Nuclear Oncology, 1 Ed.

CHAPTER 4

PARATHYROID TUMORS

Christopher J. Palestro • Kenneth J. Nichols

INTRODUCTION

Parathyroid Glands

Anatomy and Embryology

The parathyroid glands, typically four in number, are small ellipsoid-shaped structures generally located immediately posterior to the thyroid gland. The normal gland measures approximately 5 to 7 mm in length and 3 to 4 mm in width and weighs approximately 40 to 60 mg. The parathyroid glands arise from the dorsal endoderm of the third and fourth pharyngeal pouches. They undergo differentiation in the fifth week of gestation and lose their pharyngeal connections by the seventh week of gestation.1

The superior glands, together with the upper pole of the thyroid gland, descend from the base of the tongue and eventually come to lie midway along the posterior borders of the thyroid. In about 75% of the population, the superior parathyroid glands are located at the junction of the upper and middle third of the thyroid gland, posterolateral to the cricothyroid junction. In about 20% of the population, they are immediately posterior to the upper poles of the thyroid gland. In up to 5% of the population, the glands are intrathyroidal, and in about 1%, they are retroesophageal in location.1

The inferior parathyroid glands arise from the third pharyngeal pouch and migrate caudally along with the thymus. Their location is more variable than that of the superior glands; they can be found anywhere from above the carotid bifurcation to the mediastinum. In about 40% of the population, they are near, or adjacent to, the lower poles of the thyroid gland. In another 40% of the population, they are near the thymic tongue. Occasionally they are found at the angle of the mandible, in the tracheoesophageal groove, the retroesophageal and pretracheal regions, and even in the pericardium.1

Histology

The normal parathyroid gland is composed of equal amounts of parenchyma and stroma. The chief cells constitute the majority of the parenchymal cells, and are responsible for most of the hormonal secretion. Oxyphil cells usually do not appear before 5 to 7 years of age and gradually increase in number after puberty. They contain abundant mitochondria, but do not possess a significant secretory function. The stroma is made up of adipose tissue with an abundant vascular supply and provides support for the parenchyma.1

Function

The parathyroid glands secrete parathyroid hormone (PTH) which, together with calcitonin and 1,25-hydroxycholecalciferol, acts to maintain calcium homeostasis. PTH is a polypeptide that consists of an 84-amino acid sequence with a molecular weight of 95,000 Da. The 34 amino acids at the amino terminal are the active part of the hormone. PTH acts at several sites in the body to increase serum calcium. In bone, in the presence of 1,25-hydroxycholecalciferol, PTH stimulates both osteoclasts and osteoblasts, even though the net effect is osteolysis. In the kidneys, it acts on the renal tubules, promoting calcium retention and increasing phosphate, sodium, and potassium excretion. PTH also facilitates calcium absorption by the small bowel. The normal fasting serum level of PTH is less than 100 pg/mL.2

Hyperparathyroidism

Oversecretion of PTH results in hyperparathyroidism, which is classified as primary, secondary, or tertiary.

Primary Hyperparathyroidism

The most common cause of hypercalcemia, primary hyperparathyroidism (PHP), is a generalized disorder of calcium, phosphate, and bone metabolism because of the excess PTH secretion. It is a result of the loss of normal feedback control of PTH by extracellular calcium and usually leads to hypercalcemia with an elevated (or inappropriately normal) PTH concentration and resultant hypercalciuria and hypophosphatemia. The estimated incidence of PHP in the United States is approximately 22 cases per 100,000 persons per year and peaks in the seventh decade. Most cases occur in women (74%), but the incidence is similar in men and women before 45 years of age. Head and neck irradiation in childhood and long-term lithium therapy are associated with a greater prevalence of PHP. With increased detection caused by the routine calcium screening, the clinical profile of PHP in Western countries has shifted from a symptomatic disease, characterized by hypercalcemic symptoms, nephrolithiasis, frank bone disease, and neuromuscular symptoms to a disease with few or no specific symptoms.3

Although PHP most commonly occurs as a sporadic disease, it is also associated with various hereditary syndromes including familial isolated hyperparathyroidism, multiple endocrine neoplasia types 1 and 2A, hyperparathyroidism–jaw tumor syndrome, and neonatal severe PHP. Approximately 85% of all cases of PHP are caused by parathyroid adenomas; 10% to 15% are caused by parathyroid hyperplasia; parathyroid carcinoma accounts for less than 1% of the cases.3

Secondary Hyperparathyroidism

Secondary hyperparathyroidism (SHP) is the overproduction of PTH secondary to a chronically low concentration of calcium. Most often it results from chronic renal failure; occasionally it is associated with intestinal malabsorption. Calcium reabsorption by the small intestine is impaired; there is phosphate retention, and because of a lack of 1,25-hydroxycholecalciferol, the normal effect of PTH on bone calcium release is lost. In an attempt to maintain calcium homeostasis, the serum PTH level rises, causing generalized parathyroid gland hyperplasia. Other causes of SHP include rickets, osteomalacia, pseudohyperparathyroidism, and high-dose phosphate therapy in patients with x-linked hypophosphatemia.1,4

The vast majority of patients with SHP are on dialysis and present with a very variable frequency and severity of symptoms. These individuals may develop osteitis fibrosa cystica or osteomalacia which can lead to skeletal deformities and fractures. Bone pain occurs primarily in the thoracolumbar spine and lower extremities and is exacerbated by weight bearing and sudden movements. Soft tissue calcification affects about 25% of patients at the outset of dialysis and nearly 60% of those who have been on dialysis for more than 5 years. Pruritus, attributed to increased calcium concentration in the skin, can be severe and disabling, but usually improves after parathyroidectomy. Calciphylaxis, a rare condition associated with hemodialysis and renal transplantation, is characterized by high PTH concentration and increased serum phosphate product. Patients develop extensive hard, tender subcutaneous plaques that can progress to necrosis, nonhealing ulcers, and gangrene that can be life-threatening.4

There are data that suggest that high PTH levels may be cardiotoxic. In a study of 52 hemodialysis patients, one of the best predictors of left ventricular hypertrophy was an elevated PTH concentration. It was postulated that PTH acts directly on specific PTH receptors in the cardiomyocytes.5

