Predictors of tertiary hyperparathyroidism: Who will benefit from parathyroidectomy? Lindel C. Dewberry, BS, Sudha Tata, MD, Sharon Graves, MD, Collin J. Weber, MD, and Jyotirmay Sharma, MD, Atlanta, GA
Background. Tertiary hyperparathyroidism (38HPT) is hyperparathyroidism with hypercalcemia after renal transplantation. With unclear guidelines for parathyroidectomy (PTX), this study aims to determine which renal transplant patients develop 38HPT and would benefit from PTX. Methods. We performed a retrospective review of patients who received a renal transplant between 1994 and 2013; 105 patients who underwent near total PTX (NTPTX) were compared with 180 renal transplant control patients who did not undergo NTPTX. Results. Calcium and PTH varied significantly between groups (P < .001). One year before transplant, the mean serum calcium was 9.7 ± 1.1 mg/dL in the NTPTX group versus 9.1 ± 0.9 mg/dL in the control group (P < .01). One month after transplant, the mean calcium in the NTPTX group was 10.4 ± 1.1 versus 9.4 ± 0.6 mg/dL in the control group (P < .001). One year before renal transplant, the median serum PTH level was 723 pg/mL (range, 557–919) in the NTPTX group versus 212 pg/ mL (range, 160–439) in the control group (P < .01). One-month post renal transplant, the NTPTX group had a median PTH of 351 pg/mL (range, 199–497) versus 112 pg/mL (range, 73–178) pg/mL in the control group (P < .01). Conclusion. Before and after renal transplantation, PTH and calcium levels can serve as predictors of 38HPT. (Surgery 2014;156:1631-7.) From the Department of Surgery, Emory University School of Medicine, Atlanta, GA
TERTIARY HYPERPARATHYROIDISM (38HPT) is described as persistent hyperparathyroidism with hypercalcemia despite correction of secondary hyperparathyroidism (from chronic renal failure) with renal transplantation. The reported incidence of post renal transplant hyperparathyroidism ranges from 25 to 50% at 1 year after transplant1,2 and about 17% at 4 years after transplant.2 The average time for parathyroid levels to return to normal after transplantation varies between 3 and 6 months.3 Calcium levels post renal transplant are also highly variable. The natural history of posttransplant calcium levels can be categorized into 3 distinct groups: Eucalcemia, hypercalcemia, and fluctuating calcium levels.
Disclosure of Financial Interests and Potential Conflicts of Interest: None. Accepted for publication August 21, 2014. Reprint requests: Jyotirmay Sharma, MD, Department of Surgery, Emory University, The Emory Clinic, Building A, 3rd Floor, 1365 Clifton Road, Atlanta, GA 30322. E-mail: Jsharm3@emory. edu. 0039-6060/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.surg.2014.08.070
These varying calcium levels present an obstacle to identifying patients with true 38HPT as opposed to patients who are still undergoing gland involution. The incidence of true 38HPT post renal transplant ranges from 0.5 to 5.6%.4-9 Tertiary HPT leads to decreased bone mineral density, changes in mental status, myopathies, and, importantly, calcification of the renal allograft, which may affect allograft function.6,10 Parathyroidectomy (PTX) is the only definitive treatment option for 38HPT.11 Current indications for PTX include highly increased, persistent hypercalcemia, symptomatic hypercalcemia, increased parathyroid hormone levels, hypophosphatemia, decreased bone mineral density, or kidney function decline associated with hyperparathyroidism,5 but it is unclear as to what calcium or parathyroid hormone level and at what time point after renal transplantation one should refer a patient for PTX.11 At our institution, patients are referred for PTX by the transplant nephrology department, and their criteria for referral are based on the level and persistence (>3 months) of serum calcium and increases in PTH. The normal ranges of PTH vary and are determined by the patient’s glomerular filtration rate (GFR). SURGERY 1631
1632 Dewberry et al
Studies have shown similar renal graft survival between patients who undergo PTX and control groups, even though some studies report a decrease in GFR in patients who have undergone PTX.12 Benefits of PTX may include improved patient survival, improved bone mineral density, and alleviation of symptoms.6,13 Previous studies at our institution demonstrated improved patient outcomes, specifically mortality, in secondary hyperparathyroidism after NTPTX.14-17 With the lack of consensus guidelines regarding referral of 38HPT patients for PTX, this study aimed to identify factors that can be used as predictors of 38HPT to reveal patients who may benefit from PTX earlier. MATERIALS AND METHODS Patient