The American Journal of Surgery (2015) 210, 123-128
Clinical Science
Low 24-hour urine calcium levels in patients with sporadic primary hyperparathyroidism: is further evaluation warranted prior to parathyroidectomy? Kathleen O’Connell, M.D.a, Tina W. Yen, M.D.a, Joseph Shaker, M.D.b, Stuart D. Wilson, M.D.a, Douglas B. Evans, M.D.a, Tracy S. Wang, M.D., M.P.H.a,* a
Department of Surgery, Division of Surgical Oncology, bDepartment of Medicine, Division of Clinical Endocrinology, Metabolism, and Nutrition, Medical College of Wisconsin, 9200 W. Wisconsin Avenue, Milwaukee, WI 53226, USA
KEYWORDS: 24-Hour urine calcium; Familial hypocalciuric hypercalcemia; Primary hyperparathyroidism
Abstract BACKGROUND: Low 24-hour urine calcium (uCa) levels in patients with primary hyperparathyroidism (pHPT) raise concern for familial hypocalciuric hypercalcemia. This study evaluated patients with a low 24-hour uCa level for potential differences that may guide the extent of preoperative evaluation needed. METHODS: A retrospective review was conducted of 1,139 sporadic pHPT patients who underwent parathyroidectomy between December 1999 and May 2011. RESULTS: Of the 54 (5%) patients with greater than or equal to one low 24-hour uCa (,100 mg), 28 (52%) patients had only one low level, 9 (17%) had multiple low levels, and 17 (31%) had a repeat 24-hour uCa greater than 100. In the latter group, 4 of the 9 (53%) patients were on a thiazide and had normalization after cessation. Among the groups, differences existed only in serum creatinine (P 5 .0011) and glomerular filtration rate (P 5 .0007). CONCLUSION: This study suggests that sporadic pHPT patients with low 24-hour uCa levels may not require further evaluation with genetic testing for familial hypocalciuric hypercalcemia, especially if previous eucalcemia is documented. Ó 2015 Elsevier Inc. All rights reserved.
Primary hyperparathyroidism (pHPT) is the most common cause of hypercalcemia, with an estimated incidence of 22 cases per 100,000 person years in the United States.1 pHPT is characterized by increased serum calcium levels The authors declare no conflicts of interest. * Corresponding author. Tel.: 11-414-805-5755; fax: 11-414-805-5771. E-mail address: [email protected] Manuscript received February 1, 2014; revised manuscript April 29, 2014 0002-9610/$ - see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjsurg.2014.09.030
secondary to inappropriately elevated levels of parathyroid hormone (PTH); definitive therapy with curative parathyroidectomy may prevent end-organ damage, such as nephrolithiasis and osteodystrophy. The majority of patients with pHPT have sporadic disease; less commonly, pHPT occurs as a component of a familial syndrome, such as multiple endocrine neoplasia (MEN) types 1 and 2.2 Familial hypocalciuric hypercalcemia (FHH) accounts for approximately 2% of all cases of patients with hypercalcemia.3 FHH1 is an autosomal dominant disorder caused
124
The American Journal of Surgery, Vol 210, No 1, July 2015
by inactivating mutations in the calcium sensing receptor (CASR) gene, of which over 200 mutations have been identified to date.3 FHH1 accounts for about two third of FHH cases. FHH2 is caused by mutations of G protein alpha 114 and FHH3 is caused by mutations of Adapter Protein 2 Sigma 1.5 The biochemical profile of FHH is very similar to pHPT and is often difficult to differentiate. FHH is characterized by mild hypercalcemia, normal or elevated PTH levels, relatively low renal calcium excretion, and a family history of hypercalcemia.6 FHH is a benign condition that does not require treatment, whereas surgical intervention to reduce the associated long-term complications is usually advised in pHPT. Consequently, it is important to accurately differentiate whether hypercalcemia is secondary to pHPT versus FHH prior to planned parathyroidectomy. The most common measurement used to distinguish between patients with pHPT and FHH is the calcium-tocreatinine clearance ratio (CCCR), which is calculated as (24-hour urine calcium/plasma calcium)/(24-hour urine creatinine/plasma creatinine). The 2009 consensus panel on Guidelines in the Management of Asymptomatic pHPT set the CCCR threshold value at less than .01 for the diagnosis of FHH and greater than .02 for the diagnosis of pHPT.7 Although these cut-off values are not perfect and there is overlap between the diseases, the workup of patients with newly diagnosed hypercalcemia should include a 24-hour urine collection for evaluation of calcium excretion.8 Not all patients with low 24-hour urine calcium (uCa) levels have FHH; other etiologies for low 24-hour uCa levels include increasing age, use of thiazide diuretics, and chronic kidney disease. In the evaluation of patients with suspected pHPT and a low 24-hour uCa, evaluation of these causative factors (ie, discontinuation of thiazide diuretic), if possible, is recommended, followed by a repeat measurement of 24-hour uCa and creatinine levels. In patients in whom there is no other known etiology and for those with persistently low 24-hour uCa levels, genetic testing for CASR mutations is recommended, which can be time consuming and costly, particularly because FHH is rare and most patients will indeed have pHPT. Therefore, the objective of this study was to identify potential differences in patients with an initial low 24-hour uCa level that may guide the extent of evaluation recommended prior to parathyroidectomy.