The diagnosis of SHP is confirmed by hyperphosphatemia and normocalcemia or hypocalcemia, together with an elevated intact PTH level. Medical management is directed at maintaining calcium and phosphate levels close to normal levels and suppressing PTH secretion and is successful in the majority of patients. Medical therapy fails in up to 10% of patients and these individuals undergo either subtotal parathyroidectomy or total parathyroidectomy plus autotransplantation of one of the resected glands into the forearm. Typically the gland selected for autotransplantation is small and shows diffuse, rather than the more aggressive nodular, hyperplasia.4

Tertiary Hyperparathyroidism

Tertiary hyperparathyroidism (THP) is a complication of SHP. In the patient with long-standing SHP, the normal autoregulatory feedback mechanism governing PTH secretion is lost, resulting in uncontrolled hormone secretion. The sensitivity of the parathyroid glands to PTH output decreases, and therefore the threshold for inhibiting PTH output increases. The parathyroid glands exhibit gross hyperplasia and focal adenomas may develop. THP can be very difficult to distinguish from PHP because both disorders are characterized by inappropriately elevated PTH in the setting of hypercalcemia. The diagnosis is made based on the patient history and the context of disease. For all practical purposes, THP is almost exclusively a disease of patients with long-standing, advanced renal failure who have undergone successful renal transplant.1,5

Treatment

Although hyperparathyroidism can be managed conservatively if the patient’s condition is stable, surgical removal of the offending gland(s) is the only definitive treatment. At one time, the surgical approach to the patient with hyperparathyroidism consisted of bilateral neck exploration, identification of all four glands, excision of the grossly enlarged gland(s) and biopsy, with intraoperative frozen section, of the remaining glands. The success rate of this approach, for skilled surgeons, is in excess of 90%. Consequently, preoperative parathyroid localization was reserved for cases of recurrent disease or following failed parathyroidectomies. However, especially in patients with PHP, because 80% to 85% are caused by a solitary lesion, unilateral neck exploration or minimally invasive surgery often is sufficient. The success of minimally invasive parathyroidectomy depends on accurate preoperative parathyroid lesion localization, and consequently, the importance of accurate preoperative localization even in surgically naive patients, has increased dramatically in recent years.1

PARATHYROID GLAND LOCALIZATION PROCEDURES

Morphologic Imaging

Ultrasonography

Diagnostic ultrasound of the neck is a useful procedure for localizing the parathyroid glands (Fig. 4.1). The resolution of this technique actually exceeds that of computed tomography (CT) and magnetic resonance imaging (MRI) when a high-frequency probe (10 MHz) is used. Structures can be visualized in a variety of projections, and ultrasound can be used as a guide to biopsy and fine needle aspiration. Another advantage of ultrasound is that patients are not subjected to ionizing radiation. Ultrasound is limited in its ability to identify ectopic parathyroid glands in the retrotracheal and retroesophageal spaces as well as in the mediastinum. False-positive results are usually caused by thyroid nodules or lymph nodes.1 Krusback et al.6 reviewed the results of ultrasound in 100 patients with PHP and found that the sensitivity and specificity of neck ultrasound were 55% and 95%, respectively. For lesions located below the level of the thyroid gland, the sensitivity fell to 44%. Clark et al.,7 in a study of 36 patients with recurrent or persistent hyperparathyroidism, reported that the sensitivity of ultrasound was 50%. Eight of the patients in this series had parathyroid lesions below the level of the sternal notch.

FIGURE 4.1. There is a 2.5- × 2- × 1-cm hypoechoic focus (arrows ) posterior to the lower pole of the right thyroid lobe (arrowheads ) on this transaxial ultrasound image. At surgery a 5,930-mg ectopic right superior parathyroid adenoma was resected.

Computed Tomography

CT occasionally is used to identify enlarged parathyroid glands (Fig. 4.2). In contrast to ultrasound, both the neck and mediastinum can be examined with this technique. Nonopacifying vessels, lymph nodes, and thyroid nodules all can cause false-positive results, and intrathyroidal parathyroid glands can be overlooked. The overall sensitivity of CT in PHP is about 70%.1

FIGURE 4.2. There is a 10- × 6-mm ovoid-shaped soft tissue density (arrow ) anterior to the ascending aorta. A 1,080-mg ectopic left inferior parathyroid adenoma was removed at surgery. Reproduced with permission from Palestro CJ, Tomas MB, Tronco GG. Radionuclide imaging of the parathyroid glands. Sem Nucl Med. 2005;35:266–276.

Recently 4D-CT has been used for preoperative parathyroid lesion localization. This technique involves acquisition of three CT scans: Precontrast, postcontrast, and delayed imaging. 4D-CT uses CT reconstruction of tissue volumes (3D) in conjunction with changes in perfusion contrast over time (1D) to take advantage of the more rapid rate of uptake and washout in parathyroid adenomas compared to normal parathyroid glands and other tissues. In one investigation of 75 patients, sensitivity of 4D-CT for localizing parathyroid lesions to the surgically verified correct quadrant was significantly higher than that of double phase planar MIBI imaging with single photon emission computed tomography (SPECT)/CT (88% versus 65%; p < 0.001).8 A recent application of 4D-CT, augmented by ultrasound, found sensitivity, for localizing parathyroid lesions to the surgically verified correct quadrant, of 75% (N= 12) for lesion weights <150 mg, 89% (N = 65) for 150 to 500 mg, and 94% (N = 62) for >500 mg lesions.9

Magnetic Resonance Imaging

MRI can be used to examine both the neck and the mediastinum. Images can be reconstructed in multiple planes. In general, parathyroid adenomas have a longer relaxation time than thyroid tissue and exhibit higher signal intensity on T2-weighted images. The reported sensitivity of this technique, using surface coils, is about 75%.1

Other Tests

With the development of accurate noninvasive localizing tests, the role of more invasive procedures such as angiography and parathyroid venous sampling which are time consuming, costly, and highly dependent on operator skill, has diminished. These tests are reserved for those situations in which noninvasive modalities have failed to localize the offending gland(s), and even then they are rarely used.1

