ENDOCRINE PRACTICE Rapid Electronic Article in Press Rapid Electronic Articles in Press are preprinted manuscripts that have been reviewed and accepted for publication, but have yet to be edited, typeset and finalized. This version of the manuscript will be replaced with the final, published version after it has been published in the print edition of the journal. The final, published version may differ from this proof. Original Article EP-2018-0350 CALCIUM CREATININE CLEARANCE RATIO IS NOT HELPFUL IN DIFFERENTIATING PRIMARY HYPERPARATHYROIDISM FROM FAMILIAL HERPERCALCEMIC HYPOCALCIURIA: A STUDY OF 1000 PATIENTS Edwina C Moore, MD, Eren Berber, MD, Judy Jin, MD, Vikram Krishnamurthy, MD, Joyce Shin, MD, Allan Siperstein, MD Running Title: Validity of CCCR in Primary Hyperparathyroidism From: Department of Endocrine Surgery, Cleveland Clinic, F/20 9500 Euclid Avenue, Cleveland, 44195, OHIO, United States Corresponding author: Edwina C Moore, MD Cleveland Clinic - Endocrine Surgery F20/9500 Euclid Avenue Cleveland Ohio 44195 Email: edwina.bolshinsky@gmail.com / mooree@ccf.org Conflicts: There are no conflicts of interest
Abstract Objective: With increasing recognition of more subtle presentations of primary hyperparathyroidism (phpt), laboratory values are frequently seen in a range that would be expected for patients who have familial hypercalcemic hypocalciuria (FHH). Calcium creatinine clearance ratio (CCCR) has been advocated as a diagnostic tool to differentiate between these two disorders. However, it is limited by an indeterminate range (0.01-0.02). The aim of this study is to assess the relevance of CCCR in a modern series of patients with surgically managed phpt. Methods: We performed a retrospective cohort study of 1000 patients who underwent parathyroid surgery for phpt over eleven years. CCCR was evaluated by degree of biochemical derangement, single versus multiple gland disease and interfering medications. Results: Patient demographics and resected histopathology were typical for a current series of patients with phpt. In retrospect, none of the patients were suspected to have FHH post operatively. CCCR was less than 0.01 for 19.0%, between 0.01-0.02 for 43.7% and greater than 0.02 in 37.3%. Distribution of CCCR for patients free from interfering medications and different histological subtypes were the same. One third of the cohort had mild calcium elevations, more typical for FHH. Of these, almost two thirds had a CCCR in a range suspect for FHH (<0.02). Conclusion: To our knowledge this is the largest series to evaluate the validity of CCCR for patients with surgically confirmed ppht. The utility of CCCR in screening for FHH is limited, as 63% of modern patients with confirmed phpt have low values.
Introduction phpt is one of the most common endocrine disorders, with a disease prevalence of one case per 1000 population (1), but potentially higher as the diagnostic criteria broadens (2). Patients have hyper secretion of parathyroid hormone (PTH), usually from a single adenoma, leading to an elevation of serum calcium. Unfortunately, the clinical and biochemical profile are subject to significant individual variation, which can complicate diagnosis. One of the most critical steps for the endocrine surgeon is to exclude the potential for misdiagnosis with familial hypocalciuric hypercalcemia (FHH). First described in 1966 (3) and named by Foley in 1972 (4), FHH is a rare, autosomal dominant disorder usually caused by a heterozygous inactivating mutation in the calcium sensing receptor gene (CaSR) (5-6). Several additional genetic mutations have subsequently been associated with this disorder (7-9). The physiological defect is calcium hyposensitivity with associated hypercalcemia. It is important to identify these patients among hypercalcemic patients considering surgery, so as to avoid unnecessary surgery. Unlike patients with phpt, patients with FHH are asymptomatic with a normal phenotype and limited potential for sequelae (10). As phpt and FHH differ in their renal processing of calcium several markers of renal calcium excretion have been proposed to distinguish between the two disorders: 24 hour urine calcium
