Introduction

The classic presentation of primary hyperparathyroidism (PHPT), that of “stones, moans and groans”, is rarely seen in contemporary clinical practice. With the introduction of routine biochemical screening in the 1970s, the early detection of modest elevations in serum calcium and early diagnosis of PHPT in patients with very subtle or no demonstrable symptoms of parathyroid disease became possible. This led to an inevitable increase in the prevalence of PHPT worldwide, such that within a very few years after the introduction of multiphasic biochemical screening, the estimated prevalence in the USA had increased by sixfold, with more than 50% of patients reported as asymptomatic [1].

The earlier recognition of PHPT, which in the majority of cases is the result of a parathyroid adenoma (PA), has resulted in an increase in the number of patients undergoing parathyroidectomy. The “traditional” approach to the surgical treatment of hyperparathyroidism with bilateral neck exploration was done to identify all parathyroid glands, followed by removal of an adenoma with or without biopsy of the remaining glands. Although cumbersome and time consuming, this “traditional” approach has continued to find favour amongst endocrine surgeons. However, the development of minimally invasive surgery and its application to parathyroidectomy (MIP) has dramatically reduced morbidity, hospital stay and cost, with better cosmetic results. Successful MIP demands accurate pre-operative localisation of the abnormal gland(s), with reliance on a variety of imaging techniques—most recently on intra-operative guided MIP with a gamma probe, using the in vivo uptake of a radiopharmaceutical that shows a preferential retention in parathyroid adenomas.

In this article we review the value of various imaging modalities in the diagnosis of PHPT and the evolving role of scintigraphy in the pre-operative localisation and intra-operative extirpation of parathyroid adenomas.

The parathyroid gland

Anatomy and physiology of the parathyroid

The parathyroid glands originate from the endoderm of the third and fourth pharyngeal pouches and migrate during foetal life toward their final juxtathyroidal location. This migration explains why surgical localisation can sometimes be problematic. In most cases, the upper glands migrate to a position posterior to the middle and upper thirds of the thyroid lobe and posterior to the recurrent laryngeal nerves. The location of the inferior glands is more variable, and in 50% of cases they are found posterior or lateral to the lower pole of the thyroid lobe. They can also be intrathyroidal, intrathymic or within the thyrothymic ligament. It is estimated that up to 20% of parathyroid glands are ectopic and can be found anywhere in the mediastinum and occasionally within the carotid sheath. The position may vary even further when normal glands become adenomatous or hyperplastic.

The secretory balance of the parathyroid and its cellular proliferation is regulated by the calcium ion through a specific receptor expressed on the cell membrane. When activated, these cells secrete parathyroid hormone (PTH), an 84-amino acid single-chain polypeptide with a molecular weight of 9,500 that controls the level of ionised calcium in blood and extracellular fluid. The major determinant of PTH secretion is ionised calcium, where small reductions in extracellular concentrations result in increased PTH secretion with reciprocal effects on the kidney to restore normocalcaemia. This is achieved by increasing tubular reabsorption of calcium, increasing excretion of phosphate and increasing the transformation of precursors into the active form of vitamin D, which, in turn, stimulates increased absorption of calcium in the gastrointestinal tract.

Primary hyperparathyroidism

Primary hyperparathyroidism is a clinical condition characterised by excessive secretion of PTH that is inappropriate for the extracellular calcium concentration. The main abnormality in hyperparathyroidism may relate to an unidentified genetic mutation (or mutations) causing failure of parathyroid cells to reduce PTH secretion when serum calcium is elevated, and this could be a consequence of abnormalities in the expression of calcium receptors located on the cell surface. There is evidence for activation of oncogenes and inactivation of tumour suppressor genes in the non-familial disease. Exposure of the neck and chest during radiation therapy for benign diseases, including 131I treatment for Graves’ disease, is another risk factor for the development of hyperparathyroidism. Interestingly, such risk association has not been demonstrated in 131I treatment of thyroid cancer.

In the majority of cases (80–85%), PHPT is caused by one or more parathyroid adenomas, but it can be the result of parathyroid hyperplasia in 15–20% of cases. Hyperparathyroidism due to hyperplasia may also be a component of familial syndromes such as multiple endocrine neoplasia type 1 (MEN-1) (87–97%), MEN-2 (5–20%) and familial hypocalciuric hypercalcaemia. Parathyroid carcinoma is a rare cause of PHPT, accounting for less than 1% of all cases.