Molecular Imaging

201Tl/99mTc-Pertechnetate Subtraction Technique

The ability of the parathyroid glands to concentrate a variety of chemical substances, including radiopharmaceuticals, has been exploited for localization purposes. Although several radiopharmaceuticals have been used to localize the parathyroid glands over the past 50 years, the first radionuclide technique for preoperative parathyroid lesion localization to gain widespread acceptance was 201Tl/99mTcO4, or thallium/pertechnetate, subtraction imaging, which was introduced in the early 1980s (Table 4.1). 201Tl, developed as a myocardial perfusion imaging agent, is an inorganic cationic analog of potassium. It is cyclotron-produced, has a half-life of 73 hours, emits a spectrum of 68 to 80 keV mercury x-rays, as well as 135 and 167 keV γ-rays, and with the aid of the sodium–potassium–ATP pump, is actively transported into the cell, where it is integrated rapidly into the intracellular potassium pool. Thallium accumulates not only in parathyroid tissue, but in thyroid tissue as well, and consequently thyroid activity usually has to be removed in order to identify parathyroid activity. This is accomplished by administering 99mTc-pertechnetate, which accumulates in the thyroid gland but not in the parathyroid glands. 201Tl and 99mTc-pertechnetate images are obtained in a single session without moving the patient. The 99mTc-pertechnetate, or thyroid, image is digitally subtracted from the 201Tl, or parathyroid, image. Residual activity in the “subtraction” image represents activity in abnormal parathyroid tissue (Fig. 4.3). There are several limitations to the test. The principal photon energy spectrum, 69 to 80 keV, of 201Tl is suboptimal for γ-camera imaging, and the relatively high radiation-absorbed dose imparted by this tracer limits the amount that can be administered. Consequently, image quality is poor and SPECT is not performed. The sensitivity of the test has varied from as low as 44% to as high as 95%.1

TABLE 4.1

AGENTS USED FOR MOLECULAR IMAGING OF THE PARATHYROID GLANDS

Basso et al.10 reported a sensitivity of 20% for adenomas less than 300 mg in weight and 76% for adenomas weighing more than 1,250 mg. Sandrock et al.,11 in contrast, found no correlation between lesion size and test sensitivity. Hauty et al.,12 in a retrospective review of 49 thallium/technetium parathyroid scans, reported a sensitivity of 78% and an overall accuracy of 73% compared to the 82% sensitivity and 78% accuracy in an extensive review of 14 previous series of a total of 317 scans. Regardless of the variations in the overall sensitivity of the test, 201Tl/99mTc-pertechnetate imaging consistently has been reported to be less sensitive for hyperplastic glands than for adenomas.

FIGURE 4.3. There is focally increased activity (arrow ) just below the lower pole of the left thyroid lobe on the thallium image, but not on the pertechnetate thyroid image. This focus is obvious on the subtraction image (arrow  ). At surgery an ectopic 50-mg left inferior parathyroid adenoma in the thyrothymic tract was resected.

FIGURE 4.4. There is focally increased MIBI activity (arrows  ) at the lower pole of the left thyroid lobe on both early and late images. The thyroid gland, clearly seen on the early image, is barely discernible on the late image. A 690-mg eutopic left inferior parathyroid adenoma was subsequently removed.

99mTc-Sestamibi (MIBI)

Coakley et al.13 reported on MIBI parathyroid imaging more than 20 years ago, and several larger series soon followed. Because of superior image quality, more favorable dosimetry, and improved accuracy, MIBI rapidly replaced 201Tl as the radiopharmaceutical of choice for parathyroid imaging. MIBI (hexakis 2-methoxyisobutyl isonitrile) is a lipophilic, monovalent, cationic isonitrile compound that diffuses passively across the cell membrane, sequestered primarily in the mitochondria, and trapped intracellularly.14,15 The primary route of excretion (33%) is hepatobiliary; approximately 25% is excreted via the kidneys. The critical organ is the large bowel, which receives approximately 0.048 mGy/MBq (5.4 rad/30 mCi). The total body dose is approximately 5 mGy (500 mrad). Tracer retention in parathyroid lesions presumably is related to the presence of oxyphil cells, which are rich in mitochondria, in these lesions.16

Imaging Techniques

Single Isotope Double (Dual) Phase MIBI Imaging. Over the years, numerous methods of performing MIBI parathyroid imaging have been suggested. Taillefer et al.17 introduced the concept of the “single isotope, double phase technique,” which is based on the differential washout rates of MIBI from the thyroid and parathyroid glands. MIBI washes out more rapidly from both normal and abnormal thyroid tissues than from abnormal parathyroid tissue, where it is retained for a longer period of time, presumably because of the presence of mitochondria-rich oxyphil cells in these lesions.16 The single isotope double phase technique is simple and easily performed, requiring only a single injection of MIBI followed by imaging approximately 15 minutes and 1.5 to 3 hours later. A persistent focus of activity on delayed images, relative to thyroid gland activity, indicates a parathyroid lesion (Figs. 4.4 and 4.5). The reported sensitivity of the test has ranged from 43% to 91%. Taillefer et al.,17 in a study of 23 patients, reported 90% sensitivity. Neumann et al.,18 in an investigation of 15 patients, reported sensitivity, specificity, and accuracy of 53%, 86%, and 76%, respectively. In what probably is the largest series reported to date, Nichols et al.,19 in an investigation of 462 patients (534 lesions) with PHP, reported sensitivity, specificity, and accuracy of 84%, 90%, and 87%, respectively for the single isotope double phase technique. These investigators also reported that the test was significantly more sensitive (90% versus 61%) in single gland disease (SGD) than in multigland disease (MGD).

FIGURE 4.5. There is focally increased MIBI activity (arrows ) just below the lower pole of the right thyroid lobe on both early and late images. A second focus of increased activity, a thyroid nodule (arrowhead ). at the level of the midpole of the right thyroid lobe, is seen only on the early images. Figure 4.4 and this figure illustrate the principle of the single isotope double phase MIBI technique: MIBI washes out more rapidly from both normal and abnormal thyroid tissue than from abnormal parathyroid tissue, where it is retained for a longer period of time.