selection. The Institutional Review Board of Emory University approved this retrospective cohort study. A database of 38HPT patients consisted of 105 patients having undergone NTPTX between 1994 and 2013 at Emory University. Patients lacking documented laboratory values for calcium and parathyroid levels were excluded. As a comparison group for the NTPTX patients, controls who did not undergo PTX were time matched, having undergone renal transplant during the same time period between 1994 and 2013. Review of the Organ Transplant Tracking Record database entered into the hospital electronic medical record yielded a total of 3,978 patients. Of these, 250 patients were selected randomly. Patients having undergone PTX or lacking documented serum calcium and parathyroid hormone levels were excluded, yielding a total of 180 control patients. Patient characteristics (sex, age at time of renal transplant, race, cause of renal failure, duration of dialysis [vintage], cinacalcet use, and immunosuppressive regimen), and clinical outcomes were obtained from the electronic medical record. Dialysis vintage was calculated by totaling all years of dialysis, including peritoneal dialysis and hemodialysis. NTPTX technique. The procedure performed at our institution for 38HPT is a NTPTX with intraoperative PTH monitoring. This technique is a variant of a subtotal PTX, where a vascularized remnant of a parathyroid gland is left in situ. The remnant approximates the size of 2 normal parathyroid glands (80–100 mg). Remnant size is calculated intraoperatively by the following calculation: length 3 width 3 height 3 p/6. This operative technique has been described previously.16 Surgical characteristics measured included
Surgery December 2014
parathyroid gland weight, pathology, and size of remnant. OUTCOME ASSESSMENTS Patient outcomes evaluated included measurements of serum calcium, phosphorus, PTH, and 25-OH vitamin D levels, bone density, and GFR at various time points (1, 3, and 6 months, and 1 year, latest follow-up) before and after renal transplantation. Other outcome measures included all-cause mortality, cardiovascular complications, and renal allograft survival. Renal allograft function was assessed by calculating the GFR using the Modification of Diet in Renal Disease equation and by reviewing pathology results (chronic nephropathy, acute rejection, borderline rejection, chronic rejection, nephritis, or nephrotoxicity) from renal allograft biopsies. Renal allograft failure was defined as resumption of dialysis. Post renal transplant cardiac complications were evaluated by reviewing the medical record for incidence of ST segment elevation myocardial infarction, non–ST segment elevation myocardial infarction, coronary artery bypass graft, or percutaneous coronary intervention. Mortality was evaluated by reviewing the medical records, which reflected data from the Organ Transplant Tracking Record database. Data analysis. For continuous, normally distributed variables, mean and standard deviation were used to summarize data. For continuous, nonnormally distributed variables, median values and interquartile ranges were used to summarize data. Percentages were used for categorical variables. Pre and postoperative clinical values were compared using paired t tests. Differences between groups were compared using independent t tests. The odds ratio of undergoing NTPTX for different thresholds of PTH and calcium were calculated using logistic regression. SPSS software was used for all data analysis. RESULTS Baseline characteristics. We compared 105 patients in the NTPTX case group with 180 patients in the control group. The mean time between renal transplant and PTX was 5.5 years. Causes for this ostensible delay were multifactorial. The referral and practice patterns changed over the time course included in the study because the time duration spanned 19 years. For example, between 2010 and 2014 the median time between renal transplant and PTX was 2 years. The mean age in the NTPTX group was 41.8 ± 11.5 versus 46.5 ± 11.5 years in the control group (P < .01; Table I). There was no difference
Dewberry et al 1633
Surgery Volume 156, Number 6
Table I. Baseline characteristics of near total parathyroidectomy (NTPTX) cases and controls Characteristic
NTPTX
Control patients
Age at renal 41.8 ± 11.5 46.5 ± 11.5 transplant (y) Male (%) 1.1:1 1.46:1 Race (%) African American 58.1 48.3 Caucasian 34.3 45 Hispanic 3.8 4.4 Asian 1 0 Dialysis duration (y) 6 ± 4.9 2.8 ± 2.8 Cause of renal failure (%) Hypertension 43.3 22.8 Diabetes 6.7 28.9 Glomerulonephritis 32.4 32.8 Polycystic kidney 5.7 7.8 disease Cinacalcet use (%) 24.7 8.3 Immunosuppression (%) Mycophenolic acid, 47.5 92.5 tacrolimus, prednisone Mycophenolic acid, 31.7 4 cyclosporine, prednisone