Other forms of hyperparathyroidism

Secondary hyperparathyroidism is commonly associated with chronic renal failure. Renal impairment results in phosphate retention leading to hypocalcaemia, which in turn stimulates PTH secretion. In addition, there is an impaired renal response to increased PTH stimulation coupled with skeletal resistance to the action of PTH, due in part to a direct inhibiting effect of phosphorus retention. All these mechanisms progressively augment and perpetuate the hyperparathyroid state.

Tertiary hyperparathyroidism follows some cases of secondary hyperparathyroidism when biochemical abnormalities persist despite successful renal transplantation. In this instance, hyperparathyroidism is attributed to the development of functional autonomy in one or more parathyroid glands. This term is also used for patients in whom the PTH response caused by secondary hyperparathyroidism exceeds the hypocalcaemic demand, resulting in persistent hypercalcaemia.

Persistent hyperparathyroidism occurs in 5–10% of all patients who undergo surgery for PHPT, with continuation of the pre-operative abnormalities in calcium metabolism in the immediate postoperative period. It may result from failure of localisation of adenomas, inadequate resection of unrecognised multigland disease and/or the presence of metastatic parathyroid carcinoma. It is particularly frequent in patients with familial hyperparathyroidism, especially the MEN-1 syndrome. Hyperparathyroidism presenting after a period of more than 6 months of normocalcaemia following surgery is termed “recurrent hyperparathyroidism”, and is commonly due to continued growth of the remaining parathyroid tissues.

Surgery in PHPT

Bilateral neck exploration

Bilateral neck exploration is considered standard treatment of PHPT [2]. It consists in the excision of any grossly enlarged gland with or without biopsy of the remaining glands and, in experienced hands, has a 95% success rate with minimal morbidity. When multiple glands are enlarged, the operative techniques include a 3 1/2-gland resection or, less commonly, four-gland resection with subsequent autotransplantation. Success depends primarily on the experience of the surgeon in recognising the difference between enlarged and normal-sized glands, taking into consideration that the size of the gland does not always correlate with the secretion of PTH. In the absence of pre-operative or intra-operative localisation, the only certain method of finding the offending glands is a thorough neck and mediastinal exploration.

Minimally invasive parathyroidectomy

The development of fast and efficient scintigraphic techniques for pre-operative localisation of parathyroid adenomas, particularly with the use of 99mTc-MIBI, encouraged the introduction of “focussed” or minimally invasive parathyroidectomy (MIP) [3–5]. MIP is associated with less morbidity and comparable cure rates [5, 6], and is gradually replacing the traditional four-gland exploration as the procedure of choice in many institutions. Other ancillary procedures, including gamma probe-guided exploration and endoscopic techniques, have improved the accuracy and feasibility of MIP. Once successful localisation has been achieved, MIP can be employed with obvious advantages: the incision is small with minimal dissection, postoperative complications are minimal and hospital stay is shorter. MIP is less costly and may be performed as an outpatient procedure under local anaesthesia.

When available, quick PTH (QPTH) assay is extremely helpful in confirming the successful removal of the adenoma. A fall in the QPTH level by greater than 50% compared with the pre-operative level, assessed at 5–10 min after excision of a suspected adenoma, is considered indicative of successful surgery. The sensitivity, specificity and overall accuracy of QPTH are 98%, 94% and 97%, respectively [7, 8].

Role of imaging in pre-operative localisation

Anatomical imaging

Imaging plays a key role in pre-operative localisation of parathyroid adenomas (Table 1). Ultrasound (US) is commonly the first diagnostic imaging method used in the localisation of a parathyroid adenoma. US is widely available, non-invasive and cost effective, but remains highly operator dependent [9] with a sensitivity that varies as a function of the size and location of the adenoma. The typical US appearance is of a round or oval, homogeneous, hypoechoic, hypervascular nodule with an echogenic capsule, but other atypical appearances are common and may lead to false negative results. Further limitations are encountered in the presence of a multinodular goitre, which can mimic or mask a parathyroid adenoma, and in substernal, retrotracheal and retro-oesophageal adenomas due to acoustic shadowing from overlying bone or air. Therefore, an ectopic adenoma cannot be excluded on the basis of a US examination alone. The reported sensitivity of US in the detection of parathyroid adenoma is 65–85% [10]. However, the sensitivity falls to 40% in patients who have had prior failed surgical exploration [11].