excretion, fasting urinary calcium excretion, calcium creatinine ratio and CCCR (5,11-19). Classically, urinary calcium is low in FHH, but despite its name this is not universally the case (20-21). In 2008, Christensen et al proposed a two-step diagnostic algorithm to exclude FHH from patients with suspected phpt. This includes calculating CCCR followed by genetic testing for all patients with CCCR<0.02. This two-step algorithm is reportedly associated with a sensitivity of 98% (17). The CCCR diagnostic cutoffs (<0.01 is FHH and >0.02 is phpt) dates from the first published series in 1981 (13). To our knowledge, no large scale study has confirmed the efficacy of CCCR in a modern series of patients surgically treated (and cured) for phpt, using the cutoffs previously described. Our primary aim was to evaluate the validity of CCCR, as a diagnostic test for FHH, amongst patients with suspected phpt. Secondary aims were to assess its utility for particular subgroups including patients with mild biochemical derangement, patients free from medications known to interfere with urine calcium excretion and histological variants of hyperparathyroidism. Methods Patients Over an eleven year period (Jan 1 2007- Jan 12 2018), 1847 patients were managed in a tertiary referral center for parathyroid disease. Preoperative workup and surgery was performed by one
of five staff surgeons. Indications for surgery followed the guidelines of the American Association of Endocrine Surgeons (22). Specific inclusion criteria were adult patients with biochemically proven primary hyperparathyroidism. Exclusion criteria were re-operative surgery. The study was approved by the Cleveland Clinic Institutional Review Board. Data collection The institution database, which is prospectively maintained, was accessed for patients surgically treated for phpt. Basic demographics (gender, age, primary complaint, medication history), preoperative biochemical markers (serum calcium, serum creatinine, serum phosphorus, PTH, Vitamin D, 24hour urine calcium and 24hour urine creatinine), indications for surgery, histology and postoperative biochemistry (serum calcium and PTH) were extracted. CCCR was calculated as [24hour urine calcium (mg/24hrs)/total serum calcium (mg/dl)]/[24 hour urine creatinine (mg/24hrs)/total serum creatinine (mg/dl)]. Statistical analysis The data was retrospectively analyzed. Metric variables are described by median and interquartile range (IQR) and categorical variables by absolute and relative frequencies. Regression analyses were completed using Excel and SPSS. Results We identified 1000 patients out of 1847, who had primary hyperparathyroidism and a complete set of biochemistry to calculate CCCR, over 11 years. All of these patients underwent parathyroid surgery and none, in retrospect, were suspected of having FHH.
The majority of patients were female (79.7%) and the mean age was 60.7 years. Twenty-six patients had a history of prior neck radiation. Almost one in four patients (23.7%) used medications known to affect calcium excretion (thiazide 181, cinacalcet 6, bisphosphonate 62, lithium 10, biotin 6) at the time of surgery. The mean preoperative serum calcium was 10.8 mg/dl and PTH was 111 pg/ml. 1756 parathyroid glands were excised from 1000 patients. Of these, 624 patients had single adenomas (624 adenomas, 62.4%), 186 patients had double adenomas (372 adenomas, 18.6%) and 190 patients had four gland hyperplasia (760 parathyroid glands, 19.0%). The distribution of CCCR for the total study population was 19.0% for CCCR<0.01, 43.7% for CCCR between 0.01-0.02 and 37.3% for CCCR >0.02. A similar distribution was seen for each histological subtypes. (Inset Table 1 and Figure 1) When patients using medications known to interfere with calcium excretion were excluded (n=237) from the analysis, the distribution of CCCR was unchanged from the total population. More than two thirds of this subset of patients with histologically confirmed and biochemically cured primary hyperparathyroidism were still found to have a preoperative CCCR less than 0.02. (CCCR<0.01 18.0%, CCCR 0.01-0.02 43.4%, CCCR >0.02 38.6%). (Inset Figure2) There were 303 patients with mild biochemical derangement (defined as serum Ca <10.8mg/dL and PTH <100pg/mL). Distribution of CCCR for this subgroup showed that 18% (n=137) had a