secondary, or tertiary hyperparathyroidism, and those with no documentation of 24-hour uCa levels were excluded. The resulting cohort of patients was subsequently divided into 3 groups: patients with only one documented 24-hour uCa, which was low; patients with multiple low 24-hour uCa levels, and patients with a repeat 24-hour uCa greater than 100. Data collected included patient demographics, signs/symptoms of hyperparathyroidism, use of thiazide diuretic, laboratory values closest to surgery (serum calcium and creatinine, ionized calcium, PTH, vitamin D, glomerular filtration rate [GFR], uCa), postoperative laboratory values (serum calcium and PTH), number of parathyroid glands removed, and final histopathology results. Any family history of hypercalcemia or previous parathyroid disease is routinely obtained; for patients with suspicion of familial parathyroid disease, attempts are made to obtain serum calcium levels ideally from parents or other first-degree relatives. At our institution, initial evaluation of patients with hypercalcemia and suspected pHPT includes measurement of serum calcium, creatinine and glomerular filtration rate, intact PTH, and 25-hydroxy vitamin D levels, as well as a 24-hour uCa and creatinine collection. Prior to 2009, 24-hour uCa levels were obtained at the discretion of the referring provider or surgeon; since 2009, collection of 24-hour uCa and urine creatinine levels have been a routine part of the biochemical evaluation of patients with pHPT. All patients undergo preoperative localization studies including a high-quality cervical ultrasound and Sestamibi/technetium-99m scintigraphy (performed with single photon emission computed tomography after March 2009); an angiographic (‘‘four-dimensional’’) computed tomography scan is performed if the results of the former 2 studies are discordant or negative. Intraoperative PTH monitoring is routinely performed and our institutional protocol has previously been described;9 briefly, criterion for cure includes a greater than or equal to 50% decrease from the baseline or excisional PTH, whichever is higher, and returns to normal range, at 10 minutes postexcision of the abnormal parathyroid gland(s). Biochemical cure after parathyroidectomy is defined as serum calcium of less than 10.2 mg/dL at 6 months following parathyroidectomy. Patients with normocalcemia for at least 6 months after surgery who then had a serum calcium greater than or equal to 10.2 mg/dL were considered to have recurrent disease. Simple descriptive statistics were performed; a P value of less than or equal to .05 was considered statistically significant.
Methods This study was an Institutional Review Board–approved retrospective review of patients with pHPT who underwent parathyroidectomy by a single group of endocrine surgeons from December 1999 to December 2011. A prospectively collected parathyroid database was reviewed for patients with biochemically confirmed, sporadic pHPT who had at least one low 24-hour uCa measurement, defined as greater than 100 mg/24 hours. Patients with familial, recurrent, persistent,
Results Of the 1,035 patients in the database who underwent parathyroidectomy for sporadic pHPT, 735 (70%) patients had 24-hour uCa testing performed. Of these, 54 (7%) patients had at least one preoperative low 24-hour uCa measurement and comprise the study cohort. No patient
K. O’Connell et al.
Low urine calcium in primary hyperparathyroidism
had a family history suggestive of an inherited endocrinopathy. The median age of the patients was 65 years (range 33 to 85), and 44 (81%) were female. There were no statistically significant differences in patient demographics among the 3 groups (Table 1). One patient with repeat low levels had negative genetic testing for a CASR mutation. When comparing the groups by biochemical markers, patients with a 24-hour uCa that normalized had a lower median serum creatinine (.8 vs 1.2 vs 1.0; P 5 .0011) and higher GFR (86 vs 48 vs 62; P 5.0007) than those with a single low measurement or multiple low measurements (Table 2). Furthermore, when comparing 24-hour uCa levels obtained closest to the date of surgery, levels were highest in patients with a normalized 24-hour uCa level (119 mg/ 24 hours), compared with 72 mg/24 hours in the group with a single measurement and 38 mg/24 hours in the group with multiple low levels (P 5.0001). There was no difference in any other biochemical markers among the 3 groups. The use of a thiazide diuretic at the time of initial 24hour urine collection was documented in 12 patients. In the group with one low 24-hour uCa measurement, 2 patients (7%) were on a thiazide at the time of initial collection. In the group with multiple low 24-hour uCa measurements, 1 patient (11%) was on a thiazide at the time of collection. Of the patients with an initial low 24-hour uCa who subsequently normalized, 9 (53%) patients were on a thiazide at the time of initial collection. On repeat testing with cessation of the thiazide, 4 patients had uCa measurements within normal limits. There were no differences in pathologic results among the 3 groups (Table 3). Overall, 43 (80%) patients had single gland disease and 11 (20%) had multigland hyperplasia. This distribution remained consistent across the groups. The median follow-up for the entire cohort was 16.4 months; median follow-up was 13 months (range 1 to 80 months) for patients with one low 24-hour uCa measurement, 13 months (range 2 to 77 months) for patients with multiple low 24-hour uCa measurements, and 25 months (1 to 83 months) for patients with one low 24-hour uCa with subsequent normalization. Median serum calcium levels at the time of last visit were 9.5 (8.1 to 10.2), 9.8 (8.5 to 10.6), and 9.5 mg/dL (8.6 to 10.6) for the 3 groups, respectively. Only one patient, in
Table 1
125
the group with multiple low 24-hour uCa measurements, had a known recurrence.