Table 1 Imaging techniques in primary hyperparathyroidism
Full size table

The cumulative sensitivity is increased when US is combined with other imaging modalities, such as scintigraphy, particularly in differentiating enlarged parathyroid glands from thyroid nodules [12–15]—an important consideration in areas with endemic nodular goitre.

Contrast-enhanced computed tomography (CT), with thin collimation, has a sensitivity that varies from 46% to 87%, with adenomas usually demonstrating avid contrast enhancement [10]. The high anatomical definition of CT is useful in ectopic, mediastinal adenomas, particularly those located in retrotracheal, retro-oesophageal and mediastinal spaces, but the performance is poor in ectopic lesions in the lower neck and lesions close to or within the thyroid gland [16–18]. Prior neck surgery affects the sensitivity of CT imaging, and artefacts caused by metallic clips have been shown to reduce sensitivity to 46–58% [18, 19]. Other disadvantages include movement artefacts (respiration and swallowing), the use of iodinated intravenous contrast and relatively high radiation exposure.

The ability of magnetic resonance imaging (MRI) to characterise nodular lesions based on intensity changes in the T1- and T2-weighted images makes it ideal for the detection of parathyroid adenomas, with improved detection by the addition of contrast enhancement and three-dimensional reconstruction [17]. Reported sensitivity reaches 80%, but due to difficulties in distinguishing parathyroid adenomas from hyperplasia or carcinoma, specificity has been found to be consistently low [20–23]. The main disadvantages of MRI are its limited availability, high cost and long scan times. CT and MRI are therefore useful in patients with recurrent or persistent hyperparathyroidism when the probability of ectopic glands is high.

Selective venous sampling (SVS) is another imaging technique employed in the localisation of parathyroid adenoma. It involves collection of venous blood samples from the cervical and mediastinal veins, via a femoral vein approach. Localisation is successful when a 1.5-fold increase in PTH concentration compared with the femoral vein is achieved, followed by a decreasing gradient in consecutive samples [24]. SVS is an invasive procedure and is often used when non-invasive imaging modalities are inconclusive or in cases of recurrent or persistent hyperparathyroidism following surgery.

Scintigraphic imaging

The lack of an ideal anatomical imaging modality to localise parathyroid adenomas prompted a search for alternative or complementary functional imaging procedures, but this endeavour was hampered by the failure to produce radiopharmaceuticals that could specifically target the parathyroid gland. Scintigraphic assessment of parathyroid tissue is made all the more difficult by its proximity to the thyroid gland, and earlier efforts focussed on finding a scintigraphic method to separate these tissues. Earlier success was based on taking advantage of the different uptake mechanisms of radiotracers by the thyroid and parathyroid glands (75Se-methionine or 201Tl chloride) and those preferentially accumulated by the thyroid (iodine or 99mTc-pertechnetate) to obtain a “subtraction image” that represented uptake in the parathyroid only. However, subtraction scans were technically demanding and involved a high radiation burden to the patient [25], while the results were not optimal owing to variability and low reproducibility in interpretation among different centres [26].

The introduction of 99mTc-MIBI scintigraphy in clinical practice in 1989 by Coakley and co-workers [27] substantially improved the role of pre-operative scintigraphic imaging in hyperparathyroid patients. Localisation of 99mTc-MIBI in the parathyroid tissue is based on a combination of blood flow, gland size and mitochondrial activity [28], which is similar to the mechanism of uptake in the thyroid. However, the washout rate from the two glands is different, with faster release of 99mTc-MIBI from the thyroid compared with the parathyroid, allowing for successful visualisation of the parathyroid when imaging is delayed for 1.5–2 h. This differential retention might be related to some down-regulation of the P-glycoprotein system in the parathyroid adenomas, resulting in delayed efflux of 99mTc-MIBI, [29]. However, in parathyroid hyperplasia these so-called multidrug resistance-associated molecules are overexpressed, may result in a faster rate of radiotracer washout and may be a contributing factor to false negative scintigraphy [30, 31]. Further, preliminary data suggest the possibility of inhomogeneity of the histological phenotype expressed in parathyroids of patients with multiglandular disease that could lead to incomplete identification of affected glands and persistent hyperparathyroidism [32].