FIGURE 4.6. A 210-mg ectopic left inferior parathyroid adenoma (arrow ). in the thyrothymic tract, is seen only on the early MIBI image.

FIGURE 4.7. There are two foci of increased MIBI uptake seen on both the early and late images. One focus is at the level of the midpole of the right thyroid (arrows ) and the other is at the lower pole of the left thyroid lobe (arrowheads ). Based on the single isotope double phase method both likely are parathyroid lesions.

Subtraction. Some parathyroid lesions do not retain MIBI, whereas some thyroid lesions do, resulting in both false-negative and false-positive double phase studies (Figs. 4.6 and 4.7). Subtraction imaging often is helpful in these situations, especially in patients with concomitant thyroid disease (Fig. 4.8). In addition to MIBI, the subtraction method requires administration of 123I or 99mTc-pertechnetate to obtain a thyroid image. When using 123I, the thyroid image is usually acquired prior to MIBI injection. Wei et al.20 prospectively investigated MIBI/123I thyroid subtraction imaging in 20 patients, including 16 with PHP and 4 with SHP. Among the patients with PHP all 11 solitary adenomas were successfully localized. In four of five patients with diffuse hyperplasia, bilateral localization consistent with enlarged glands was seen. In one patient who previously had undergone parathyroidectomy, recurrent disease was found. Among the four patients with SHP, three had bilateral localization consistent with enlarged glands. Recurrent disease was identified in one patient who previously had undergone parathyroidectomy. Casas et al.,21in an investigation of 22 patients with PHP, including 16 with single adenomas, 5 with diffuse hyperplasia, and 1 with a double adenoma compared MIBI/123I subtraction imaging to high-resolution ultrasound. MIBI/123I subtraction imaging correctly localized all 18 adenomas. All five patients with diffuse hyperplasia had images consistent with diffuse hyperplasia, although delineation of individual glands was not possible. High-resolution ultrasound, in contrast, identified only 11 of the 18 adenomas. Among the five patients with diffuse parathyroid hyperplasia, ultrasound was completely negative in one case, identified a single enlarged gland in two cases, and identified two enlarged glands in two cases. The authors concluded that MIBI/123I subtraction imaging is more sensitive than high-resolution ultrasound for preoperative parathyroid lesion localization.

FIGURE 4.8. By adding thyroid subtraction (with pertechnetate thyroid imaging in this case) to the case illustrated in Figure 4.7, it is apparent that only the left-sided focus (arrowheads ) is a parathyroid lesion; the focus on the right (arrows ) is a thyroid lesion. At surgery a 1,025-mg left inferior parathyroid adenoma was resected.

Wakamatsu et al.22 in an investigation of 39 patients reported that MIBI/123I subtraction was more sensitive (56%) than double phase MIBI (39%), MRI (43%), and ultrasound (51%). Chen et al.23 compared MIBI/123I subtraction to double phase MIBI imaging in 25 patients with recurrent PHP. The sensitivity of MIBI/123I, 70% (19/27), was higher, but not significantly, than that of double phase MIBI, 59% (16/27).

Hindie et al.,24 in a study of 30 patients with PHP, performed simultaneous dual isotope MIBI/123I acquisition. As with sequential acquisition, these investigators found that the subtraction technique using simultaneous dual isotope acquisition was more sensitive than double phase MIBI imaging (94% versus 79%; p < 0.04). The false-positive rate for the subtraction technique was 3% versus 10% for the double phase technique. An advantage of the simultaneous dual isotope acquisition was the shortened study time, which resulted in fewer motion artifacts associated with the prolonged immobilization required for sequential acquisition.

Caveny et al.25 also investigated simultaneous dual isotope 123I/MIBI imaging in 37 patients with PHP. Patients were injected with MIBI 2 hours after 123I was given. Fifteen minutes and three hours after MIBI administration, simultaneous dual isotope acquisitions were performed. These investigators compared three different protocols: Early plus late MIBI images (conventional double phase MIBI imaging) with single phase (early) 123I/MIBI subtraction and double phase (early plus late) 123I/MIBI imaging. The localization success rate of 66% for double phase MIBI was significantly less (p < 0.01) than the 94% success rate for the dual tracer single phase technique and the 90% success rate for dual tracer double phase technique. There was no significant difference between the single and double phase dual tracer techniques. The degree of certainty of localization was significantly higher (p < 0.001) with the dual tracer protocols than with MIBI imaging alone.

Neumann et al.18 investigated the utility of 123I/MIBI SPECT subtraction. They studied 15 patients with PHP who underwent preoperative double phase MIBI SPECT and simultaneous l23I/MIBI subtraction SPECT. At surgery, 17 parathyroid adenomas were found. The sensitivity, specificity, and accuracy were 88%, 97%, and 94%, respectively, for simultaneous 123I/MIBI SPECT subtraction versus 53%, 86%, and 76%, respectively, for double phase MIBI SPECT. The differences in sensitivity and accuracy were statistically significant (p = 0.031 and p = 0.016, respectively).

Although some investigators have used 123I for thyroid imaging, others have used pertechnetate for this purpose. Wei et al.26 performed preoperative MIBI/pertechnetate subtraction imaging on 30 patients with PHP, SHP, or THP. Patients underwent thyroid imaging 30 minutes after pertechnetate injection. Although still under the camera, they were injected with MIBI, after which they immediately underwent parathyroid imaging. The sensitivity of the test was 92% (12/13) for detecting parathyroid adenomas and 79% (37/47) for other parathyroid lesions (46 hyperplasias and one carcinoma). These results were similar to the 123I/MIBI subtraction technique results that these same investigators previously had reported.20 The authors felt that the MIBI/pertechnetate subtraction technique was less cumbersome and faster than the 123I/MIBI technique.

Chen et al.27 compared double phase MIBI imaging with MIBI/pertechnetate and MIBI SPECT in 55 patients. In this investigation, pertechnetate was injected first and thyroid imaging was performed followed by MIBI injection and early MIBI imaging. Two-and-one-half to four hours afterward, late MIBI imaging and SPECT were performed. Using visual comparison the sensitivity of early MIBI/pertechnetate and late MIBI/pertechnetate were 72% to 75% and 73% to 78%, respectively. The sensitivity of computer subtraction of pertechnetate from early MIBI images was 71% to 74%. The sensitivity of double phase MIBI and MIBI SPECT were 62% to 65% and 79%, respectively. The authors concluded that visual comparison of pertechnetate thyroid imaging with early MIBI imaging was sufficient for preoperative parathyroid lesion localization.