Comments The majority of patients with pHPT will have sporadic disease, with a small subset of patients having an inherited syndrome, such as MEN1, MEN2, isolated familial hyperparathyroidism, or hyperparathyroidism jaw-tumor syndrome. In patients with elevated serum calcium levels undergoing evaluation for pHPT, FHH, for which parathyroidectomy is not indicated, should be excluded with measurement of 24-hour uCa levels. When the 24-hour uCa measurement is low (,100 mg/24 hours), concern is raised for FHH, and may require repeat urine biochemical testing and/or genetic evaluation. The results of this study suggest that, given the rarity of FHH, further evaluation with genetic testing may not be necessary prior to parathyroidectomy, particularly if eucalcemia is documented in the patient previously and/or in family members. The most common type of FHH (FHH1) results from an autosomal dominant transmission of a mutation in the CASR, a G protein-coupled receptor located on the surface of parathyroid chief cells, thyroidal C-cells, and renal tubular epithelial cells. Normally, this receptor binds to calcium, resulting in reduced PTH production and renal calcium reabsorption. With an inactivating mutation in one copy of the CASR gene, the CASR is relatively insensitive to serum calcium levels and requires higher calcium levels to activate the normal feedback mechanisms.10 The paradoxical finding of hypocalciuria in the setting of mild to moderate hypercalcemia, normal to mildly elevated PTH, and family history of hypercalcemia are characteristic findings of FHH because of loss-of-function of the CASR in the renal tubule. Although FHH is a benign condition that does not require treatment, it is important to distinguish FHH from pHPT, as surgical intervention in FHH patients is ineffective at reducing serum calcium levels. As such, FHH is sometimes recognized in patients with persistent hypercalcemia after parathyroidectomy. In a landmark study of a cohort of 67 patients with persistent hypercalcemia after failed parathyroidectomy,
Demographics of the 54 patients with low 24-hour uCa levels
Median age (years) Female Caucasian African American Hispanic Median BMI (kg/m2)
One low 24-hour uCa (n 5 28)
Multiple low 24-hour uCa (n 5 9)
One low 24-hour uCa with normalization (n 5 17)
67 22 25 2 1 27
65 7 7 1 1 29
60 15 9 7 1 31
(33–84) (79%) (89%) (7%) (4%) (16–46)
Value ranges are listed in parentheses. BMI 5 body mass index; uCa 5 urine calcium.
(58–85) (78%) (78%) (11%) (11%) (21–41)
(37–83) (88%) (53%) (41%) (6%) (21–42)
126
The American Journal of Surgery, Vol 210, No 1, July 2015
Table 2
Biochemical markers for patients with low 24-hour uCa levels One low 24-hour uCa (n 5 28)
Median preop calcium (mg/dL) Median maximum preop calcium (mg/dL) Ionized calcium (mmol/L) Preop PTH (pg/mL) Maximum PTH (pg/mL) Vitamin D (ng/mL) Serum creatinine (mg/dL) GFR (mL/min/1.73 sqm) Urine calcium closest to surgery CCCR (lowest) CCCR (highest)
10.9 (9.9–11.8) 11.4 (10.5–12.7) 1.44 117 145 28 1.2 48 72 .00745 .00745
(1.32–6.9) (27–463) (53–484) (7–54) (.8–1.9) (34–75) (3–95) (.00285–.0534) (.00285–.0534)
Multiple low 24-hour uCa (n 5 9)
One low 24-hour uCa with normalization (n 5 17)
10.6 (8.7–11.5) 11.2 (10.6–11.6)
10.9 (9.4–11.8) 11.3 (9.9–12.5)
1.43 94 177 35 1.0 62 38 .00551 .00706
(1.25–6.0) (67–428) (85–428) (19–42) (.8–1.7) (30–89) (15–72) (.00317–.00704) (.00315–.01097)
1.44 94 137 19 .8 86 119 .00542 .01065
P value .59 .30
(1.28–5.45) (7–311) (77–431) (7–50) (.6–1.8) (29–115) (42–471) (.0005–.0103) (.00229–.11932)
.82 .51 .74 .19 .0011 .0007 .0001 .26 .44
Value ranges are listed in parentheses. CCCR 5 calcium to creatinine clearance ratio; GFR 5 glomerular filtration rate; PTH 5 parathyroid hormone; sqm 5 square meters; uCa 5 urine calcium.
6 (9%) patients were subsequently identified as members of kindreds with FHH.11 These patients were found to have lower CCCR, as compared with the remaining patients with persistent hypercalcemia after parathyroidectomy; thus, the authors recommended assessment of uCa excretion in all hypercalcemic patients. A subsequent study has shown CCCR to have a sensitivity of 80% and specificity of 88% when the cut-off point was less than .01 for diagnosing patients with FHH, and a CCCR of less than .02 identified 53 of the 54 patients with FHH. However, the use of this test alone would potentially misdiagnose a significant proportion of patients with FHH, and thus the authors recommended use of the CCCR for initial FHH screening, followed by CASR gene analysis for patients with CCCR levels suggestive of FHH.12 Differentiating between pHPT and FHH is often not straightforward, as there is considerable overlap in both the clinical and biochemical features of these disease processes. In general, patients with sporadic pHPT are older, more often to be symptomatic and to have decreased bone density, and to have higher serum calcium levels. In contrast, hypercalcemia in FHH is lifelong, and patients are identified as early as the second or third decade of life. Importantly, while patients with pHPT have normal/ elevated uCa levels despite the effects of PTH on the kidney because of the increased filtered load of calcium, patients with FHH patients typically have a hypocalciuric hypercalcemia, because the mutated CASR is also present Table 3
in the kidney, although patients with FHH may also have normal 24-hour uCa levels.10,12 FHH is not the only etiology of low 24-hour uCa levels and other factors must also be considered. Urinary calcium excretion has been shown to vary with sex, age, race, and with the use of certain medications, such as thiazide diuretics. One group examined a cohort of African American patients with pHPT and found that this group had an unexpectedly high prevalence of low uCa compared with non-African American controls.13 In our study, there was not a statistically significant difference in the number of African American, Caucasian, or Hispanic patients between the 3 study groups, but our sample sizes are small. However, we did identify the use of thiazide diuretics at the time of 24-hour uCa collection as a complicating factor in accurately measuring urinary calcium excretion. With cessation of thiazide and repeat 24-hour uCa, 4 of the 9 (44%) patients had normalization. It is unclear why the other 5 patients continued to have low uCa excretion, but this finding highlights the importance of discontinuing medications that are known to affect renal calcium excretion prior to the time of 24-hour uCa measurement. We identified statistically significant lower median serum creatinine levels and higher GFR values in the patients with repeat normal 24-hour uCa measurements. This suggests that this group of patients had better overall renal function compared with the other groups, and on
Operative findings for patients with 24-hour uCa undergoing parathyroidectomy
Single gland disease Multiple gland disease uCa 5 urine calcium.
