Parathyroid hormone (PTH) maintains calcium homeostasis by binding to its receptors in bone and kidney. In pseudohypoparathyroidism (PHP), the major target organ (the kidney) resists the effects of PTH. Fuller Albright reported the first cases of PTH resistance in 1942.1 Classically, affected patients presented with hyperparathyroidism, hypocalcemia, and hyperphosphatemia and exhibited a number of distinctive physical characteristics including short stature, round facies, shortened fourth metacarpal bones, and subcutaneous ossifications—a constellation of findings commonly referred to as Albright's hereditary osteodystrophy (AHO). Subsequent studies have led to an appreciation of the variable presentations of PHP and subclassification based on phenotypic, biochemical, and molecular features.
GNAS is the gene encoding the alpha subunit of the stimulatory G protein that mediates coupling of receptors to adenylate cyclase via Gsα. GNAS encodes for Gsα and other transcripts from the maternal and paternal alleles. PHP type 1a is due to mutations in the coding exons of Gsα, resulting in impaired renal cyclic adenosine monophosphate (cAMP) responses to PTH.2–4 It is further appreciated that GNAS is a complex gene that is imprinted at four distinct differentially methylated regions (DMRs). Because of this, some transcripts are derived only from one of the parental alleles. The Gsα expression from the paternal allele is silenced in tissues like the proximal tubule of the kidney. This means there is little or no Gsα expression when a patient carries a maternal allele with a loss of function mutation. PHP type 1b is a paternally-imprinted disorder. Thus, patients who inherit the mutation in GNAS from their mothers show the mineral and PTH abnormalities of PHP. Patients who inherit the mutation in GNAS from their fathers do not. Patients with PHP-1a generally have the AHO phenotype. PHP-1b may occur sporadically or in families, and the genetic basis for the disorder is abnormalities or defects in GNAS methylation. Patients with PHP-1b present with hypocalcemia, hyperparathyroidism, and impaired cAMP responses to PTH but lack the typical clinical features of AHO; moreover, they have normal Gsα activity levels in tissues like fibroblasts and red blood cells, where paternal Gsα expression is not silenced.
The molecular defects responsible for type 1b PHP include epigenetic imprinting abnormalities of the GNAS gene (20q13.3).2, 4, 5 Many patients have lost key regulatory sites of DNA methylation through gene deletion.2, 4 A recent report has also demonstrated that a heterozygous mutation in the regulatory subunit of protein kinase A (PKA) causes reduced PKA signaling.6 This mutation interferes with the capacity of the regulatory subunit of PKA to dissociate from the catalytic subunit and thereby the activation of PKA. Patients demonstrate features of PHP1a with low or low normal serum calcium levels, hyperphosphatemia, secondary hyperparathyroidism, acrodysostosis, and varying degrees of multihormone resistance. This indicates that PHP1a has both molecular and phenotypic variability.
It is noteworthy that, accompanying resistance to PTH in the kidney, many patients with both type 1a and 1b PHP demonstrate responsiveness to PTH in bone, manifested by the development of osteitis fibrosa cystica.3, 7 Although classically PTH targets bone and kidney, PTH receptors also exist in other tissues, such as the brain, pancreas, and zona glomerulosa of the adrenal cortex.8–10 These include both type 1 and 2 PTH receptors. Some of these receptors respond to the ligand tubular infundibular peptide 39 (TIP 39) as well as to PTH. The role of the PTH receptor in many of these sites remains unclear. Evidence associating hyperparathyroidism and hyperaldosteronism, both in vitro10, 11and in vivo,12–16 suggest a functional relationship.
We present a case of an initially normotensive woman with both primary hyperaldosteronism and PTH resistance who developed progressively severe secondary hyperparathyroidism and parathyroid gland autonomy that eventually required total parathyroidectomy. Four hyperplastic parathyroid glands were removed, followed immediately by severe hypocalcemia, attributed to “hungry bone syndrome”17 and hypoparathyroidism, which required prolonged hospitalization, calcium infusions, and oral calcitriol. Although her bone pain resolved, hyperaldosteronism persisted.