Leslie et al.28 compared MIBI/pertechnetate subtraction imaging alone to double phase MIBI and to double phase MIBI plus subtraction imaging in 88 patients, including 68 with single adenomas. After pertechnetate injection patients underwent continuous dynamic imaging for 30 minutes. Midway through the acquisition, patients were injected with MIBI. Delayed MIBI images were performed about 2 hours later. Subtraction images only, double phase MIBI images and subtraction images together with early and late MIBI images were interpreted separately. Among the 68 patients with single adenomas, double phase MIBI imaging correctly localized 49 (72%) of the lesions significantly less than the 58 (85%) that were localized on subtraction images and the 61 (90%) that were localized with the double phase plus subtraction images. Reader confidence was greater with the subtraction-only and combined images than with the MIBI-only images. Results in the 20 patients with MGD were much less satisfactory for all three methods, with no significant differences among them.

FIGURE 4.9. False-negative subtraction study following iodinated contrast. A 240-mg right inferior parathyroid adenoma is clearly seen on early and late MIBI images. The pertechnetate thyroid image is virtually identical to the late MIBI image resulting in a false-negative subtraction image. This patient had undergone a CT study with contrast approximately 1 week previously. The importance of obtaining a thorough pertinent history cannot be overemphasized.

Thyroid uptake of MIBI can be three to five times higher than parathyroid uptake, which potentially could mask low MIBI uptake by some parathyroid lesions. Rubello et al.29 reasoned that this problem could be minimized by administering potassium per­chlorate (KClO4), which causes a rapid washout of pertechnetate from the thyroid gland. In their investigation of MIBI/pertechnetate imaging, they administered KClO4 after completing thyroid imaging and before injecting MIBI. All 18 adenomas were correctly localized with this method.

Nichols et al.,19 as part of a larger investigation, retrospectively studied MIBI/pertechnetate subtraction imaging. In their study, pertechnetate was injected midway through the late MIBI acquisition. Following image normalization and background subtraction, the pertechnetate thyroid image was subtracted from the late MIBI image. Sensitivity and specificity of double phase MIBI imaging were 84% and 90%. Sensitivity and specificity of the subtraction images read in isolation were 80% and 93%, respectively. When the subtraction images were read together with the early and late MIBI images sensitivity and specificity both were 88%. For all three methods, sensitivity was significantly higher for SGD than for MGD.

The subtraction technique, though useful, has limitations. Patient motion during data acquisition may lead to misregistration of the MIBI and thyroid images, resulting in a false-positive study. When 123I is used for thyroid imaging, this potential problem can be reduced by simultaneous dual isotope acquisition. In our institution, where pertechnetate is used for the thyroid image, we try to minimize patient motion by placing an intravenous line in the patient’s arm before beginning the test, which can be accessed for pertechnetate injection without disturbing the patient or interrupting imaging. Instead of acquiring separate late MIBI and thyroid images, we perform a dynamic acquisition consisting of 25 2-minute images; pertechnetate is injected at about the 11th frame. Frames with excessive movement can be eliminated without discarding the entire data set after which, with computer manipulation, the parathyroid and thyroid images are separated, group-added, pixel re-registered, and normalized for proper digital subtraction.

Another important pitfall of the subtraction technique is decreased or absent thyroid uptake of pertechnetate or iodine, which may render the subtraction image invalid and result in a false-negative study (Fig. 4.9). Obtaining a detailed history, including recent iodinated contrast administration, and performing the thyroid image before MIBI injection can reduce the likelihood of this occurring. In our institution, because pertechnetate is administered after MIBI, we review the dynamic imaging after pertechnetate injection to confirm uptake by the thyroid gland.

Very intense MIBI uptake by a parathyroid lesion occasionally may “shine through” on the thyroid image and be interpreted erroneously as a “thyroid nodule.” Sometimes a parathyroid lesion is situated immediately behind a thyroid lesion and is eliminated with the subtraction technique (Fig. 4.10). Occasionally thyroid lesions concentrate MIBI, but not pertechnetate or iodine (Fig. 4.11).1 Subtraction images are most useful when reviewed together with early and late images, rather than in isolation.19

FIGURE 4.10. False-negative subtraction study in a patient with a multinodular goiter. The double phase study demonstrates a focus of increased MIBI activity at the upper pole of the left thyroid lobe and another at the midpole of the right thyroid lobe. The pertechnetate thyroid and subtraction images suggest, incorrectly, that both foci are thyroid lesions. At surgery, however, a 910-mg left superior parathyroid adenoma immediately posterior to a thyroid adenoma was removed.

FIGURE 4.11. A large MIBI-avid lesion at the lower pole of the left thyroid lobe is seen on both the early and late MIBI images. This lesion does not concentrate pertechnetate resulting in a false-positive subtraction image. Note the small focus of activity along the lower pole of the right thyroid lobe (arrows ) seen only on the late MIBI and subtraction images. At surgery a 670-mg hypercellular right inferior parathyroid gland was removed.

Single Photon Emission Computed Tomography. Single photon emission computed tomography (SPECT) frequently is incorporated into MIBI parathyroid imaging protocols. Tomography improves contrast and provides a 3D visualization, increasing diagnostic confidence and assisting in more precise lesion localization, especially in the case of ectopic parathyroid lesions. This in turn can affect surgical planning. A variety of acquisition protocols have been used; some incorporate SPECT and planar imaging whereas others use SPECT alone. In some protocols, early SPECT has been performed and in others late SPECT was used. Dual isotope SPECT and pinhole SPECT also have been investigated.

Billotey et al.30 compared early SPECT to planar imaging and factor analysis of dynamic structures (FADS) in patients undergoing preoperative MIBI parathyroid imaging and reported slightly higher sensitivity for SPECT (90.5% versus 86%) than for planar imaging. They also observed, not surprisingly, that SPECT provided superior localization of ectopic parathyroid glands.