The single-tracer dual-phase scintigraphy protocol, based on differential washout, was originally described by Taillefer et al. [33] and gained acceptance owing to its ease of use. However, the observation that 99mTc-MIBI can accumulate in solid thyroid nodules reduced specificity, particularly in geographic areas with a high prevalence of nodular goitre [34, 35], while reduced sensitivity was also noted in association with rapid 99mTc-MIBI washout from some parathyroid adenomas [36].

This led to the adoption of a modified protocol, dual-tracer subtraction scintigraphy, whereby a second radiopharmaceutical accumulating specifically in the thyroid gland was added. The suggested protocols (with inclusion of the entire chest in the imaging field) have included:

  1. 1.

    123I/99mTc-MIBI dual-tracer subtraction technique. This protocol is difficult to adopt clinically owing to the high cost and limited availability of 123I.

  2. 2.

    99m\({\text{TcO}}^{{\text{ - }}}_{{\text{4}}} \)/99mTc-MIBI dual-tracer subtraction technique. Patients are injected with 40 MBq of 99mTc-pertechnetate and imaging is performed 20 min post injection. Immediately after completion, patients are injected with 400–500 MBq 99mTc-MIBI without changing the patient’s position, and a 20-min dynamic acquisition is performed. Using this protocol, Geatti and co-workers achieved 95% sensitivity, without any false-positive result [37]. The addition of potassium perchlorate \({\left( {{\text{KClO}}^{{\text{ - }}}_{{\text{4}}} } \right)}\), to achieve a rapid washout of 99mTc-pertechnetate from the thyroid tissue, and neck US enhances the sensitivity of this protocol [12, 38].

With the application of an appropriate imaging protocol, hyperfunctioning parathyroid glands as small as 100 mg can be detected [12, 39–42], with an estimated sensitivity of 85–95%. The addition of single-photon emission computed tomography (SPECT) considerably improves localisation of ectopic adenomas in the retro-oesophageal space or mediastinum [43–47]. It is recommended that SPECT be acquired early rather than late as maximum parathyroid uptake takes place shortly after 99mTc-MIBI injection [48], and in order to avoid early, rapid washout seen in some adenomas [5, 44].

In addition to SPECT, other scintigraphic modifications such as dynamic imaging, pinhole SPECT [49] and hybrid SPECT/CT imaging [50] have been suggested to improve sensitivity. The use of 99mTc-MIBI scintigraphy has resulted in an appreciable improvement in pre-operative localisation and has improved the success rate of MIP with unilateral neck exploration [51–54].

99mTc-tetrofosmin shares some qualities with 99mTc-MIBI though its mechanism of uptake is different, with retention occurring primarily in the cytosol rather than within mitochondria. When used for parathyroid imaging, it has shown slow washout from the thyroid, making it unsuitable for single-tracer, dual-phase scintigraphy [55]. However, the sensitivity of 99mTc-tetrofosmin is substantially improved when it is used in a dual-tracer subtraction protocol with SPECT [56].

Positron emission tomography (PET) using 18F-fluorodeoxyglucose (18F-FDG) has been investigated to detect increased metabolic activity in adenomas, with variable success. Some reports suggest that 18F-FDG PET is more sensitive but less specific than 99mTc-MIBI SPECT [57], while others have reported very low sensitivity [58].

The use of PET with a labelled amino acid such as 11C-methionine has been investigated in parathyroid adenomas. In a study involving 34 patients with primary or recurrent adenomas, Sundin et al., using semi-quantitative standard uptake values (SUV) and transport rate constants, achieved a true positive rate of 85% compared with 59% and 57% for CT and US respectively [59]. In another study, Beggs and Hain investigated 51 patients with suspected adenomas who had negative or equivocal conventional imaging, using 11C-methionine. Their results showed a sensitivity of 83%, a specificity of 100% and an accuracy of 88% [60]. The major drawback of using 11C is its extremely short half-life of 20 min, which limits use to centres in close proximity to a cyclotron.