One low 24-hour uCa (n 5 28)
Multiple low 24-hour uCa (n 5 9)
One low 24-hr uCa with normalization (n 5 17)
22 (79%) 6 (21%)
7 (78%) 2 (22%)
14 (82%) 3 (18%)
K. O’Connell et al.
Low urine calcium in primary hyperparathyroidism
127
Figure 1 Algorithm for evaluation of patients with initial low 24-hour uCa measurement. *CCCR 5 calcium-creatinine clearance ratio; **Clinical discretion of providers, with respect to the need for further evaluation.
repeat uCa measurement, were no longer concerning for FHH. Therefore, impaired baseline renal function may impact the usefulness of 24-hour uCa in ruling out FHH as a cause of hypercalcemia. Limitations to this study include the retrospective nature, single institution experience and small sample size. Twenty-four hour urine creatinine levels were also not collected in all patients; therefore, the CCCR was unable to be calculated in all patients with low 24-hour uCa measurements. In addition, the cohort comprises only patients who underwent parathyroidectomy; information (number and characteristics) on patients with persistently low 24-hour uCa levels who were either not referred for surgical evaluation or were evaluated and did not undergo surgery is unknown. The percentage of patients with hypocalciuria was only 5% of the total number of patients with pHPT; a larger cohort with preoperative hypocalciuria may identify other significant differences between these groups. However, in our patient cohort, all patients had curative parathyroidectomy with normalization of serum calcium levels, evidence that the diagnosis of pHPT was correct and that further testing with genetic evaluation is likely not necessary. Although previously recognized, there is little written in the literature on patients with pHPT and low uCa levels. This study identifies a small but discrete subgroup of patients with pHPT who were found to have hypocalciuria on preoperative workup. The optimal screening strategy for FHH has yet to be defined and currently is a compilation of clinical and biochemical findings. Although genetic testing
is the gold standard for diagnosing FHH, it is expensive and time consuming. The results of this study propose that patients with low 24-hour uCa, previous eucalcemia, and no family history to suggest FHH, including normal serum calcium levels in family members, may not need further evaluation before parathyroidectomy (Fig. 1). Elimination of genetic testing in patients who meet these criteria will decrease costs and expedite surgical cure in this group of patients with pHPT. In the future, less expensive genetic testing may alter this algorithm.
References 1. Wermers RA, Khosla S, Atkinson EJ, et al. Incidence of primary hyperparathyroidism in Rochester, Minnesota, 1993-2001: an update on the changing epidemiology of the disease. J Bone Miner Res 2006; 21:171–7. 2. Bilezikian JP, Silverberg SJ. Clinical practice. Asymptomatic primary hyperparathyroidism. N Eng J Med 2004;350:1746–51. 3. Varghese J, Rich T, Jimenez C. Benign familial hypocalciuric hypercalcemia. Endocr Pract 2011;17:13–7. 4. Nesbit MA, Hannan FM, Howles SA, et al. Mutations affecting Gprotein subunit a11 in hypercalcemia and hypocalcemia. N Engl J Med 2013;368:2476–86. 5. Nesbit MA, Hannan FM, Howles SA, et al. Mutations in AP2S1 causes familial hypocalciuric hypercalcemia type 3. Nat Genet 2013;45:93–7. 6. Heath DA. Familial hypocalciuric hypercalcemia. Rev Endocr Metab Disord 2000;1:291–6. 7. Bilezikian JP, Potts Jr JT, Fuleihan Gel-H, et al. Summary statement from a workshop on asymptomatic primary hyperparathyroidism: a perspective for the 21st century. J Clin Endocrinol Metab 2002;87: 5353–61.
128
The American Journal of Surgery, Vol 210, No 1, July 2015
8. Bilezikian JP, Khan AA, Potts Jr JT, et al. Guidelines for the management of asymptomatic primary hyperparathyroidism: summary statement from the third international workshop. J Clin Endocrinol Metab 2009;94:335–9. 9. Wade TJ, Yen TW, Amin AL, et al. Focused parathyroidectomy with intraoperative parathyroid hormone monitoring in patients with lithiumassociated primary hyperparathyroidism. Surgery 2013;153:718–22. 10. Shinall MC, Dahir KM, Broome JT. Differentiating familial hypocalciuric hypercalcemia from primary hyperparathyroidism. Endocr Pract 2013;19:697–702.
11. Marx SJ, Stock JL, Attie MF, et al. Familial hypocalciuric hypercalcemia: recognition among patients referred after unsuccessful parathyroid exploration. Ann Intern Med 1980;92:351–6. 12. Christensen SE, Nissen PH, Vestergaard P, et al. Discriminative power of three indices of renal calcium excretion for the distinction between familial hypocalciuric hypercalcaemia and primary hyperparathyroidism: a follow-up study on methods. Clin Endocrinol 2008;69:713–20. 13. Taha W, Singh N, Flack JM, et al. Low urine calcium excretion in African Americans with primary hyperparathyroidism. Endocr Pract 2011;17:867–72.