Martínez-Rodríguez et al.31 compared early and late planar imaging, plus early SPECT, to early planar imaging plus early SPECT and found that the sensitivity was identical: 93.4%.

Lorberboym et al.32 compared double phase MIBI, MIBI/pertechnetate subtraction, and early SPECT in 52 patients with PHP. SPECT had the highest sensitivity (96%), followed by subtraction (79%) and double phase imaging (60%). In another investigation, Lorberboym et al.33 compared double phase MIBI and MIBI/pertechnetate subtraction to early SPECT in 41 patients with PHP and multinodular goiter. Sensitivities were: SPECT: 95%, subtraction: 68%, double phase MIBI: 61%.

Nichols et al.19 performed early SPECT on 462 patients with 534 parathyroid lesions. In their investigation, SPECT sensitivity (83%) was not significantly different than planar double phase MIBI (84%) or MIBI/pertechnetate subtraction (88%) for all 534 lesions. SPECT was significantly less sensitive ( p < 0.05) for MGD lesions (59%) than for SGD lesions (90%).

Chen et al.27 compared double phase MIBI, MIBI/pertechnetate subtraction, and late SPECT imaging in 55 patients with hyperparathyroidism. The sensitivity of visual comparison of early images and pertechnetate was 72% to 75%; for late images and pertechnetate images it was 73% to 78%, and for double phase (early and late) MIBI images, sensitivity was 62% to 65%. Sensitivity of computer subtraction of pertechnetate from early images was 71% to 74%; sensitivity of SPECT was 79%. The authors concluded that SPECT may not be necessary for preoperative parathyroid lesion detection since it provides only marginally increased sensitivity compared with visual comparison of early MIBI images with pertechnetate images.

Jorna et al.34 reported that late SPECT (90 minutes post injection) increased diagnostic confidence and changed surgical strategy in 21% of the 64 patients studied. Civelek et al.35 prospectively studied 338 patients with PHP and reported a sensitivity of 87%, specificity of 94%, and positive predictive value of 86% for delayed (2.5 hours post injection) MIBI SPECT. These authors reported that SPECT precisely localized 82% of the abnormal glands.

Moka et al.36 found that in 92 patients with PHP and parathyroid adenomas, the addition of delayed (2 hours post injection) SPECT to planar MIBI/pertechnetate subtraction imaging increased the sensitivity of the test from 87% to 95%. In addition to providing more precise information about lesion location, SPECT was more sensitive than planar imaging for detecting lesions weighing less than 500 mg.

Perez-Monte et al.37 compared early (15 to 30 minutes post injection) and delayed (2 to 4 hours post injection) MIBI SPECT in 37 patients with PHP. Thirty-four patients had parathyroid adenomas and three had hyperplasia. The sensitivity of early SPECT for detection and localization was 91% (31/34), whereas the sensitivity of delayed SPECT images was 74% (25/34) for detection and 32% (11/34) for localization. Early SPECT was significantly better for localization ( p < 0.001) and detection ( p = 0.03).

Thomas et al.38 investigated 36 patients with hyperparathyroidism. All patients underwent double phase MIBI planar and double phase SPECT imaging. These investigators reported that double phase SPECT was significantly more sensitive than double phase planar imaging for adenomas (79% versus 67%; p < 0.05). Neither planar nor SPECT imaging were sensitive (0% versus 25%; p = ns) for detecting hyperplastic glands.

Neumann et al.18 compared double phase MIBI SPECT and simultaneous early 123I/MIBI subtraction SPECT in 15 patients with PHP. At surgery, 17 parathyroid adenomas were identified. The sensitivity, specificity, and diagnostic accuracy were 88%, 97%, and 94%, respectively, for 123I/MIBI subtraction SPECT and 53%, 86%, and 76%, respectively, for double phase MIBI SPECT. The differences in sensitivity and diagnostic accuracy were statistically significant (p = 0.031 and p = 0.016, respectively).

As should be evident from the preceding summary, there are no well-established criteria regarding either the timing of SPECT imaging in relation to the injection of MIBI, or the number of SPECT acquisitions that should be performed. We perform a single, early, SPECT acquisition approximately 30 to 45 minutes after tracer injection, just after completing early planar imaging. Our rationale is that some parathyroid lesions demonstrate rapid washout and may go undetected if only late tomography is performed. In addition, when SPECT is performed shortly after MIBI injection there is sufficient activity remaining in surrounding structures, such as the thyroid gland, to generate the anatomic information necessary to localize the lesion. This of course is not an issue when performing SPECT/CT.

FIGURE 4.12. Selected coronal images from a MIBI SPECT study (anterior to posterior, from left to right, top to bottom.) There is focally increased MIBI activity in the plane of the thyroid gland (arrow ). which is a thyroid nodule. There is a second focus of increased MIBI activity posterior to the thyroid gland (arrowhead ) which is a parathyroid lesion. SPECT is useful for differentiating thyroid from parathyroid lesions. When located in the axial plane of the thyroid gland, as in this case, the parathyroid glands are almost always posterior to and only rarely within the thyroid gland. They are never anterior to the thyroid gland. (This is the same case as illustrated in Fig. 4.10.)

Regardless of when or how many times it is performed, SPECT is a useful complement to planar imaging. Although it may provide only marginal, if any, increase in sensitivity compared with planar imaging, SPECT provides information, not readily available on planar imaging, about the location of a lesion. This information is valuable for localizing a parathyroid lesion as well as for differentiating a parathyroid from a thyroid lesion. The parathyroid glands usually are located posterior to the thyroid gland, and occasionally a parathyroid lesion lies directly behind a thyroid lesion. It may not be possible, on planar images, to differentiate between the two. By determining the precise location of the lesion SPECT can help differentiate a thyroid lesion from a parathyroid lesion, and can identify the parathyroid lesion located behind a thyroid lesion (Fig. 4.12).