value less than 0.01 and 68.0% (n=206) had a value less than 0.02. (Inset Figure 3). The distribution of CCCR was unaffected gender. On regression analysis, there was no correlation between PTH and Ca with CCCR. Figure 4 compares the regression analysis of Ca and CCCR for the total population and the subset with four gland hyperplasia. The proportion of patients with hyperplasia who had a low CCCR and a borderline serum calcium, was not statistically different to the proportion of patients in the entire population with a similar biochemical profile (19% v/v 31%). population. (Inset Figure 4). Discussion The treatment for patients with phpt is surgery, with a long term cure a rate of 95-99% depending on criteria used (22-23). By contrast, for patients with FHH who often have a similar clinical and biochemical presentation, surgical cure is 0%. Therefore it is imperative to differentiate between these two disorders preoperatively. The purpose of this study was to evaluate the validity of a widely accepted diagnostic test, CCCR, for a modern population with surgically proven phpt. Our results show that CCCR is not a useful index to preoperatively differentiate phpt from FHH and its routine use should be abandoned. Almost two thirds (62.7%) of our patients had a CCCR<0.02 and therefore by current recommendations would have been recommended to have had preoperative genetic analysis for
CaSR mutations. Notwithstanding that this is neither a practical nor a cost effective strategy, it is also completely unnecessary. All of the patients in our population had abnormal parathyroid glands removed at operation and were biochemically cured post operatively. As the prevalence of FHH is extremely rare, reportedly between 1:10000 and 1:100000 (11, 19), it is possible that our population did not include a patient with FHH based on probability. In addition, with increasing recognition of FHH, such patients may have been screened out by primary care providers and never referred to us. However, in our subgroup of patients with mild biochemical derangement, who by definition are more likely to overlap with patients with FHH, two thirds still had a CCCR <0.02. Despite this all of them were found to have parathyroid disease and improved after surgery, thereby eliminating the possibility of misdiagnosis. Historical case series have identified that almost one third of patients with mild phpt have CCCR<0.01 and an additional one third have CCCR<0.02 (25). Our subgroup analysis is complementary. 21.8% of patients with mild phpt in our population had CCCR<0.01 and a further 46.2% had CCCR between 0.01 and 0.02. Glendenning et al rationalized that as the ratio is influenced by both the degree of urinary calcium excretion and the degree of elevation of serum calcium, CCCR is a sensitive but increasingly less specific for milder presentations (16). In our study, CCCR was non-discriminatory for all biochemical presentations, subtle and more severe. The distribution of pathology in our study was similar to other studies: 61.1 % single adenomas, 20% double adenomas and 18.9% four gland hyperplasia (22-24). We did not find a significant
difference in the distribution of CCCR for each histological subtype. For single adenomas, double adenomas and four gland hyperplasia alike, approximately one third of each cohort had a CCCR<0.01 and two thirds had a CCCR <0.02. To our knowledge, this has not previously been described. Genetic testing is presently viewed as the gold standard to diagnose FHH. Notwithstanding, false negatives are possible if there are novel CaSR mutations (26-27) and the vast number of possible mutations makes the diagnostic process even more complex (27-28). Long prior to the availability of genomic tests, Marx et al proposed the diagnosis of FHH depended on observation of characteristic features within a family (13). This remains true and should be upheld. Clinical red flags for FHH include: asymptomatic hypercalcemia beginning in early life, relative hypocalciuria and multiple affected family members in an autosomal dominant inheritance pattern. Fasting urinary calcium excretion (CaE) has also been advocated, in place of CCCR, to diagnose FHH (17, 30). It is advantageous for the ease of spot urinary testing without the need for acidification compared with 24 hour collection (16). CaE<30umol/L GFR indicates avid renal calcium conservation typical of FHH however overlap can occur with phpt in the setting of very high PTH, calcium poor diets and use of thiazide medication (30, 31). For these reasons we did not evaluate CaE. Kalderli et al reviewed factors influencing urinary calcium excretion in phpt (32). They reported that 1,25 (OH)2 D3 and Osteocalcin were the only factors to independently correlate