Gamma probe-guided minimally invasive parathyroidectomy (GP-MIP)

99mTc-MIBI is currently the only radiopharmaceutical employed in the pre- or intra-operative localisation of parathyroid adenoma(s). Intra-operative guidance with a hand-held gamma probe is commonly used at 2–3 h after 99mTc-MIBI injection. A number of protocols have been introduced with the aim of improving parathyroid-to-thyroid count ratios and optimising the performance of the gamma probe.

A single-day protocol has been advocated whereby a full imaging dose of 99mTc-MIBI (740 MBq) is administered, followed by a standard dual-phase imaging protocol at 20 min and 2 h. Gamma probe-guided surgery follows approximately 2.5–3 h later. This protocol offers the advantage of performing both parathyroid scintigraphy and surgery on the same day [61]. However, a prerequisite is that the adenoma has been identified by prior parathyroid scintigraphy and/or other imaging modalities.

A separate-day protocol involves performing 99mTc-MIBI dual-tracer subtraction scintigraphy a few days before surgery to localise the adenoma. On the day of surgery, a small dose of 99mTc-MIBI (37 MBq) is injected in the theatre before the start of the operation, followed by a search in the relevant area using the gamma probe [62, 63]. Intra-operative PTH monitoring is also employed to confirm complete removal of the hyperfunctioning parathyroid tissue. This low-dose protocol offers reduced radiation exposure to the surgical team and is particularly useful in patients with concomitant nodular goitre when separating parathyroid adenomas from adjacent thyroid nodules becomes difficult [12, 37].

The procedure for radio-guided parathyroid surgery starts with external scanning with the gamma probe to locate the highest radioactive spot on the skin surface. After making a small incision, the probe is inserted over the presumed location of the adenoma. The high-pitch signals produced by the gamma probe lead the surgeon towards the location with highest radioactivity. A parathyroid-to-thyroid radioactivity ratio higher than 1.5 suggests the presence of a parathyroid adenoma. After the lesion has been removed, the surgical bed is scanned again to ensure complete removal of the adenoma by establishing a new level of background radioactivity. The errors due to equivocal or false-positive scans are thus decreased by the use of an intra-operative gamma probe. Final assessment of radioactivity in all four quadrants increases confidence in the completeness of the parathyroidectomy. Gamma probe guidance enables the surgeon to perform a small skin incision with an improved cosmetic appearance. The technique can be performed under local anaesthesia with reduced operative time and allows early discharge from hospital [4, 64, 65].

In summary, the GP-MIP is an appropriate approach in the following conditions that are encountered in 60–70% of all cases of PHPT [34, 35, 66]:

  • High probability of a solitary parathyroid adenoma

  • Significant 99mTc-MIBI uptake in the parathyroid adenoma

  • No coexisting 99mTc-MIBI-avid thyroid nodules

  • No history of familial hyperparathyroidism or multiple endocrine neoplasia

  • No history of previous neck irradiation

  • Re-operation for persistent or recurrent hyperparathyroidism and ectopic adenomas

Gamma probe guidance has also been useful when performing a standard bilateral neck exploration, as it increases the accuracy of pre-operative 99mTc-MIBI scintigraphy [42, 67, 68].

Although GP-MIP is helpful in distinguishing a parathyroid adenoma from a thyroid nodule, the presence of a significant nodular goitre is a contraindication to the procedure [62]. Thyroid nodules can give false positive results at both pre-operative and intra-operative localisation [33, 63, 68, 69] owing to 99mTc-MIBI retention [68, 70], and the use of dual-tracer subtraction protocols is generally preferred, particularly in areas where nodular goitre is endemic [68, 71].

Conclusion

The role of scintigraphy in the pre- and intra-operative localisation of parathyroid adenomas, as the main cause of PHPT, has evolved to match the widespread use of minimally invasive parathyroidectomy, which demands accurate pre-operative localisation, particularly in patients undergoing treatment for recurrent or persistent hyperparathyroidism. Anatomical imaging has limited sensitivity and specificity compared with the high efficacy of 99mTc-MIBI using dual-tracer subtraction SPECT protocols. However, the combination of US and scintigraphy offers sufficiently high sensitivity and is successful in accurately identifying parathyroid adenomas. A low dose of 99mTc-MIBI injected immediately before surgery can be easily detected with a hand-held gamma probe and used to guide the surgeon to the location of the parathyroid adenoma during surgery.