A 33-year-old African American woman was referred to the General Clinical Research Center (GCRC) at San Francisco General Hospital Medical Center in 1986 for evaluation of hypokalemia and hypocalcemia. Three years previously, she noted facial twitching and muscle spasms and was found to have hypokalemia (2.8 mEq/L) and hypocalcemia (5.9 mg/dL). Although potassium and calcium supplements partially ameliorated her muscle spasms, she also complained of pain in the hips, knees, and feet for approximately 5 years. After an outpatient workup demonstrated renal potassium and calcium wasting, the patient was admitted for further evaluation.
On admission, she complained of chronic hip pain and pain on motion of her knees and back. There was no evidence of intellectual disability. Her family history was unremarkable. Blood pressure was 103/71 mm Hg, weight 90.5 kg, height 160 cm, body mass index (BMI) 35 kg/m2. The patient was obese but had no other stigmata of AHO. Neither Chvostek nor Trousseau signs were present. Her initial laboratory results demonstrated hypokalemia (2.4 mEq/L), hypocalcemia (7.5 mg/dL), elevated alkaline phosphatase activity (155 IU/L: reference range, 30–115 IU/L), and normal levels of albumin (4.4 g/dL), phosphate (3.4 mg/dL), and magnesium (1.8 mg/dL). Serum creatinine was normal (0.7 mg/dL) as was creatinine clearance (103 mL/min). During this and subsequent admissions to the GCRC, the patient underwent extensive metabolic evaluations. The results of these studies and her subsequent clinical course are reviewed in the sections below.
Evaluation of calcium homeostasis
The patient and 6 healthy control subjects were placed on identical low-calcium (5.4 mg/kg/d) and normal-sodium (2 mmol/kg/d) metabolic diets. After equilibration, three to five consecutive 24-hour urine samples were collected, and the results were averaged (Table 1).18 Despite the low-calcium diet, the patient's 24-hour urinary calcium excretion was ∼2.5-fold greater than the average of the control subjects. The patient also had low normal serum calcium levels with markedly elevated PTH levels.19 1,25(OH)2 vitamin D was twice the mean value in the controls studied at the same time on the same diet but within the laboratory's normal range (15–60 pg/mL). The 25-OH vitamin D level was 15 ng/mL (reference range, 10–55 ng/mL). In order to characterize the hypercalciuria further, the patient underwent a test designed to distinguish between renal and alimentary hypercalciuria20 while consuming the diet described above. After an overnight fast, urine samples were collected prior to and for 4 hours after an oral calcium load (900 mg). Urine calcium excretion under fasting conditions was elevated (0.205 mg/mg creatinine) in comparison to normal control subjects (mean ± SD, 0.050 ± 0.040). Following oral calcium administration, urine calcium excretion did not increase in the patient (0.085 mg/mg creatinine) whereas it increased substantially in controls (0.131 ± 0.035). This was interpreted as renal (not alimentary) hypercalciuria.
Table 1. Summary of Baseline Patient Characteristics Relative to Six Healthy Controls on a Low-Calcium Diet
Values in the patient were compared to those in 4 to 6 healthy controls after they had equilibrated on a low-calcium (5.4 mg/kg/d), normal-sodium (2 mol/kg/d) metabolic diet. Values for the controls are mean ± SEM.
PTH = parathyroid hormone; PRA = plasma renin activity.