with CCCR. Riss et al also reviewed the impact of thiazides on biochemical parameters in phpt and concluded that withdrawal was necessary for accurate diagnosis prior to surgery (33). By contrast in our study, there was little difference in the distribution of CCCR between patients taking and not taking similar medications and therefore we conclude that it is not relevant. The major limitation of this study was that the timing in the acquisition of preoperative blood and urine tests was variable. This was related to surgical scheduling and high number of external referrals where repeating tests would be unnecessary. However we consistently selected laboratory results that were drawn together and as close as possible to the date of surgery. Secondly, we did not assess 24 hour urine volume (nor fasting status at the time of urine collection) which can theoretically affect urine flow rate and mineral concentration, however in a population with normal renal function this is not likely to be significant. Thirdly, we only included patients with complete sets of laboratory values required to calculate CCCR. Conclusion The role of CCCR to preoperatively differentiate between phpt and FHH is questionable. Its utility is not enhanced by any histological subtype, exclusion of confounding medications or even for mild biochemical presentations. CCCR is cumbersome to perform and interpretation has low yield. It should not be prioritized in the diagnostic algorithm for patients with phpt. References
1. Clarke BL. Epidemiology of primary hyperparathyroidism. Endocrine Practice. 2013; 18; 450-455 2. Press DM, Siperstein AE, Berber E, et al. The prevalence of undiagnosed and unrecognized primary hyperparathyroidism: A population-based analysis from the electronic medical record. Surgery. 2013: 154; 1232-1238 3. Jackson CE, Boonstra CE. Hereditary Hypercalcemia and parathyroid hyperplasia without definite hyperparathyroidism. Journal of Laboratory and Clinical Medicine. 1966; 668-883 4. Foley TP, Harrison HC, Arnaud CD and Harrison HE. Familial Benign Hypercalcemia. Journal of Pediatrics. 1972; 81: 1060-1067 5. Brown EM, Gamba G, Riccardi D et al. Cloning and characterization of an extracellular Ca2+ receptor from bovine parathyroid. Nature. 1993; 366; 575-580 6. Thakker RV. Disease associated with the extracellular calcium-sensing receptor. Cell Calcium 2004; 35:275-282 7. Ward BK, Magno AL, Blitvich BJ, et al. Novel mutations in the calcium-sensing receptor gene associated with biochemical and functional differences in familial hypocalciuric hypercalcaemia. Clin Endocrinol (Oxf). 2006; 64:580-587. 8. D'Souza-Li L, Yang B, Canaff L, et al. Identification and functional characterization of novel calcium-sensing receptor mutations in familial hypocalciuric hypercalcemia and autosomal dominant hypocalcemia. J Clin Endocrinol Metab. 2002;87:1309-1318 9. Lietman SA, Tenenbaum-Rakover Y, Jap TS, et al. A novel loss-of-function mutation, Gln459Arg, of the calcium-sensing receptor gene associated with apparent autosomal recessive inheritance of familial hypocalciuric hypercalcemia. J Clin Endocrinol Metab. 2009; 94:4372-4379
10. Pearce SHS, Trump D, Wooding C et al. Calcium-sensing receptor mutations in familial benign hypercalcemia and neonatal hyperparathyroidism. Journal of Clinical Investigation. 1995; 96:2683-2692 11. Shinall MC, McCrystal Dahir K, Broome JT. Differentiating familial hypocalciuric hypercalcemia from primary hyperparathyroidism. Endocrine Practice 2013;19:697-702 12. Tfelt-Hansen J, Brown EM. The calcium-sensing receptor in hereditary disorders of calcium homeostasis. Clinical Cases of Mineral Bone Metabolism. 2006; 3: 150-161 13. Glendenning, P. Diagnosis of primary hyperparathyroidism: controversies, practical issues and the need for Australian guidelines. Internal Medicine Journal. 2003; 33: 598-603 14. Christensen SE, Nissen PH, Vestergaard P, Heickendorff L, Brixen K, Mosekilde L. Discriminative power of three indices of renal calcium excretion for the distinction between familial hypocalciuric hypercalcemia and primary hyperparathyroidism: a follow up study on methods. Clinical Endocrinology. 