©2015 Elsevier
Clinical Science
Low 24-hour urine calcium levels in patients with sporadic primary hyperparathyroidism: is further evaluation warranted prior to parathyroidectomy? Kathleen O’Connell, M.D.a, Tina W. Yen, M.D.a, Joseph Shaker, M.D.b, Stuart D. Wilson, M.D.a, Douglas B. Evans, M.D.a, Tracy S. Wang, M.D., M.P.H.a,* a
Department of Surgery, Division of Surgical Oncology, bDepartment of Medicine, Division of Clinical Endocrinology, Metabolism, and Nutrition, Medical College of Wisconsin, 9200 W. Wisconsin Avenue, Milwaukee, WI 53226, USA
KEYWORDS: 24-Hour urine calcium; Familial hypocalciuric hypercalcemia; Primary hyperparathyroidism
Abstract BACKGROUND: Low 24-hour urine calcium (uCa) levels in patients with primary hyperparathyroidism (pHPT) raise concern for familial hypocalciuric hypercalcemia. This study evaluated patients with a low 24-hour uCa level for potential differences that may guide the extent of preoperative evaluation needed. METHODS: A retrospective review was conducted of 1,139 sporadic pHPT patients who underwent parathyroidectomy between December 1999 and May 2011. RESULTS: Of the 54 (5%) patients with greater than or equal to one low 24-hour uCa (,100 mg), 28 (52%) patients had only one low level, 9 (17%) had multiple low levels, and 17 (31%) had a repeat 24-hour uCa greater than 100. In the latter group, 4 of the 9 (53%) patients were on a thiazide and had normalization after cessation. Among the groups, differences existed only in serum creatinine (P 5 .0011) and glomerular filtration rate (P 5 .0007). CONCLUSION: This study suggests that sporadic pHPT patients with low 24-hour uCa levels may not require further evaluation with genetic testing for familial hypocalciuric hypercalcemia, especially if previous eucalcemia is documented. Ó 2015 Elsevier Inc. All rights reserved.
Primary hyperparathyroidism (pHPT) is the most common cause of hypercalcemia, with an estimated incidence of 22 cases per 100,000 person years in the United States.1 pHPT is characterized by increased serum calcium levels The authors declare no conflicts of interest. * Corresponding author. Tel.: 11-414-805-5755; fax: 11-414-805-5771. E-mail address: [email protected] Manuscript received February 1, 2014; revised manuscript April 29, 2014 0002-9610/$ - see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjsurg.2014.09.030
secondary to inappropriately elevated levels of parathyroid hormone (PTH); definitive therapy with curative parathyroidectomy may prevent end-organ damage, such as nephrolithiasis and osteodystrophy. The majority of patients with pHPT have sporadic disease; less commonly, pHPT occurs as a component of a familial syndrome, such as multiple endocrine neoplasia (MEN) types 1 and 2.2 Familial hypocalciuric hypercalcemia (FHH) accounts for approximately 2% of all cases of patients with hypercalcemia.3 FHH1 is an autosomal dominant disorder caused
124
The American Journal of Surgery, Vol 210, No 1, July 2015
by inactivating mutations in the calcium sensing receptor (CASR) gene, of which over 200 mutations have been identified to date.3 FHH1 accounts for about two third of FHH cases. FHH2 is caused by mutations of G protein alpha 114 and FHH3 is caused by mutations of Adapter Protein 2 Sigma 1.5 The biochemical profile of FHH is very similar to pHPT and is often difficult to differentiate. FHH is characterized by mild hypercalcemia, normal or elevated PTH levels, relatively low renal calcium excretion, and a family history of hypercalcemia.6 FHH is a benign condition that does not require treatment, whereas surgical intervention to reduce the associated long-term complications is usually advised in pHPT. Consequently, it is important to accurately differentiate whether hypercalcemia is secondary to pHPT versus FHH prior to planned parathyroidectomy. The most common measurement used to distinguish between patients with pHPT and FHH is the calcium-tocreatinine clearance ratio (CCCR), which is calculated as (24-hour urine calcium/plasma calcium)/(24-hour urine creatinine/plasma creatinine). The 2009 consensus panel on Guidelines in the Management of Asymptomatic pHPT set the CCCR threshold value at less than .01 for the diagnosis of FHH and greater than .02 for the diagnosis of pHPT.7 Although these cut-off values are not perfect and there is overlap between the diseases, the workup of patients with newly diagnosed hypercalcemia should include a 24-hour urine collection for evaluation of calcium excretion.8 Not all patients with low 24-hour urine calcium (uCa) levels have FHH; other etiologies for low 24-hour uCa levels include increasing age, use of thiazide diuretics, and chronic kidney disease. In the evaluation of patients with suspected pHPT and a low 24-hour uCa, evaluation of these causative factors (ie, discontinuation of thiazide diuretic), if possible, is recommended, followed by a repeat measurement of 24-hour uCa and creatinine levels. In patients in whom there is no other known etiology and for those with persistently low 24-hour uCa levels, genetic testing for CASR mutations is recommended, which can be time consuming and costly, particularly because FHH is rare and most patients will indeed have pHPT. Therefore, the objective of this study was to identify potential differences in patients with an initial low 24-hour uCa level that may guide the extent of evaluation recommended prior to parathyroidectomy.