PTH levels were measured in a mid-molecule PTH assay.19
Values of p (two-tailed) were determined from the Z-score derived by comparing the patient to the controls.18
An Ellsworth-Howard test to assess renal sensitivity to infused PTH was performed using human PTH (1–34).21, 22 Unlike the 10 control subjects who all showed prompt increases in cAMP excretion, no such increase was seen in the patient, indicating renal resistance to PTH (Fig. 1, top). Fractional excretion of phosphate increased minimally (Fig. 1, bottom), but far less than the robust phosphaturic response reported by Ellsworth and Howard.21
Evaluation of the skeleton
Skeletal X-rays demonstrated multiple cystic lesions, some of which were expansive, consistent with osteitis fibrosa cystica (Fig. 2). The spine initially showed a pattern compatible with a “rugger jersey” spine (Fig. 2A). On later images over time, a few of the vertebral bodies developed cystic changes compatible with brown tumors (not shown). The pelvis demonstrated multiple cystic lesions of various sizes in the sacrum and iliac wings (Fig. 2B). That these were cystic lesions indicative of brown tumors (and not overlying bowel gas) was confirmed on images of the lateral spine showing the bony cystic lesions in areas without overlying bowel loops. Additionally, the right acromion showed expansion of the bone and endosteal resorption and multiple cystic lesions with septations of residual trabeculae (Fig. 2C). Iliac crest bone biopsy following dual tetracycline labeling performed in 1987 showed increased bone formation and resorption indicative of high bone turnover (Fig. 3A, B).23 The intense stimulus for activation of remodeling was manifested in the increased numbers of both osteoblasts and osteoclasts lining the bone as seen on Goldner staining (Fig. 3C–F). In addition, there was evidence of significant fibrosis, increased osteoid accumulation, and extensive woven mineralized and unmineralized osteoid. Bone histomorphometric analysis showed increased total bone volume, osteoid volume and thickness, eroded surface, and mineral apposition and bone formation rates (Table 2). These findings are compatible with hyperparathyroid bone disease coupled with a mineralization defect, the latter perhaps from relative or absolute vitamin D deficiency. The elevated mineralization rate is seen with woven bone formation.
Table 2. Bone Histomorphometric Indices in the Patient After Dual Tetracycline Labeling23
Trabecular bone volume
Relative osteoid volume
Mineral apposition rate
Bone formation rate
At multiple points during her evaluation, the patient was offered GNAS sequencing and more recently analysis of GNAS methylation status, but she consistently declined.
Evaluation of potassium homeostasis
On a normal-sodium (2 mmol/kg/d) and normal-potassium (1.5 mmol/kg/d) diet, the patient continued to excrete large quantities of potassium despite persistent hypokalemia (Table 1), providing evidence for renal potassium wasting. Urinary aldosterone excretion rate was elevated and plasma renin activity was suppressed in comparison to controls (Table 1). Plasma aldosterone was 23.5 ng/dL after overnight recumbency (reference range, 4–12 ng/dL) and increased to 43.6 ng/dL in response to upright posture, accompanying an increase in plasma renin activity (0.2 to 0.9 ng/mL/h). Plasma aldosterone levels increased markedly in response to infusion of des-Asp1-angiotensin II (angiotensin III) in the patient compared with the response seen in 8 control subjects (Fig. 4).24 No adrenal adenoma was noted on abdominal CT scan.
Both dietary sodium restriction (16 mmol/d) and spironolactone therapy (300 mg/d) reduced hypercalciuria and ameliorated hypocalcemia (Fig. 5), but hyperparathyroidism persisted with values in the then-current PTH mid-molecule assay of 500 to 590 µL-Eq/mL (reference range < 40).19 In 1990, the patient underwent sequential intravenous infusions of calcium gluconate and ethylenediaminetetraacetic acid (EDTA) to evaluate whether or not her parathyroid glands were responsive to changes in serum calcium.25 Figure 6 shows that her intact PTH levels could be suppressed when serum calcium was increased, and PTH increased further when calcium levels were lowered. However, the minimum and maximum PTH levels observed with infusion of calcium and EDTA, respectively, were much higher than those in healthy men25, 26 and women26 studied in a similar fashion. Such secretory dynamics are consistent with a large increase in parathyroid cell mass.27
Calcitriol (up to 1 µg/d) and calcium supplements (1800 mg/d) were added in an attempt to suppress her markedly elevated PTH levels (Fig. 7). However, PTH levels remained elevated even though her serum calcium levels normalized. Her suboptimal PTH suppression in response to medical therapy was attributed in part to patient nonadherence and to hyperplasia of the parathyroid glands. During this period, her metabolic bone disease progressed, with several pathologic fractures and persistent bone pain.