2008; 69:713-720 15. Marx SJ, Attie MF, Levine MA et al. The hypocalciuric or benign variant of familial hypercalcemia: clinical and biochemical features in 15 kindreds. Medicine (Baltimore). 1981;69:397-412 16. Christensen SE, Nissen PH, Vestergaard P, et al. Skeletal consequences of familial hypocalciuric hypercalcaemia vs. primary hyperparathyroidism. Clinical Endocrinology. 2009; 71: 798-807 17. Christensen SE. PhD thesis: Calcium and bone metabolism in familial hypocalciuric hypercalcemia and primary hyperparathyroidism, Faculty of Health Sciences, University of Aarhus, Aarhus, Denmark 2009; 1-106
18. Carling T, Szabo E, Bai M, et al. Familial hypercalcemia and hypercalciuria caused by a novel mutation in the cytoplasmic tail of the calcium receptor. Journal of Clinical Endocrinology and Metabolism. 2000; 85:2042-2047 19. Pasieka JL, Andersen MA and Hanley DA. Familial benign hypercalcemia: hypercalciuria and hypocalcemia in affected members of a small kindred. Clinical Endocrinology. 1990; 33:429-433 20. Wilhelm SM, Wang TS, Ruan DT, et al. The American Association of Endocrine Surgeons Guidelines for Definitive Management of Primary Hyperparathyroidism. JAMA. 2016; 151:959 968 21. Udelsman R1, Donovan P, Shaw C. Cure predictability during parathyroidectomy. World Journal of Surgery. 2014; 38: 525-533 22. Abboud B, Sleilaty G, Helou E, et al. Existence and anatomic distribution of double parathyroid adenoma. Laryngoscope. 2005; 115: 1128-1131 23. Gunn IR, Wallace JR. Urine Calcium and Serum Ionized Calcium, Total Calcium and Parathyroid Hormone Concentrations in the Diagnosis of Primary Hyperparathyroidism and Familial Benign Hypercalcaemia. Annals of Clinical Biochemistry 1992; 29:52-58 24. Guarnieri V1, Canaff L, Yun FH, et al. Calcium-sensing receptor (CASR) mutations in hypercalcemic states: studies from a single endocrine clinic over three years. Journal of Clinical Endocrinology and Metabolism. 2010; 95: 1819-1829 25. Chou YH, Brown EM, Levi T, et al. The gene responsible for familial hypocalciuric hypercalcemia maps to chromosome 3q in four unrelated families. Nat Genet. 1992; 1:295-300
26. Nissen PH, Christensen SE, Heickendirff L, Brixen K, Mosekilde L. Molecular Genetic Analysis of the Calcium Sensing Receptor Gene in Patients Clinically Suspected to Have Familial Hypocalciuric Hypercalcemia: Phenotypic Variation and Mutation Spectrum in a Danish Population. The Journal Clinical Endocrinology & Metabolism. 2007; 92: 4373-4379 27. Hendy GN, Guarnieri V, Canaff L. Calcium-sensing receptor and associated diseases. Prog Mol Biol Transl Sci. 2009; 89: 31-95 28. Stuckey BG, Kent GN, Gutteridge DH, Pullan PT, Price RI, Bhagat C. Fasting calcium excretion and parathyroid horomone together distinguish familian hypocalciuric herpercalcemia from primary hyperparathyroidism. Endocrinology. 1987; 27: 525-533 29. Friedman PA, Bushinsky DA. Diuretic effects on calcium metabolism. Seminars in nephrology. 1999; 19: 551-556 30. Kaderli RM, Riss P, Geroldinger A, Selberherr A, Scheuba C, Niederle B. Factors influencing pre operative urinary calcium excretion in primary hyperparathyroidism. Clinical Endocrinology 2017; : 97 102 31. Riss P, KammerM, Selberherr A, et al. The influence of thiazide intake on calcium and parathyroid hormone levels in patients with primary hyperparathyroidism. Clinical Endocrinology. 2016; 85: 196-201
Table 1: Proportion of patients with phpt and their CCCR, stratified by histology Histology Number CCCR<0.01 % CCCR % CCCR>0.02 % of 0.01-0.02 patients All patients 1000 190 19.0 437 43.7 373 37.3 Single 624 95 15.2 269 43.1 260 41.7 Adenomas Double 186 40 21.5 80 43.0 66 35.5 Adenomas Four Gland 190 55 29.0 88 46.3 47 24.7 Hyperplasia
Serum Calcium (mg/dl) 15 Regression Analysis of Serum Calcium and CCCR for the Total Population Suspect for FHH Marginal for FHHUnlikely FHH 14 13 12 11 R² = 0.0088 10 9 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 Calcium Creatinine Clearance Ratio (CCCR)
Serum Calcium (mg/dl) 16 Regression Analysis of Serum Calcium and CCCR for Patients with Four Gland Hyperplasia Suspect for FHHMarginal for FHHUnlikely FHH 15 14 13 12 11 R² = 0.0009 10 9 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 Calcium Creatinine Clearance Ratio (CCCR)