secondary, or tertiary hyperparathyroidism, and those with no documentation of 24-hour uCa levels were excluded. The resulting cohort of patients was subsequently divided into 3 groups: patients with only one documented 24-hour uCa, which was low; patients with multiple low 24-hour uCa levels, and patients with a repeat 24-hour uCa greater than 100. Data collected included patient demographics, signs/symptoms of hyperparathyroidism, use of thiazide diuretic, laboratory values closest to surgery (serum calcium and creatinine, ionized calcium, PTH, vitamin D, glomerular filtration rate [GFR], uCa), postoperative laboratory values (serum calcium and PTH), number of parathyroid glands removed, and final histopathology results. Any family history of hypercalcemia or previous parathyroid disease is routinely obtained; for patients with suspicion of familial parathyroid disease, attempts are made to obtain serum calcium levels ideally from parents or other first-degree relatives. At our institution, initial evaluation of patients with hypercalcemia and suspected pHPT includes measurement of serum calcium, creatinine and glomerular filtration rate, intact PTH, and 25-hydroxy vitamin D levels, as well as a 24-hour uCa and creatinine collection. Prior to 2009, 24-hour uCa levels were obtained at the discretion of the referring provider or surgeon; since 2009, collection of 24-hour uCa and urine creatinine levels have been a routine part of the biochemical evaluation of patients with pHPT. All patients undergo preoperative localization studies including a high-quality cervical ultrasound and Sestamibi/technetium-99m scintigraphy (performed with single photon emission computed tomography after March 2009); an angiographic (‘‘four-dimensional’’) computed tomography scan is performed if the results of the former 2 studies are discordant or negative. Intraoperative PTH monitoring is routinely performed and our institutional protocol has previously been described;9 briefly, criterion for cure includes a greater than or equal to 50% decrease from the baseline or excisional PTH, whichever is higher, and returns to normal range, at 10 minutes postexcision of the abnormal parathyroid gland(s). Biochemical cure after parathyroidectomy is defined as serum calcium of less than 10.2 mg/dL at 6 months following parathyroidectomy. Patients with normocalcemia for at least 6 months after surgery who then had a serum calcium greater than or equal to 10.2 mg/dL were considered to have recurrent disease. Simple descriptive statistics were performed; a P value of less than or equal to .05 was considered statistically significant.
Methods This study was an Institutional Review Board–approved retrospective review of patients with pHPT who underwent parathyroidectomy by a single group of endocrine surgeons from December 1999 to December 2011. A prospectively collected parathyroid database was reviewed for patients with biochemically confirmed, sporadic pHPT who had at least one low 24-hour uCa measurement, defined as greater than 100 mg/24 hours. Patients with familial, recurrent, persistent,
Results Of the 1,035 patients in the database who underwent parathyroidectomy for sporadic pHPT, 735 (70%) patients had 24-hour uCa testing performed. Of these, 54 (7%) patients had at least one preoperative low 24-hour uCa measurement and comprise the study cohort. No patient
K. O’Connell et al.
Low urine calcium in primary hyperparathyroidism
had a family history suggestive of an inherited endocrinopathy. The median age of the patients was 65 years (range 33 to 85), and 44 (81%) were female. There were no statistically significant differences in patient demographics among the 3 groups (Table 1). One patient with repeat low levels had negative genetic testing for a CASR mutation. When comparing the groups by biochemical markers, patients with a 24-hour uCa that normalized had a lower median serum creatinine (.8 vs 1.2 vs 1.0; P 5 .0011) and higher GFR (86 vs 48 vs 62; P 5.0007) than those with a single low measurement or multiple low measurements (Table 2). Furthermore, when comparing 24-hour uCa levels obtained closest to the date of surgery, levels were highest in patients with a normalized 24-hour uCa level (119 mg/ 24 hours), compared with 72 mg/24 hours in the group with a single measurement and 38 mg/24 hours in the group with multiple low levels (P 5.0001). There was no difference in any other biochemical markers among the 3 groups. The use of a thiazide diuretic at the time of initial 24hour urine collection was documented in 12 patients. In the group with one low 24-hour uCa measurement, 2 patients (7%) were on a thiazide at the time of initial collection. In the group with multiple low 24-hour uCa measurements, 1 patient (11%) was on a thiazide at the time of collection. Of the patients with an initial low 24-hour uCa who subsequently normalized, 9 (53%) patients were on a thiazide at the time of initial collection. On repeat testing with cessation of the thiazide, 4 patients had uCa measurements within normal limits. There were no differences in pathologic results among the 3 groups (Table 3). Overall, 43 (80%) patients had single gland disease and 11 (20%) had multigland hyperplasia. This distribution remained consistent across the groups. The median follow-up for the entire cohort was 16.4 months; median follow-up was 13 months (range 1 to 80 months) for patients with one low 24-hour uCa measurement, 13 months (range 2 to 77 months) for patients with multiple low 24-hour uCa measurements, and 25 months (1 to 83 months) for patients with one low 24-hour uCa with subsequent normalization. Median serum calcium levels at the time of last visit were 9.5 (8.1 to 10.2), 9.8 (8.5 to 10.6), and 9.5 mg/dL (8.6 to 10.6) for the 3 groups, respectively. Only one patient, in
Table 1
125
the group with multiple low 24-hour uCa measurements, had a known recurrence.
