPO, 1,25- dihydroxy vitamin D, and PTH were compared among the three groups using the Wilcoxon rank sum test. The study was designed to have 80% power to detect a 50% difference between groups in 1,25- dihydroxy vitamin D levels.
Association of calcium with PTE
Excluding 16 patients with early graft loss, a total of 283 patients underwent renal transplantation at Duke University Medical Center during the study period. Seventy-three patients (26%) developed PTE. Post transplant erythrocytosis occurred in 45/131 (34%) patients with HC, and in 28/152 (18%) patients without HC. The association was highly significant (p = 0.002) by the chi-square test (Table 1). Among the PTE patients there were no significant differences in demographic variables or distribution of the cause of renal disease between those with and without HC (data not shown).
Table 1. Frequency of hypercalcemia in the patients with post transplant erythrocytosis
X2; p = 0.002
HC+: hypercalcemia present; HC –: hypercalcemia absent.
PTE+: post transplant erythrocytosis present; PTE – post transplant erythrocytosis absent.
In the 45 patients who had both HC and PTE, 40 patients developed HC before or concurrent with the development of PTE. The median time from transplant to the development of HC was 3 months, and from transplant to the development of PTE was 6 months. Of these patients, eight developed HC within 4 weeks of transplantation. Twenty-seven patients (60%) had sustained HC, defined as HC on successive laboratory measurements, and 25 patients (56%) had sustained PTE. Of the patients with sustained PTE, 64% also had sustained HC. Only two patients had early HC that was not sustained. None of the patients included in the prospective phase of the study had transient hypercalcemia.
Linear regression analysis confirmed that maximum serum calcium post transplant is a univariate predictor of maximum hematocrit (p = 0.0007). In the multivariable model, maximum serum calcium remained a significant predictor of maximum hematocrit post transplant (p = 0.004). Female gender was also associated with lower maximum hematocrit (p < 0.001). Results of the multivariable logistic regression analysis in which the outcome variable was the presence or absence of PTE are shown in Table 2. Maximum serum calcium was an independent predictor of PTE (p = 0.009). Higher estimated GFR was also associated with PTE (p = 0.04), and female gender was a negative predictor of PTE (p < 0.001). When maximum serum calcium was replaced in the model by the discrete variable HC indicated by serum calcium ≥10.2 mg/dL, the adjusted OR for HC was 2.1 (1–3,8).
Table 2. Multivariate predictors of post transplant erythrocytosis
(per 1-mg/dL increments)
(per 1 mL/min/1.73 m2)
Female vs. male
(per 1-year increments)
White vs. Black race
Association of calcium and hormone levels
In the second phase of the study, 17 patients with HC and PTE or PTE alone were studied. Patients with HC alone served as the control group. The demographic characteristics of these patients are shown in Table 3. The cause of renal failure was diabetes in 63% of the patients. Patients with PTE alone were more likely to be male compared with the other two subgroups. The length of time since transplantation was somewhat greater for the patients with both HC and PTE compared with the patients in the other two subgroups, however, this difference was not statistically significant.
Table 3. Clinical characteristics of the prospective study group
All patients n = 26
HC + PTE n = 7
PTE only n = 10
HC only n = 9
Values are mean ± SE except where indicated.
p-values are for differences between the three subgroups.
Months since transplant
Current ACEI use
Patients with both HC and PTE and those with PTE alone had a mean hematocrit of 54 ± 1% (Table 4). The mean serum calcium in the patients with HC and PTE was 10.5 ± 0.1 mg/day compared with the patients with PTE alone, who had a mean serum calcium of 9.6 ± 0.1 mg/dL. There were no significant differences in levels of EPO, intact PTH or 1,25-dihydroxy vitamin D between the three groups. Among the PTE patients, when calcium and EPO levels were considered as continuous variables, there was a significant association between serum calcium and serum EPO, such that higher serum calcium was associated with a lower EPO level (Figure 1, p = 0.05 by least-squares regression). This relationship was also true when using the corrected calcium concentration.
Table 4. Laboratory characteristics of the prospective study group
Although EPO is a well-established mediator of erythropoiesis, previous studies have reported conflicting results with respect to its role in the pathogenesis of PTE. Up to half of all PTE patients have undetectable levels of EPO (4,9), suggesting that PTE may be mediated by other factors.
In this study, HC was a strong independent predictor of PTE in a large cohort of renal transplant recipients. A serum calcium level of 10.2 mg/dL was associated with a greater than two-fold increased odds of PTE, and was independent of other known predictors of PTE including male gender and estimated GFR (2,4,10). There were no significant differences in levels of intact PTH, 1,25-dihydroxy vitamin D, and EPO between patients with HC and PTE, PTE alone, and HC alone, suggesting that the development of PTE may be directly mediated by serum calcium.