SPECT is also useful for detecting and localizing ectopic parathyroid lesions. Although these lesions may be seen on planar imaging, tomographic images provide more detailed topographic information about the lesion and its relation to other structures (Fig. 4.13).1

FIGURE 4.13. A: There is a well-circumscribed MIBI-avid focus in the midline of the neck, between the two lobes of the thyroid gland. B: On the axial SPECT image this focus (arrow ) lies well posterior to the left thyroid lobe (arrowhead ). At surgery a 2,100-mg retroesophageal left superior parathyroid adenoma was found. Note the linear photopenic defect between the thyroid and parathyroid glands on this image. This finding is invariably present in the setting of a retroesophageal parathyroid gland.

SPECT/CT. Radiotracers primarily reflect function. Only gross anatomic detail can be inferred from the images and the precise anatomic detail necessary to localize radiotracer accumulation often is lacking, even when SPECT is performed. Integrating radionuclide and anatomic images with SPECT/CT improves diagnostic confidence and accuracy. Patel et al.39 compared double phase MIBI planar imaging plus late SPECT/CT to ultrasound in 63 patients with PHP, including 59 with SGD and 4 with MGD. MIBI sensitivity for SGD was 90% versus 64% for ultrasound. Sensitivity of MIBI plus ultrasound was 95%. None of the MGD lesions were detected with either method. There were several cases in which planar MIBI imaging was equivocal but SPECT/CT demonstrated a definite abnormality.

Serra et al.40 studied MIBI double phase planar with SPECT and SPECT/CT in 16 patients with hyperparathyroidism, including 10 with PHP and 6 with SHP. The sensitivity of the double phase technique in PHP was 57% versus 100% for SPECT and SPECT/CT. The sensitivity of the double phase technique in SHP was 43% versus 64% for SPECT and SPECT/CT. Although the sensitivity of SPECT and SPECT/CT were the same, in 39% of the cases SPECT/CT provided additional information concerning the precise location of lesions, simplifying the surgical procedure.

Ciappuccini et al.41 evaluated double phase planar MIBI plus late MIBI SPECT/CT in 54 patients with PHP. The sensitivity of the test was 92%, the specificity was 83%. The authors concluded that double phase MIBI scintigraphy with SPECT/CT has a major impact on parathyroid surgery for patients having parathyroid tumors in expected as well as unexpected locations.

Gayed et al.42 compared MIBI double phase planar plus early SPECT to SPECT/CT in 32 patients with PHP. They reported that the sensitivity of the test was 89% with and without SPECT/CT. SPECT/CT changed the diagnosis in only one patient and better located abnormal glands in only four patients. SPECT/CT was helpful in locating the two ectopic parathyroid adenomas in this investigation. They concluded that SPECT/CT has no significant clinical value beyond that of SPECT except for localizing ectopic parathyroid glands.

Pata et al.43 studied 33 patients with PHP and concomitant nodular thyroid goiter, all of whom underwent double phase MIBI planar imaging. Eighteen patients underwent delayed MIBI SPECT and fifteen underwent delayed MIBI SPECT/CT. The authors compared the ability of the two techniques to correctly localize lesions both to the right or left side of the neck as well to the four quadrants of the neck. Although there were no significant differences in sensitivity and specificity between SPECT and SPECT/CT for lateralizing lesions to one side of the neck or the other, SPECT/CT was significantly more sensitive (87.5% versus 55.6%; p < 0.0001) than SPECT for localizing lesions to the correct quadrant of the neck. The mean operative time was shorter for patients who underwent SPECT/CT than for those who underwent SPECT (38 versus 56 minutes; p = 0.034). Though the results are not surprising, it should be noted that two different populations were compared and that the role of planar imaging in study interpretation was not provided.

Krausz et al.44 evaluated the contribution of early SPECT/CT in 36 patients with PHP who underwent double phase MIBI planar imaging, some of whom also underwent pertechnetate thyroid subtraction imaging. The overall sensitivity of MIBI imaging was 92% (33/36). Three patients, two with MGD and one with SGD, were completely negative on MIBI imaging. Among the 33 patients with positive studies, 23 parathyroid lesions were in the neck and 10 were in the lower neck/mediastinum. SPECT/CT contributed to the localization of parathyroid lesions in patients with PHP and to planning the surgical exploration in 14 of 36 (39%) patients, predominantly those with ectopic lesions or who had distorted neck anatomy.

Oksuz et al.45 evaluated MIBI imaging in 60 patients with PHP. All 60 patients underwent planar imaging, including thyroid subtraction; 35 also underwent SPECT/CT. The sensitivity of planar imaging was 76%, whereas sensitivity of SPECT and SPECT/CT both were 95%. The authors reported that SPECT/CT provided superior topographic information, especially in patients with ectopic lesions, facilitating the surgical approach by providing preoperative anatomic mapping. The authors found that the increased sensitivity of SPECT and SPECT/CT compared with planar scintigraphy can be attributed to improved detection of small lesions related to superior contrast resolution, depth information, and accurate 3D localization.

Lavely et al.46 retrospectively studied 98 patients with SGD, all of whom underwent early and late planar MIBI imaging and SPECT/CT. Six image sets (early and delayed planar imaging, SPECT, and SPECT/CT) and combinations of the two image sets were reviewed for lesion localization at 13 possible sites by two reviewers in two reviewer groups. Sensitivity, specificity, area under the curve, positive predictive and negative predictive values, as well as κ inter-rater agreement values, were determined for each method. The highest values were for double phase studies that included SPECT/CT (0.76 to 0.79). Double phase planar imaging, SPECT, and SPECT/CT were statistically significantly better than single phase early or delayed imaging in sensitivity, area under the curve, and positive predictive value. Neither single phase nor double phase SPECT was statistically superior to double phase planar imaging. Early phase SPECT/CT together with any delayed imaging method was superior to double phase planar imaging or SPECT in sensitivity, area under the curve, and positive predictive value. Double phase acquisition was more accurate than single phase MIBI scintigraphy for planar imaging, SPECT, and SPECT/CT. The authors noted that a major advantage of SPECT/CT was its ability to differentiate inferior glands from ectopic superior glands in the tracheoesophageal groove.