Comments The majority of patients with pHPT will have sporadic disease, with a small subset of patients having an inherited syndrome, such as MEN1, MEN2, isolated familial hyperparathyroidism, or hyperparathyroidism jaw-tumor syndrome. In patients with elevated serum calcium levels undergoing evaluation for pHPT, FHH, for which parathyroidectomy is not indicated, should be excluded with measurement of 24-hour uCa levels. When the 24-hour uCa measurement is low (,100 mg/24 hours), concern is raised for FHH, and may require repeat urine biochemical testing and/or genetic evaluation. The results of this study suggest that, given the rarity of FHH, further evaluation with genetic testing may not be necessary prior to parathyroidectomy, particularly if eucalcemia is documented in the patient previously and/or in family members. The most common type of FHH (FHH1) results from an autosomal dominant transmission of a mutation in the CASR, a G protein-coupled receptor located on the surface of parathyroid chief cells, thyroidal C-cells, and renal tubular epithelial cells. Normally, this receptor binds to calcium, resulting in reduced PTH production and renal calcium reabsorption. With an inactivating mutation in one copy of the CASR gene, the CASR is relatively insensitive to serum calcium levels and requires higher calcium levels to activate the normal feedback mechanisms.10 The paradoxical finding of hypocalciuria in the setting of mild to moderate hypercalcemia, normal to mildly elevated PTH, and family history of hypercalcemia are characteristic findings of FHH because of loss-of-function of the CASR in the renal tubule. Although FHH is a benign condition that does not require treatment, it is important to distinguish FHH from pHPT, as surgical intervention in FHH patients is ineffective at reducing serum calcium levels. As such, FHH is sometimes recognized in patients with persistent hypercalcemia after parathyroidectomy. In a landmark study of a cohort of 67 patients with persistent hypercalcemia after failed parathyroidectomy,
Demographics of the 54 patients with low 24-hour uCa levels
Median age (years) Female Caucasian African American Hispanic Median BMI (kg/m2)
One low 24-hour uCa (n 5 28)
Multiple low 24-hour uCa (n 5 9)
One low 24-hour uCa with normalization (n 5 17)
67 22 25 2 1 27
65 7 7 1 1 29
60 15 9 7 1 31
(33–84) (79%) (89%) (7%) (4%) (16–46)
Value ranges are listed in parentheses. BMI 5 body mass index; uCa 5 urine calcium.
(58–85) (78%) (78%) (11%) (11%) (21–41)
(37–83) (88%) (53%) (41%) (6%) (21–42)
126
The American Journal of Surgery, Vol 210, No 1, July 2015
Table 2
Biochemical markers for patients with low 24-hour uCa levels One low 24-hour uCa (n 5 28)
Median preop calcium (mg/dL) Median maximum preop calcium (mg/dL) Ionized calcium (mmol/L) Preop PTH (pg/mL) Maximum PTH (pg/mL) Vitamin D (ng/mL) Serum creatinine (mg/dL) GFR (mL/min/1.73 sqm) Urine calcium closest to surgery CCCR (lowest) CCCR (highest)
10.9 (9.9–11.8) 11.4 (10.5–12.7) 1.44 117 145 28 1.2 48 72 .00745 .00745
(1.32–6.9) (27–463) (53–484) (7–54) (.8–1.9) (34–75) (3–95) (.00285–.0534) (.00285–.0534)
Multiple low 24-hour uCa (n 5 9)
One low 24-hour uCa with normalization (n 5 17)
10.6 (8.7–11.5) 11.2 (10.6–11.6)
10.9 (9.4–11.8) 11.3 (9.9–12.5)
1.43 94 177 35 1.0 62 38 .00551 .00706
(1.25–6.0) (67–428) (85–428) (19–42) (.8–1.7) (30–89) (15–72) (.00317–.00704) (.00315–.01097)
1.44 94 137 19 .8 86 119 .00542 .01065
P value .59 .30
(1.28–5.45) (7–311) (77–431) (7–50) (.6–1.8) (29–115) (42–471) (.0005–.0103) (.00229–.11932)
.82 .51 .74 .19 .0011 .0007 .0001 .26 .44
Value ranges are listed in parentheses. CCCR 5 calcium to creatinine clearance ratio; GFR 5 glomerular filtration rate; PTH 5 parathyroid hormone; sqm 5 square meters; uCa 5 urine calcium.
6 (9%) patients were subsequently identified as members of kindreds with FHH.11 These patients were found to have lower CCCR, as compared with the remaining patients with persistent hypercalcemia after parathyroidectomy; thus, the authors recommended assessment of uCa excretion in all hypercalcemic patients. A subsequent study has shown CCCR to have a sensitivity of 80% and specificity of 88% when the cut-off point was less than .01 for diagnosing patients with FHH, and a CCCR of less than .02 identified 53 of the 54 patients with FHH. However, the use of this test alone would potentially misdiagnose a significant proportion of patients with FHH, and thus the authors recommended use of the CCCR for initial FHH screening, followed by CASR gene analysis for patients with CCCR levels suggestive of FHH.12 Differentiating between pHPT and FHH is often not straightforward, as there is considerable overlap in both the clinical and biochemical features of these disease processes. In general, patients with sporadic pHPT are older, more often to be symptomatic and to have decreased bone density, and to have higher serum calcium levels. In contrast, hypercalcemia in FHH is lifelong, and patients are identified as early as the second or third decade of life. Importantly, while patients with pHPT have normal/ elevated uCa levels despite the effects of PTH on the kidney because of the increased filtered load of calcium, patients with FHH patients typically have a hypocalciuric hypercalcemia, because the mutated CASR is also present Table 3
in the kidney, although patients with FHH may also have normal 24-hour uCa levels.10,12 FHH is not the only etiology of low 24-hour uCa levels and other factors must also be considered. Urinary calcium excretion has been shown to vary with sex, age, race, and with the use of certain medications, such as thiazide diuretics. One group examined a cohort of African American patients with pHPT and found that this group had an unexpectedly high prevalence of low uCa compared with non-African American controls.13 In our study, there was not a statistically significant difference in the number of African American, Caucasian, or Hispanic patients between the 3 study groups, but our sample sizes are small. However, we did identify the use of thiazide diuretics at the time of 24-hour uCa collection as a complicating factor in accurately measuring urinary calcium excretion. With cessation of thiazide and repeat 24-hour uCa, 4 of the 9 (44%) patients had normalization. It is unclear why the other 5 patients continued to have low uCa excretion, but this finding highlights the importance of discontinuing medications that are known to affect renal calcium excretion prior to the time of 24-hour uCa measurement. We identified statistically significant lower median serum creatinine levels and higher GFR values in the patients with repeat normal 24-hour uCa measurements. This suggests that this group of patients had better overall renal function compared with the other groups, and on
Operative findings for patients with 24-hour uCa undergoing parathyroidectomy
Single gland disease Multiple gland disease uCa 5 urine calcium.