There are two alternative hypotheses that may explain the relationship between HC and PTE. Post transplant erythrocytosis and HC may be linked by other factors that mediate or are associated with both conditions, or PTE may modulate serum calcium concentration. Post transplantation HC has been attributed to persistent hyperparathyroidism, increased renal synthesis of 1,25-dihydroxy vitamin D, and bone resorption resulting from resolution of skeletal resistance to PTH (11). In vitro and clinical data have suggested a role for vitamin D in erythropoiesis (12–14). However we failed to find a difference in 1,25-dihydroxy vitamin D levels to support this conclusion.
Pretransplantation PTH values were not available for most patients in the retrospective study. Thus, we could not evaluate the role of pretransplantation hyperparathyroidism as a risk factor for PTE. In the prospective phase of the study, there were no significant differences in post-transplantation PTH levels between the three patient subgroups. Post transplantation PTH levels correlate with pretransplant values in transplant recipients with serum creatinine <1.5 mg/dL (15). Moreover, secondary hyperparathyroidism is typically associated with EPO resistance and anemia in dialysis patients. Thus, it is unlikely that PTH mediates the observed association between HC and PTE.
Adynamic bone disease has been associated with HC in dialysis patients (16). Although we did not have clinical data regarding the presence of adynamic bone disease, we did not find a significant difference in the prevalence of risk factors predisposing to adynamic bone disease, including age and diabetes, between patients with HC and PTE and patients with PTE alone. Finally, PTE may be a cause rather than a result of HC. The temporal relationship between the development of the two conditions in our study argues against this possibility. Hypercalcemia preceded the development of PTE in the majority of the patients with both conditions, supporting a possible causal role of HC in the development of PTE.
Emerging evidence suggests that calcium may be an important modulator of erythroid proliferation. Several in vitro reports have proposed that Ca2+ flux and intracellular Ca2+ concentration are important determinants of erythroid proliferation and differentiation. Erythropoietin has been shown to induce increases in intracellular Ca2+ concentration in late human erythroblasts through an EPO-regulated Ca2+ channel (17–19). The EPO-induced increase in intracellular free Ca2+ is dependent on the extracellular Ca2+ concentration and can be blocked by nifedipine (18).
Carozzi et al. studied erythroid precursors from PTE patients and demonstrated that proliferation of these cells was enhanced three-fold in the presence of increasing extracellular Ca2+ concentration while maintaining a constant low EPO concentration in the culture medium (20). In our study, the frequency of PTE was almost twice as high in patients with elevated serum calcium levels compared with patients with normal serum calcium levels. Taken together, these results suggest that an elevated serum calcium concentration is important in the pathogenesis of PTE. Consistent with previous studies, EPO levels were within the normal range, and were not significantly different between the three patient subgroups studied (2,10,21,21). However in the patients with PTE, there was a significant inverse correlation between serum calcium and EPO levels. Erythropoietin is known to modulate erythroid calcium channels (18); thus this finding may indicate increased sensitivity of calcium channels to EPO in the presence of elevated serum calcium. Alternatively, increased hematocrit in persons with PTE may have reduced EPO levels independently of serum calcium.
Angiotensin-converting enzyme inhibitors and, more recently, ARBs have been demonstrated to be effective in the treatment of PTE (4,22,23). Most reports have indicated the reduction in hematocrit in patients treated with ACE inhibitors or ARBs to be independent of circulating EPO levels (5,24). Accumulating evidence suggests that calcium may play a key role in EPO-independent, angiotensin II mediated erythropoiesis. Recent in vitro studies have demonstrated erythroid proliferation by stimulation of the AT1 receptor on erythroid progenitors by angiotensin II (7). Furthermore, stimulation by angiotensin II was associated with increases in intracellular calcium in erythroid progenitors derived from PTE patients (25). Thus, increases in intracellular calcium may occur via stimulation of the AT1 receptor by angiotensin II. Although further studies are needed, these results suggest that calcium may act as a second messenger for both EPO and angiotensin II-stimulated erythropoiesis.
There are several limitations of the present study. Clinical and laboratory parameters of metabolic bone disease were not available for most patients before transplantation. Thus we could not evaluate the possibility that pretransplantation bone disease is causally related to the observations reported in this study. Medication use other than ACE inhibitors, including calcium channel blockers, calcium-containing supplements, and vitamin D preparations, was not ascertained. Finally, given the retrospective nature of the current study, our results can only suggest but not prove causality.
In summary, HC, defined as serum calcium of ≥10.2 mg/dL, is associated with the development of PTE. Maximum serum calcium is a predictor of maximum hematocrit in renal transplant patients, even after adjustment for gender and renal function. These data in concert with other studies indicate that calcium plays an important role in erythroid proliferation in PTE patients, and may be involved in angiotensin II-mediated PTE. Further studies are needed to clarify the mechanisms by which HC leads to increased erythroid proliferation.
There were no external funding sources for this study.