Objectives: Persistent hyperparathyroidism can have a deleterious effect on graft function in kidney transplant recipients, although serum calcium, phosphorus, and parathyroid hormone levels tend to normalize after successful transplant. Parathyroidectomy can result in sustained amelioration of persistent hyperpara-thyroidism despite graft failure risk and unfavorable graft outcomes. Data on this issue are limited and conflicting. Here, we evaluated the effects of parathyroidectomy on graft function in kidney transplant recipients.
Materials and Methods: This retrospective study included 249 adult kidney transplant recipients (121 deceased-donor/128 living-donor; 142 males/107 females; mean age of 39.3 ± 11.6 y; mean follow-up of 46.5 ± 23.5 mo). Participants were grouped as those without (n = 222), those with pretransplant (n = 12), and those with posttransplant (n = 15) parathyroidectomy. Graft outcomes and serum calcium, phosphorus, and parathyroid hormone levels were studied.
Results: Serum calcium levels at baseline and at 1, 3, 6, and 12 months and parathyroid hormone levels at baseline and at 6 and 12 months were higher and serum phosphorus levels at 3, 6, and 12 months were lower in the posttransplant parathyroidectomy group versus the other groups (P < .001). We observed no significant differences between groups regarding serum calcium, phosphorus, and parathyroid hormone levels at last visit. Estimated glomerular filtration rates at 3, 6, and 12 months and at last visit in the pretransplant parathyroidectomy group were higher than in those without parathyroidectomy (P < .05) and higher at 6 and 12 months than in the posttransplant parathyroidectomy group (P < .05). No significant differences regarding graft loss and patient mortality were observed among the 3 groups (P > .05).
Conclusions: Parathyroidectomy resulted in sustained decreased levels of serum calcium and parathyroid hormone. We observed no graft failure risk associated with parathyroidectomy in our study. Parathyroid-ectomy before transplant is advantageous with better graft function.
Key words : Graft outcome, Hypercalcemia, Hyper-parathyroidism, Parathyroid hormone, Renal transplant
Chronic kidney disease is commonly associated with disorders of mineral and bone metabolism, including secondary hyperparathyroidism. The prevalence of secondary hyperparathyroidism increases as kidney function declines, and approximately 10% of patients with end-stage renal disease undergo parathyroidectomy.1 In addition, a portion of patients on wait lists have untreated and unrecognized hyperparathyroidism. Hyperparathyroidism resolves in most patients after transplant; however, some have persistent hyperparathyroidism (PHPT). Persistent hyperparathyroidism and hypercalcemia can be associated with deleterious effects on graft function via tubulointerstitial calcifications and renal vasoconstriction.2,3 Parathyroidectomy is one of the treatment approaches for PHPT after transplant and usually provides permanent improvement in hypercalcemia, but it may lead to allograft function deterioration.4,5 Some candidates have required parathyroidectomy while on wait lists, thus resolving this issue before transplant.
The timing of parathyroidectomy varies among transplant centers and is critical because of possible negative effects on graft outcome. Few studies have addressed the favorable timing of parathyroidectomy (before or after transplant) and its effect on long-term graft function.6-9 Here, we analyzed the effects of timing of parathyroidectomy on graft function.
Materials and Methods
We retrospectively evaluated 390 consecutive adult recipients who underwent kidney transplant between January 2008 and December 2014. Of these, 72 recipients with less than 12-month follow-up (39 died, 23 had graft loss, 10 had missing data) were excluded from the study. Of the remaining 318 recipients with at least 12-month follow-up, 69 were excluded because of missing data, resulting in 249 patients (121 deceased/128 living donor; 142 men/107 women; mean age of 39.3 ± 11.6 y; mean follow-up of 46.5 ± 23.5 mo) included in the study.
Participants were divided into 3 groups according to parathyroidectomy history and timing: those without parathyroidectomy (n = 222; non-PTX group), those who underwent parathyroidectomy pretransplant (n = 12; old-PTX group), and those who underwent parathyroidectomy posttransplant (n= 15; new-PTX group). Persistent hyperparathyroidism was defined as serum-corrected calcium levels of > 10.2 mg/dL (at least twice in a 6-mo period) and parathyroid hormone (PTH) level of > 150 pg/mL at 6 months posttransplant. Patients provided written informed consent for analyses of anonymized data.