P value
Background. Tertiary hyperparathyroidism (38HPT) is hyperparathyroidism with hypercalcemia after renal transplantation. With unclear guidelines for parathyroidectomy (PTX), this study aims to determine which renal transplant patients develop 38HPT and would benefit from PTX. Methods. We performed a retrospective review of patients who received a renal transplant between 1994 and 2013; 105 patients who underwent near total PTX (NTPTX) were compared with 180 renal transplant control patients who did not undergo NTPTX. Results. Calcium and PTH varied significantly between groups (P < .001). One year before transplant, the mean serum calcium was 9.7 ± 1.1 mg/dL in the NTPTX group versus 9.1 ± 0.9 mg/dL in the control group (P < .01). One month after transplant, the mean calcium in the NTPTX group was 10.4 ± 1.1 versus 9.4 ± 0.6 mg/dL in the control group (P < .001). One year before renal transplant, the median serum PTH level was 723 pg/mL (range, 557–919) in the NTPTX group versus 212 pg/ mL (range, 160–439) in the control group (P < .01). One-month post renal transplant, the NTPTX group had a median PTH of 351 pg/mL (range, 199–497) versus 112 pg/mL (range, 73–178) pg/mL in the control group (P < .01). Conclusion. Before and after renal transplantation, PTH and calcium levels can serve as predictors of 38HPT. (Surgery 2014;156:1631-7.) From the Department of Surgery, Emory University School of Medicine, Atlanta, GA
TERTIARY HYPERPARATHYROIDISM (38HPT) is described as persistent hyperparathyroidism with hypercalcemia despite correction of secondary hyperparathyroidism (from chronic renal failure) with renal transplantation. The reported incidence of post renal transplant hyperparathyroidism ranges from 25 to 50% at 1 year after transplant1,2 and about 17% at 4 years after transplant.2 The average time for parathyroid levels to return to normal after transplantation varies between 3 and 6 months.3 Calcium levels post renal transplant are also highly variable. The natural history of posttransplant calcium levels can be categorized into 3 distinct groups: Eucalcemia, hypercalcemia, and fluctuating calcium levels.
Disclosure of Financial Interests and Potential Conflicts of Interest: None. Accepted for publication August 21, 2014. Reprint requests: Jyotirmay Sharma, MD, Department of Surgery, Emory University, The Emory Clinic, Building A, 3rd Floor, 1365 Clifton Road, Atlanta, GA 30322. E-mail: Jsharm3@emory. edu. 0039-6060/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.surg.2014.08.070
These varying calcium levels present an obstacle to identifying patients with true 38HPT as opposed to patients who are still undergoing gland involution. The incidence of true 38HPT post renal transplant ranges from 0.5 to 5.6%.4-9 Tertiary HPT leads to decreased bone mineral density, changes in mental status, myopathies, and, importantly, calcification of the renal allograft, which may affect allograft function.6,10 Parathyroidectomy (PTX) is the only definitive treatment option for 38HPT.11 Current indications for PTX include highly increased, persistent hypercalcemia, symptomatic hypercalcemia, increased parathyroid hormone levels, hypophosphatemia, decreased bone mineral density, or kidney function decline associated with hyperparathyroidism,5 but it is unclear as to what calcium or parathyroid hormone level and at what time point after renal transplantation one should refer a patient for PTX.11 At our institution, patients are referred for PTX by the transplant nephrology department, and their criteria for referral are based on the level and persistence (>3 months) of serum calcium and increases in PTH. The normal ranges of PTH vary and are determined by the patient’s glomerular filtration rate (GFR). SURGERY 1631
1632 Dewberry et al
Studies have shown similar renal graft survival between patients who undergo PTX and control groups, even though some studies report a decrease in GFR in patients who have undergone PTX.12 Benefits of PTX may include improved patient survival, improved bone mineral density, and alleviation of symptoms.6,13 Previous studies at our institution demonstrated improved patient outcomes, specifically mortality, in secondary hyperparathyroidism after NTPTX.14-17 With the lack of consensus guidelines regarding referral of 38HPT patients for PTX, this study aimed to identify factors that can be used as predictors of 38HPT to reveal patients who may benefit from PTX earlier. MATERIALS AND METHODS Patient selection. The Institutional Review Board of Emory University approved this retrospective cohort study. A