The patient was lost to follow-up, but re-established care in 2004. At that time, she had discontinued all medical therapy and was noted to have impaired renal function (creatinine 1.5 mg/dL) and hypertension (blood pressure 179/95 mm Hg). Laboratory studies showed hypercalcemia (total Ca 11.2 mg/dL) and hyperparathyroidism (intact PTH 1328 pg/mL; reference range, 14–72 pg/mL). Cinacalcet (30 mg given twice daily) failed to lower her serum calcium or PTH levels. Sestamibi scan in 2005 suggested a left inferior parathyroid tumor with no evidence of activity outside the neck.
Parathyroidectomy was recommended, but the patient declined. She was seen only intermittently in a local primary care clinic over the next 4 years where her hypertension was managed with enalapril, nifedipine, hydrochlorthiazide, and spironolactone. By June 2008, the patient's condition continued to deteriorate, and she was admitted to an outside hospital where she was noted to have hyperkalemia (K 5.8 mEq/L), hypercalcemia (total Ca 12.4 mg/dL), worsening renal function (creatinine 2.7 mg/dL), and severe normocytic anemia (Hgb 8.0 g/dL). Intact PTH was markedly elevated (2730 pg/mL) as was alkaline phosphatase activity (821 IU/L; reference range, 42–98 IU/L). 25-OH vitamin D was 19.9 ng/mL (reference range, 20–100 ng/mL) and 1,25(OH)2 vitamin D was 28 pg/mL (reference range, 18–72 pg/mL). Bone X-rays showed progressive osteitis fibrosa cystica with involvement of the iliac bones, femoral neck and intertrochanteric regions, and all lumbar vertebrae. Hyperkalemia resolved with discontinuation of enalapril and spironolactone, and hypercalcemia improved with saline hydration and intravenous furosemide. Anemia was treated with erythropoietin (Epogen) and oral iron.
Neck ultrasound revealed a left-sided solid nodule measuring approximately 5 cm in greatest dimension, that appeared to lie within the border of the thyroid gland and that corresponded to the region of increased activity on delayed washout noted on sestamibi parathyroid scan from 2005 and on a more recent scan done at the referring hospital. Given her worsening bone pain, hyperparathyroidism, and hypercalcemia, the patient finally consented to surgical intervention.
Due to her reluctance to undergo surgery, her significant anemia, and her poor response to Epogen therapy, the chosen surgical approach was total parathyroidectomy with cryopreservation of parathyroid tissue and possible autotransplantation at a subsequent date. Neck exploration was done on January 20, 2009. Preoperative serum PTH was 3064 pg/mL, calcium 10.0 mg/dL, magnesium 1.9 mg/dL, phosphate 3.8 mg/dL, alkaline phosphatase 1174 U/L (reference range, 42–98 IU/L), and serum creatinine 2.25 mg/dL. Bilateral cervical exploration revealed a partially intrathyroidal left inferior parathyroid gland, which was removed along with the left thyroid lobe. Subsequently, the remaining three enlarged parathyroid glands were also removed, and a portion of the left superior parathyroid gland was cryopreserved. PTH levels were monitored intraoperatively, falling to 240 pg/mL 10 minutes after the last gland was removed and to 10.9 pg/mL the following morning. Severe postoperative hypocalcemia due to hypoparathyroidism and the “hungry bone syndrome” ensued (serum Ca 6.3 mg/dL with normal albumin, magnesium, and phosphate on postoperative day 4). Continuous calcium infusion and oral calcitriol (up to 3 µg bid) eventually controlled the hypocalcemia. A brief trial of teriparatide was attempted, but was promptly discontinued due to immediate recurrence of severe bone pain, which had resolved promptly following surgery. High doses of calcium carbonate (1500 mg qid) and calcitriol (3 µg bid) allowed eventual discharge from the hospital after 4 weeks of calcium infusions. Serum alkaline phosphatase decreased from a peak of 1574 to 473 IU/L at discharge. Although controlling the PTH levels ameliorated bone pain, hyperaldosteronism persisted.