All recipients received basiliximab and methyl-prednisolone for induction therapy. Maintenance treatment consisted of calcineurin inhibitors (tacrolimus and cyclosporine) or everolimus combined with mycophenolic acid or mycophenolate mofetil and corticosteroids. The distribution of immunosuppressive regimens was similar between the 3 groups (Table 1).
Demographic and laboratory data of recipients were obtained from medical records. We recorded serum levels of creatinine, calcium, and phosphorus before transplant and at 1, 3, 6, and 12 months after transplant. Intact PTH levels before transplant and at 6 and 12 months after transplant were also noted. Estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration formula.10
Statistical analyses were performed with SPSS software (SPSS: An IBM Company, version 23.0, IBM Corporation, Armonk, NY, USA). Shapiro-Wilks test was used to assess the normal distribution of variables. Categorical variables are presented as frequency with corresponding percentages. Continuous variables are presented as means and standard deviation or median (with minimum to maximum). Pearson chi-square or Fisher exact tests were used to test differences in proportion of the categorical variables. The Wilcoxon signed rank test was used for comparisons within groups. Comparisons between groups were made using Kruskal-Wallis test. We conducted 2 group comparisons using the Mann-Whitney test. To compare values before and after treatment, percent changes were computed. Pearson correlation was used to analyze the data. P < .05 was considered statistically significant.
We observed no significant differences in age, sex distribution, posttransplant follow-up time, and immunosuppressive regimens among the 3 groups. The percentage of transplants from living donors in the non-PTX group (55.4%) was significantly higher than in the new-PTX (20%) and old-PTX (16.7%) groups (P = .001). Time on dialysis was also significantly shorter in the non-PTX group (P < .001). In the 3 groups, there were no significant differences in biopsy-proven acute rejection, delayed graft function, chronic allograft nephropathy, cytomegalovirus infection, and BK nephropathy rates (Table 1).
Compared with pretransplant results, serum calcium levels were significantly increased at 3, 6, and 12 months after transplant (P < .001) in the non-PTX group, at 1 (P = .048) and 3 months (P = .001) after transplant in the new-PTX group, and at 3 (P = .002) and 6 months (P = .003) after transplant in the old-PTX group (Table 2). Serum calcium levels were significantly decreased at the last visit in the new-PTX group (P = .027). Serum calcium levels at baseline and at 1, 3, 6, and 12 months in the new-PTX group were higher than those shown in the old-PTX and non-PTX groups (P < .001). Changes in calcium levels among groups were significant at 1 month (P = .019), 3 months (P = .026), and at last visit (P = .007). Serum calcium elevation rates in the non-PTX group (1.05%) were lower at month 1 compared with rates shown in the new-PTX (8.49%) and old-PTX groups (6.65%; P = .040). They were also lower (4.30%) than rates at month 3 in the new-PTX group (10.0%; P = .020). Furthermore, calcium decline rate (-10.1%) at last visit in the new-PTX group was significantly prominent compared with the non-PTX (0%; P = .003) and the old-PTX (3.9%; P = .010) groups (P = .007).
Baseline serum phosphorus levels among the groups were similar (Table 2). Serum phosphorus levels were significantly decreased after transplant compared with pretransplant from month 1 posttransplant throughout to the last visit (P < .01) in all groups. Serum phosphorus levels at 3, 6, and 12 months in the new-PTX group were lower than those shown in the old-PTX and non-PTX groups (P < .001). Changes in phosphorus levels in the groups were significant at months 1 (P = .036), 3 (P = .002), 6 (P = .001), and 12 (P = .001) and at last visit (P = .016). Declining phosphorus levels in the new-PTX group were significantly more prominent compared with those shown in the non-PTX group from month 1 to the last visit (-64.6% vs -51.3%, P = .010; -59% vs -41.4%, P < .001; -59.8% vs -38.3%, P < .001; -55.3% vs -33.3%, P < .001; -47.1 vs 32.7%, P = .010, respectively) and old-PTX groups at month 3 (-59% vs -49.9%; P = .006) and month 6 (-59.8% vs -42.6%; P = .027).