database of 38HPT patients consisted of 105 patients having undergone NTPTX between 1994 and 2013 at Emory University. Patients lacking documented laboratory values for calcium and parathyroid levels were excluded. As a comparison group for the NTPTX patients, controls who did not undergo PTX were time matched, having undergone renal transplant during the same time period between 1994 and 2013. Review of the Organ Transplant Tracking Record database entered into the hospital electronic medical record yielded a total of 3,978 patients. Of these, 250 patients were selected randomly. Patients having undergone PTX or lacking documented serum calcium and parathyroid hormone levels were excluded, yielding a total of 180 control patients. Patient characteristics (sex, age at time of renal transplant, race, cause of renal failure, duration of dialysis [vintage], cinacalcet use, and immunosuppressive regimen), and clinical outcomes were obtained from the electronic medical record. Dialysis vintage was calculated by totaling all years of dialysis, including peritoneal dialysis and hemodialysis. NTPTX technique. The procedure performed at our institution for 38HPT is a NTPTX with intraoperative PTH monitoring. This technique is a variant of a subtotal PTX, where a vascularized remnant of a parathyroid gland is left in situ. The remnant approximates the size of 2 normal parathyroid glands (80–100 mg). Remnant size is calculated intraoperatively by the following calculation: length 3 width 3 height 3 p/6. This operative technique has been described previously.16 Surgical characteristics measured included
Surgery December 2014
parathyroid gland weight, pathology, and size of remnant. OUTCOME ASSESSMENTS Patient outcomes evaluated included measurements of serum calcium, phosphorus, PTH, and 25-OH vitamin D levels, bone density, and GFR at various time points (1, 3, and 6 months, and 1 year, latest follow-up) before and after renal transplantation. Other outcome measures included all-cause mortality, cardiovascular complications, and renal allograft survival. Renal allograft function was assessed by calculating the GFR using the Modification of Diet in Renal Disease equation and by reviewing pathology results (chronic nephropathy, acute rejection, borderline rejection, chronic rejection, nephritis, or nephrotoxicity) from renal allograft biopsies. Renal allograft failure was defined as resumption of dialysis. Post renal transplant cardiac complications were evaluated by reviewing the medical record for incidence of ST segment elevation myocardial infarction, non–ST segment elevation myocardial infarction, coronary artery bypass graft, or percutaneous coronary intervention. Mortality was evaluated by reviewing the medical records, which reflected data from the Organ Transplant Tracking Record database. Data analysis. For continuous, normally distributed variables, mean and standard deviation were used to summarize data. For continuous, nonnormally distributed variables, median values and interquartile ranges were used to summarize data. Percentages were used for categorical variables. Pre and postoperative clinical values were compared using paired t tests. Differences between groups were compared using independent t tests. The odds ratio of undergoing NTPTX for different thresholds of PTH and calcium were calculated using logistic regression. SPSS software was used for all data analysis. RESULTS Baseline characteristics. We compared 105 patients in the NTPTX case group with 180 patients in the control group. The mean time between renal transplant and PTX was 5.5 years. Causes for this ostensible delay were multifactorial. The referral and practice patterns changed over the time course included in the study because the time duration spanned 19 years. For example, between 2010 and 2014 the median time between renal transplant and PTX was 2 years. The mean age in the NTPTX group was 41.8 ± 11.5 versus 46.5 ± 11.5 years in the control group (P < .01; Table I). There was no difference
Dewberry et al 1633
Surgery Volume 156, Number 6
Table I. Baseline characteristics of near total parathyroidectomy (NTPTX) cases and controls Characteristic
NTPTX
Control patients
Age at renal 41.8 ± 11.5 46.5 ± 11.5 transplant (y) Male (%) 1.1:1 1.46:1 Race (%) African American 58.1 48.3 Caucasian 34.3 45 Hispanic 3.8 4.4 Asian 1 0 Dialysis duration (y) 6 ± 4.9 2.8 ± 2.8 Cause of renal failure (%) Hypertension 43.3 22.8 Diabetes 6.7 28.9 Glomerulonephritis 32.4 32.8 Polycystic kidney 5.7 7.8 disease Cinacalcet use (%) 24.7 8.3 Immunosuppression (%) Mycophenolic acid, 47.5 92.5 tacrolimus, prednisone Mycophenolic acid, 31.7 4 cyclosporine, prednisone
P value