The left lobe of the thyroid gland had no pathologic findings. All four parathyroid glands were considered to be hyperplastic: left inferior (4.7 × 2.3 × 1.5 cm), right inferior (2.0 × 1.7 × 0.4 cm), right superior (2.3 × 1.5 × 0.5 cm), and left superior (one “half” was cryopreserved; the remainder was 3.0 × 0.7 × 0.5 cm).
This patient presented with two distinct endocrine disorders, both with unique variations: PHP and normotensive primary hyperaldosteronism. The initial evaluation for hypocalcemia with high PTH levels revealed an impaired urinary cAMP response to PTH infusion which, in the absence of phenotypic features of AHO, supported a diagnosis of PHP-1b. Of note was marked bone responsiveness to PTH as evidenced by severe osteitis fibrosa cystica. The bone biopsy in this patient was done prior to the onset of renal insufficiency and showed features of both hyperparathyroid bone disease and mineralization abnormalities. These biopsy findings plus osteosclerosis have been previously reported in PHP-1 patients.7, 28–31 The target organ manifestations seen in PHP-1b include renal resistance to PTH and varying degrees of bone sensitivity to PTH, ranging from skeletal resistance to the hormone to normal sensitivity manifested by osteitis fibrosa cystica or low bone mineral density on dual-energy X-ray absorptiometry.32–35 Cultured osteoblast-like cells from a patient with PHP type 1b demonstrated normal cAMP responsiveness to PTH despite the lack of a renal response.7
Another atypical aspect of this patient's presumed PHP type 1b is the severity of her hypercalciuria. Urine calcium excretion rates tend to be normal off therapy in patients with PHP.36–38 The portion of the kidney responsible for calcium reabsorption normally responds to PTH in this disorder. Atypically for PHP, this patient had 1,25(OH)2D concentrations in the upper portion of the normal range.39 The 1,25(OH)D levels tend to be low in PHP given the impairment of renal 1-hydroxylation step which PTH drives in the proximal tubule. However, relatively few patients have such chronically elevated PTH and chronically low serum calcium levels as this patient did.40
The initial evaluation of hypokalemia and renal potassium wasting supported a diagnosis of primary hyperaldosteronism. The robust increase in plasma aldosterone with upright posture, as well as the exaggerated aldosterone response to infusion of angiotensin III, is suggestive of idiopathic adrenal hyperplasia rather than an aldosterone-producing adenoma.41, 42 This formulation is further supported by the absence of an adrenal mass on CT imaging.
Atypically for hyperaldosteronism, this patient did not have hypertension when she first presented. Only later in her clinical course, when she developed chronic kidney disease and became hypercalcemic, did she develop hypertension. Normotensive primary hyperaldosteronism is distinctly rare.43–45 The absence of volume expansion or ingestion of a low salt diet might account for the lack of hypertension in such individuals, but unlikely in this patient given her markedly elevated urinary sodium excretion on a self-selected diet. It is possible that the presence of PHP with hypocalcemia and renal resistance to PTH may have attenuated or prevented the hypertensive effects of aldosterone. Acute hypocalcemia causes relative vasodilatation and orthostatic hypotension in normal subjects and those with chronic kidney disease.46, 47 The mechanism for this is thought to be decreased myocardial as well as smooth muscle contractility because calcium is important in generating muscle contractions.48 However, patients with chronic hypocalcemia, with conditions like PHP, have not shown a tendency to hypotension.46 Thus, the cause for our patient's normal blood pressure, in the setting of primary hyperaldosteronism, remains unexplained.