After transplant, serum PTH levels were signi-ficantly decreased from month 6 posttransplant to the last visit in the new-PTX (P < .01) and non-PTX (P < .001) groups compared with levels pretransplant; however, decreases in the old-PTX group were not significant (Table 2). Serum PTH levels at baseline and at months 6 and 12 in the new-PTX group were higher than those shown in the old-PTX and non-PTX groups (P < .001). Serum PTH levels at months 6 and 12 were also lower in the old-PTX versus the non-PTX group (P < .05). Changes in PTH levels among the groups were significant at last visit (P < .001). Serum PTH levels decreased significantly more in the new-PTX group (-88.9%) than in the non-PTX (-57.3%) and old-PTX groups (-50.6%) at last visit (P < .001 and P = .003, respectively).
When compared with results shown at month 1 posttransplant, serum creatinine levels were significantly increased at month 6 (P < .001), at month 12 (P = .011), and at last visit (P < .001) in the non-PTX group and at month 6 (P = .045) in new-PTX group (Table 2). Serum creatinine levels at month 3 in the old-PTX group were lower than in the non-PTX group (P < .05). Compared with results shown at month 1 posttransplant, decreases in eGFR values were significant at month 6 (P < .001), at month 12 (P = .004), and at last visit (P < .001) in the non-PTX group and at month 6 (P = .010) and month 12 (P = .011) in the new-PTX group. At months 3, 6, and 12 and at last visit, eGFR results in the old-PTX group were higher than in the non-PTX group (P < .05) and higher at months 6 and 12 than in the new-PTX group (P < .05). Changes in serum creatinine and eGFR values among groups were not significant.
In all 3 groups, the prevalence of hypercalcemia (> 10.2 mg/dL), hypocalcemia (< 8.5 mg/dL), hyperphosphatemia (> 4 mg/dL), and hypo-phosphatemia (< 2.3 mg/dL) at pretransplant and at months 1, 3, 6, and 12 posttransplant and at last visit were 10.4%, 9.3%, 17.7%, 17.7%, 19%, and 5.2%; 12.9%, 9.7%, 2.8%, 0.8%, 2.5%, and 6.0%; 74.9%, 3%, 2.3%, 3.5%, 10.4%, and 9.5%; and 0.5%, 42.9%, 16.8%, 7%, 1.7%, and 4.1%, respectively (Figure 1).
The prevalence of hypocalcemia in the old-PTX group at 12 months posttransplant was higher than that shown in the non-PTX group (16.7% vs 1.8%; P = .033). In the new-PTX group, the prevalence of hypercalcemia at pretransplant and at 1, 3, 6, and 12 months posttransplant and the prevalence of hypophosphatemia at 3, 6, and 12 months were higher than those shown in the old-PTX and non-PTX groups (P < .001). At last visit, there were no differences in prevalences of hypercalcemia, hypocalcemia, hyperphosphatemia, and hypophos-phatemia among the groups (Table 3).
The prevalence of PHPT (>150 pg/mL) in all patients at pretransplant and at months 6 and 12 posttransplant and at last visit were 82.3%, 63.5%, 53.0%, and 56.6%, respectively (Figure 1). The prevalence of PHPT in the old-PTX group at pretransplant was lower at 50% than results shown in the non-PTX and new-PTX groups (82.9% and 100%, respectively; P < .01). At 6 and 12 months, the PHPT rates (93.3% and 86.7%) in the new-PTX group were higher than those shown in the non-PTX (64%, P = .022 and 52.3%, P = .01) and the old-PTX (16.7%, P < .001 and 25%, P = .001) groups, respectively. At month 6, the PHPT rate in the non-PTX group was higher than that shown in the old-PTX group (P = .001). There was no significant difference between PHPT rates in groups at last visit.
The PTX procedure was performed post-transplant in 15 recipients, with mean time between transplant and PTX of 25.8 ± 23.6 months. Of these 15 recipients, 2 had surgery 5 months after transplant and 1 had surgery 7 months after transplant because of severe hypercalcemia (11.4, 12.3, and 11.2 mg/dL, respectively). Seven of the 15 recipients used cinacalcet because of persistent hypercalcemia before PTX.
During the study period, there were no significant differences regarding graft loss and patient mortality rates among the 3 groups (P > .05). Three recipients in the new-PTX group had graft loss due to chronic allograft nephropathy. One patient who had stage 4 chronic kidney disease before PTX returned to dialysis 1 month after PTX surgery; the others returned after 23 and 59 months. In the non-PTX group, 22 recipients lost their grafts because of chronic allograft nephropathy, 2 because of BK virus nephropathy, 1 for urologic reason, and 1 because of nonadherence to treatment. There were no reports of graft loss in the old-PTX group. In total, 1 patient (8.3%) in the old-PTX group died because of pneumonia and 7 patients (3.6%) in the non-PTX group died because of different causes (pneumonia in 5 and acute myocardial infarction in 2 patients). There were no significant differences regarding graft loss (P = .699) and mortality (P = .469) among the groups (Table 1).