One low 24-hour uCa (n 5 28)
Multiple low 24-hour uCa (n 5 9)
One low 24-hr uCa with normalization (n 5 17)
22 (79%) 6 (21%)
7 (78%) 2 (22%)
14 (82%) 3 (18%)
K. O’Connell et al.
Low urine calcium in primary hyperparathyroidism
127
Figure 1 Algorithm for evaluation of patients with initial low 24-hour uCa measurement. *CCCR 5 calcium-creatinine clearance ratio; **Clinical discretion of providers, with respect to the need for further evaluation.
repeat uCa measurement, were no longer concerning for FHH. Therefore, impaired baseline renal function may impact the usefulness of 24-hour uCa in ruling out FHH as a cause of hypercalcemia. Limitations to this study include the retrospective nature, single institution experience and small sample size. Twenty-four hour urine creatinine levels were also not collected in all patients; therefore, the CCCR was unable to be calculated in all patients with low 24-hour uCa measurements. In addition, the cohort comprises only patients who underwent parathyroidectomy; information (number and characteristics) on patients with persistently low 24-hour uCa levels who were either not referred for surgical evaluation or were evaluated and did not undergo surgery is unknown. The percentage of patients with hypocalciuria was only 5% of the total number of patients with pHPT; a larger cohort with preoperative hypocalciuria may identify other significant differences between these groups. However, in our patient cohort, all patients had curative parathyroidectomy with normalization of serum calcium levels, evidence that the diagnosis of pHPT was correct and that further testing with genetic evaluation is likely not necessary. Although previously recognized, there is little written in the literature on patients with pHPT and low uCa levels. This study identifies a small but discrete subgroup of patients with pHPT who were found to have hypocalciuria on preoperative workup. The optimal screening strategy for FHH has yet to be defined and currently is a compilation of clinical and biochemical findings. Although genetic testing
is the gold standard for diagnosing FHH, it is expensive and time consuming. The results of this study propose that patients with low 24-hour uCa, previous eucalcemia, and no family history to suggest FHH, including normal serum calcium levels in family members, may not need further evaluation before parathyroidectomy (Fig. 1). Elimination of genetic testing in patients who meet these criteria will decrease costs and expedite surgical cure in this group of patients with pHPT. In the future, less expensive genetic testing may alter this algorithm.
References 1. Wermers RA, Khosla S, Atkinson EJ, et al. Incidence of primary hyperparathyroidism in Rochester, Minnesota, 1993-2001: an update on the changing epidemiology of the disease. J Bone Miner Res 2006; 21:171–7. 2. Bilezikian JP, Silverberg SJ. Clinical practice. Asymptomatic primary hyperparathyroidism. N Eng J Med 2004;350:1746–51. 3. Varghese J, Rich T, Jimenez C. Benign familial hypocalciuric hypercalcemia. Endocr Pract 2011;17:13–7. 4. Nesbit MA, Hannan FM, Howles SA, et al. Mutations affecting Gprotein subunit a11 in hypercalcemia and hypocalcemia. N Engl J Med 2013;368:2476–86. 5. Nesbit MA, Hannan FM, Howles SA, et al. Mutations in AP2S1 causes familial hypocalciuric hypercalcemia type 3. Nat Genet 2013;45:93–7. 6. Heath DA. Familial hypocalciuric hypercalcemia. Rev Endocr Metab Disord 2000;1:291–6. 7. Bilezikian JP, Potts Jr JT, Fuleihan Gel-H, et al. Summary statement from a workshop on asymptomatic primary hyperparathyroidism: a perspective for the 21st century. J Clin Endocrinol Metab 2002;87: 5353–61.
128
The American Journal of Surgery, Vol 210, No 1, July 2015
8. Bilezikian JP, Khan AA, Potts Jr JT, et al. Guidelines for the management of asymptomatic primary hyperparathyroidism: summary statement from the third international workshop. J Clin Endocrinol Metab 2009;94:335–9. 9. Wade TJ, Yen TW, Amin AL, et al. Focused parathyroidectomy with intraoperative parathyroid hormone monitoring in patients with lithiumassociated primary hyperparathyroidism. Surgery 2013;153:718–22. 10. Shinall MC, Dahir KM, Broome JT. Differentiating familial hypocalciuric hypercalcemia from primary hyperparathyroidism. Endocr Pract 2013;19:697–702.
11. Marx SJ, Stock JL, Attie MF, et al. Familial hypocalciuric hypercalcemia: recognition among patients referred after unsuccessful parathyroid exploration. Ann Intern Med 1980;92:351–6. 12. Christensen SE, Nissen PH, Vestergaard P, et al. Discriminative power of three indices of renal calcium excretion for the distinction between familial hypocalciuric hypercalcaemia and primary hyperparathyroidism: a follow-up study on methods. Clin Endocrinol 2008;69:713–20. 13. Taha W, Singh N, Flack JM, et al. Low urine calcium excretion in African Americans with primary hyperparathyroidism. Endocr Pract 2011;17:867–72.
©2015 Elsevier