Renal hypercalciuria in this patient may have been due primarily to the concomitant primary hyperaldosteronism combined with a high-salt diet. Expansion of effective circulating volume in patients with hyperaldosteronism decreases proximal tubule reabsorption of calcium as well as sodium because calcium reabsorption is coupled to transtubular sodium uptake.49 Although the distal nephron can reabsorb calcium, excessive delivery can overwhelm this absorptive capacity. In addition, potassium depletion causes intracellular acidosis50 that further inhibits calcium reabsorption in the distal tubule. Finally, aldosterone, by augmenting sodium chloride co-transport in the distal nephron,51 can lead indirectly to impaired calcium reabsorption. The reduction in urine calcium excretion in response to either a low-salt diet or spironolactone supported the notion that primary hyperaldosteronism was the main cause of the hypercalciuria in this patient. The very high PTH levels with bone responsiveness to PTH may have played some role in determining the degree of aldosterone-induced hypercalciuria, specifically by limiting the severity of hypocalcemia. Given the severity of the osteitis fibrosa cystica and multiple lytic lesions, one speculation is that the reduction in renal calcium reabsorption was insufficiently counteracted by hypocalcemia-related reduced glomerular filtration of calcium because excessive calcium released from bone prevented serum calcium levels from decreasing to the levels to which it otherwise would have decreased from the aldosterone-induced volume expansion–caused renal calcium wasting. Thus, the convergence of these two endocrine disorders and the duration they were unchecked may have ultimately caused the patient's chronic kidney disease.
Once the chronic kidney disease progressed, autonomous hyperparathyroidism clearly emerged. Her parathyroid gland pathology supports the history of progressive hyperplasia due to PTH resistance (ie, PHP) combined with the emergence of autonomous function confirmed by nonsuppressible PTH levels in the face of hypercalcemia, calcitriol, and the calcimimetic cinacalcet. Based on evidence from in vitro studies that PTH may raise intracellular calcium levels in cultured adrenal zona glomerulosa cells10, 11 and the ability of increases in intracellular calcium to stimulate aldosterone secretion,11, 13 we hypothesized that the markedly elevated PTH levels in this patient might be the cause for the enhanced aldosterone secretion. During her early clinical course, we attempted to test this hypothesis by a variety of approaches designed to reduce the patient's PTH levels, but we never succeeded. By the time the PTH levels were reduced by total parathyroidectomy, the patient had developed advanced renal insufficiency and was being treated with multiple medications that precluded the ability to discern whether or not aldosterone was affected, even by dramatic lowering of PTH. Similarly, Maniero and colleagues16 described a patient with concomitant primary hyperparathyroidism and primary aldosteronism, and postulated a role for PTH in the stimulation of aldosterone secretion. That group demonstrated the presence of PTH receptors in aldosterone-producing adrenocortical nodules; however, the correction of hyperaldosteronism by adrenalectomy prior to the discovery and treatment of the hyperparathyroidism precluded an in vivo confirmation of their hypothesis.
All authors state that they have no conflicts of interest.
We thank the GCRC staff for their assistance in conducting the clinical studies, and Mike Wen for his technical assistance with the figures.
Authors' roles: Study design: BRD, AS, AAP, DS, MS. Study conduct: JP-S, BRD, AP, RR, AS, Q-YD, AAP, DS, MS. Data collection: JP-S, BRD, AS, AAP, DS, MS. Data analysis: JP-S, BRD, RR, VG, AS, AAP, DS, MS. Data interpretation: JP-S, BRD, RR, AS, AAP, DS, MS. Drafting manuscript: JP-S, BRD, AS, AAP, DS, MS. Revising manuscript content: JP-S, BRD, AS, AAP, DS, MS. Approving final version of the manuscript: All. JP-S, DS, and MS take responsibility for the integrity of the data analysis.