Serum creatinine levels at month 12 for all patients were correlated with PTH at month 12 (r = 0.374, P < .001) and were correlated with percent changes in PTH at month 6 (r = 0.197, P = .038), month 12 (r = 0.175, P = .011), and last visit (r = 0.140, P = .048). Serum creatinine levels at month 12 were negatively correlated with follow-up time (r = -0.125, P = .047), dialysis duration (r = -0.194, P = .005), and percent change in calcium at months 3 (r = -0.172, P = .007), 6 (r = -0.151, P = .018), and 12 (r = -0.161, P = .011). Serum creatinine levels at last visit for all patients were correlated with PTH levels at month 12 (r = 0.204, P = .003) and at last visit (r = 0.217,P < .001) and phosphorus levels at last visit (r = 0.237, P < .001). Serum creatinine levels at last visit were negatively correlated with serum calcium levels at last visit (r = -0.147, P = .032).
At month 12, eGFR values for all patients were correlated with dialysis duration (r = 0.149, P = .034). In addition, eGFR values at month 12 were negatively correlated with transplant age (r = -0.158, P = .013) and PTH levels at month 12 (r = -0.266, P < .001) and at last visit (r = -0.152; P = .028). Values for eGFR at last visit for all patients were negatively correlated with transplant age (r = -0.175, P = .011) and PTH levels at month 12 (r = -0.187, P = .007) and at last visit (r = -0.192, P = .005).
Dialysis duration of all patients was correlated with transplant age (r = 0.198, P = .004), pretransplant albumin level (r = 0.175, P = .034), pretransplant PTH level (r = 0.235, P = .001), and percent change in eGFR at month 12 (r = 0.167, P = .017) and at last visit (r = 0.170, P = .023). Dialysis duration was negatively correlated with percent change in creatinine at month 12 (r = -0.158, P = .017). There were no significant correlations between changes in creatinine and eGFR at month 12 and at last visit versus other studied parameters.
Treatment of PHPT and hypercalcemia is essential for recipients to allow successful graft outcomes and to avoid its deleterious effects. In our study, we showed that parathyroidectomy after transplant resulted in a sustained decrease in serum calcium and PTH levels without any significant risk of graft failure. However, graft function was better in recipients who had parathyroidectomy before transplant. We also found that graft function was inferior in recipients without parathyroidectomy versus those who had parathy-roidectomy before transplant.
Serum calcium, phosphorus, and PTH levels tend to normalize within about 1 year after transplant.11 Although PTH levels decline within 3 months after a successful kidney transplant, approximately 50% of recipients have risk of PHPT, and the severity of preexisting PHPT largely affects its persistence after kidney transplant.12,13 The incidence of recipients with abnormal PTH levels decreased from 38% at 6 months after transplant to 21.7% at 5-year follow-up, including those who had excellent graft function.14 In our study, we observed that serum calcium, phosphorus, and PTH levels were generally within normal ranges and ameliorated compared with pretransplant levels except for those in the new-PTX group. We also observed that prevalences of hypercalcemia (19%) and PHPT (53.3%) 12 months after transplant were in accordance with the literature. Furthermore, PHPT and hypercalcemia may be associated with important complications, including bone loss, fractures, vascular calcification, graft dysfunction, and graft loss.15-17 Longer dialysis duration (> 6 y), higher CaxP products [>55 (mg/dL)2], and cinacalcet use before transplant are independent risk factors for PHPT in recipients.18 High pretransplant PTH levels have a significant and nonquadratic exponential influence on the risk of graft failure censored for death. At a PTH level of 7 pmol/L, graft failure risk is 1; however, at a level of 90 pmol/L, the risk for graft failure doubles.19 Compatible with the previous findings, in our cohort, renal replacement duration and high pretransplant PTH and calcium levels were positively correlated with PTH and calcium at first year posttransplant.
Parathyroidectomy and cinacalcet after transplant are treatment options for PHPT. In addition to the deleterious effects of hypercalcemia and PHPT on graft function, there are concerns about graft outcomes associated with PTX and cinacalcet use (Table 4). Although PTX and cinacalcet therapy can ameliorate PTH levels, graft function may worsen due to hemodynamic changes mediated by decreased PTH. Experimental studies have shown vasodilatory effects of PTH on preglomerular renal vasculature. After parathyroidectomy, decreases in PTH levels may contribute to decreased eGFR because of the lost vasodilatory effects of PTH.20,21 Some studies have shown that parathyroidectomy is associated with reduced graft function and survival in recipients.6,7,22 According to hemodynamic changes, serum creatinine levels tend to increase in the early PTX period.4,8,23 In the long term, serum creatinine levels showed no significant detrimental effects on graft function.23-25 In our study, serum creatinine levels in new-PTX group did not significantly change after PTX. Graft loss occurred in 3 recipients in whom we did not consider that it was associated with parathyroidectomy. Although serum calcium and PTH levels were similar between all groups at last visit, revealing a sustained amelioration with these parameters, it was remarkable that eGFR levels were significantly better in the old-PTX group.
Knowledge about parathyroidectomy timing in recipients is insufficient, and there are limited studies about this issue. Because of concerns about graft outcomes with high PTH levels, treatment approaches are essential for patients before and after transplant. Callender and associates9 showed that high pre-transplant PTH levels were associated with graft failure. In this study, recipients who had para-thyroidectomy procedures pretransplant had less risk of graft failure. Jeon and associates8 compared recipients who had parathyroidectomy before and after transplant. They concluded that pretransplant parathyroidectomy is advantageous with less risk of graft dysfunction compared with posttransplant parathyroidectomy. In that study,8 early parathy-roidectomy, defined as occurring in the first year of transplant, was associated with deteriorated graft function compared with pretransplant and late-PTX groups.
We found that eGFR levels showed better graft function in the old-PTX group compared with the new-PTX and non-PTX groups. Low eGFR levels in the non-PTX group were also an engrossing finding. Serum calcium levels were within normal ranges, but there were markedly higher values at the maximum side of the range. Unrecognized and untreated persistent hypercalcemia could lead to graft deterioration. However, the amelioration of hyper-parathyroidism before transplant may preserve possible posttransplant hypercalcemia and provide better graft function. In addition, the high levels of PTH in the non-PTX group were controversial, whether it was the cause or the consequence of decreased eGFR. Decreased median PTH levels in the non-PTX group posttransplant from 374 to 154 pg/mL indicated that high PTH levels are independent of decreased eGFR. We considered that unrecognized hypercalcemia after transplant and disregard of hyperparathyroidism treatment while on the wait list were possible reasons for worse graft outcomes of non-PTX recipients. Furthermore, events affecting graft outcomes, including acute rejection, delayed graft function, chronic allograft nephropathy, cytomegalovirus disease, and BK nephropathy, were evaluated, and their ratios were similar among the groups.
There are several limitations of our study. First, it is a retrospective study. Because a bone mineral disorder has dynamic status and heterogeneity among patients, bone biopsy is a criterion standard procedure to uncover precise bone status. We did not perform bone biopsy; therefore, we could not detect the exact causes of hypercalcemia like adynamic bone disease. Furthermore, the distribution of participant numbers was unbalanced among the groups, which may have negatively affected the statistical results.
In conclusion, PHPT remains important in recipients and has an unanswered and controversial management approach. Although hypercalcemia and hyperparathyroidism can resolve with parathy-roidectomy after transplant, parathyroidectomy before transplant is associated with better graft function. According to our findings, to obtain appropriate treatment options for PHPT, candidates should be evaluated before transplant and PTX should be performed in appropriate patients.
DOI : 10.6002/ect.2018.0221
From the 1Department of Nephrology and the 2Department of Endocrinology, Uludag
University Medical School, Bursa, Turkey
Acknowledgements: The authors have no sources of funding for this study and have no conflicts of interest to declare.
Corresponding author: Aysegul Oruc, Department of Nephrology, Uludag University Medical School, 16059 Bursa, Turkey
Phone: +90 224 2951425
Table 1. Patient Characteristics
Table 2. Changes in Laboratory Parameters in the Groups
Table 3. Prevalences of Hypercalcemia, Hypocalcemia, Hyperphosphatemia, and Hypophosphatemia in the Groups
Table 4. Review of Literature
Figure 1. Prevalence of Hyperparathyroidism and Hypercalcemia in Groups