As with SPECT, it is not clear that SPECT/CT improves the sensitivity of MIBI parathyroid imaging. SPECT/CT does however facilitate the differentiation of parathyroid from thyroid lesions and affords more precise lesion localization than does SPECT, especially in the case of ectopic lesions (Figs. 4.14 and 4.15). We have found MIBI SPECT/CT particularly valuable for localizing mediastinal parathyroid lesions. In fact, the anatomic detail provided by this technique is sufficiently precise that, at our institution, diagnostic CTs no longer are performed to localize scinitigraphically detected mediastinal lesions (Fig. 4.16).

Limitations of MIBI Parathyroid Imaging

False-Positive Results. The most frequent cause of false-positive MIBI parathyroid imaging results is the solid thyroid nodule, either solitary or in a multinodular gland (Table 4.2). False-positive results are associated with various benign and malignant tumors as well as with lymph node disease, both benign, such as inflammation and sarcoidosis, and malignant including lymphoma, and metastatic disease (Fig. 4.17). MIBI uptake in brown tumors of hyperparathyroidism has been reported.1

TABLE 4.2

FACTORS ADVERSELY AFFECTING SPECIFICITY OF MIBI PARATHYROID IMAGING

False-Negative Results. The failure of MIBI imaging to identify some parathyroid lesions is likely caused by several factors. Certainly, lesion size is important, as it relates to both the system resolution and to the amount of tracer uptake by the parathyroid tissue. Nichols et al.19 found that, regardless of the imaging protocol used, MIBI was significantly more sensitive for heavier (larger) than for lighter (smaller) lesions. Parathyroid lesions, even when markedly enlarged, are relatively small structures, and may go unrecognized. Using a pinhole collimator, rather than a parallel hole collimator, can reduce false-negative planar MIBI imaging. The pinhole collimator, which provides the highest resolution of all the collimators used in nuclear medicine, magnifies the structures being imaged. Its conical shape fits well in the neck, which is recessed in between the head and the chest when the patient is supine, making the pinhole collimator ideally suited for imaging the small structures in this region. Studies have shown that the use of pinhole collimation significantly improves the sensitivity of the test.47,48 Tomas et al.48 reported that planar imaging performed with a pinhole collimator is significantly ( p = 0.0003) more sensitive (89%; 48/54) than planar imaging performed with a parallel hole collimator (56%; 30/54) (Fig. 4.18). In their investigation, 18 lesions were detected only on pinhole images. All lesions detected on parallel hole collimator images were identified on pinhole collimator images. There were no lesions seen only on parallel hole collimator images. Furthermore, there was no significant difference ( p = 0.29) in specificity between pinhole collimator (93%) and parallel hole collimator (96%) images.

FIGURE 4.14. A: The pinhole images demonstrate an MIBI-avid focus at the lower pole of the left thyroid lobe. B: On the SPECT/CT this focus is in the retroesophageal space (arrows ). A 2,780-mg retroesophageal left superior parathyroid gland was found at surgery.

FIGURE 4.15. A: 1,650-mg retropharyngeal right superior parathyroid adenoma. There is a MIBI-avid focus in the midline of the neck above the multinodular thyroid gland (arrows ). More precise localization of this focus is not possible on these images. B: Although the SPECT images (top row) indicate that the focus is in the same coronal plane as the thyroid gland (arrows ). the retropharyngeal location of the lesion (arrows ) is evident only on the fused SPECT/CT images (bottom row).

The sensitivity of MIBI parathyroid imaging also may be affected because of the cellular function. There are data that indicate that MIBI uptake is related to parathyroid cellular function. Sun et al.49 reported that parathyroid tissue that expresses P-glycoprotein (PgP) does not accumulate MIBI. They observed that normal parathyroid glands, as well as some parathyroid adenomas, express PgP. They found that large parathyroid adenomas that express either PgP or the multidrug resistance-related protein, MRP, did not accumulate MIBI whereas adenomas lacking both proteins accumulated the tracer (Table 4.3).

TABLE 4.3

FACTORS ADVERSELY AFFECTING SENSITIVITY OF MIBI PARATHYROID IMAGING

MIBI is less sensitive for detecting hyperplastic parathyroid glands than for detecting adenomatous ones and it has been suggested that this is because hyperplastic glands usually are smaller than adenomatous ones. Recent investigations however indicate that, at least in SHP, MIBI uptake is more closely related to cell cycle than to gland size. Torregrosa et al.50 reported that higher MIBI uptake correlated with the active growth phase of the cells.

The presence of mitochondria-rich oxyphil cells presumably accounts for MIBI uptake in parathyroid tissue, and glands with fewer oxyphil cells, and hence fewer mitochondria, may explain both lower uptake and rapid washout of MIBI from some parathyroid lesions.15,16,51

FIGURE 4.16. Mediastinal parathyroid lesion. Although SPECT localizes the ectopic parathyroid lesion to the anterior mediastinum, SPECT/CT localizes it precisely to the thymus. At surgery a 1,700-mg hypercellular parathyroid gland, attached to the anterior surface of the thymus, was resected. If only SPECT had been performed, the patient would have undergone a separate diagnostic CT scan.

Although most cases of PHP are caused by SGD, up to 20% are caused by MGD, which is one of the more intriguing causes of false-negative results in patients with PHP.52 Milas et al.53 reported that MIBI imaging identified two sites of disease in only 5 of 84 patients (6% sensitivity) with hyperparathyroidism and double adenomas. Chiu et al.54 reviewed the results of MIBI imaging in 401 patients with PHP and reported that whereas the test was 76% sensitive for detecting SGD lesions, among 38 patients with MGD, the test identified more than two lesions in 7 patients, one lesion in 13 patients, and no lesions in the remaining 18 patients. Nichols et al.19 in a retrospective investigation of 409 patients, including 53 with MGD, found that MIBI imaging was significantly less sensitive for detecting MGD lesions than for SGD lesions. This was true regardless of how images were interpreted: Double phase MIBI alone, MIBI/pertechnetate subtraction alone, SPECT alone, and all images together. Even if all lesions in a patient with MGD are not detected, it would be useful for surgical planning if patients with MGD could be identified preoperatively. In that same investigation, however, MIBI was no more sensitive for identifying patients with MGD than for identifying the lesions themselves.

In a subsequent investigation, Nichols et al.55 studied 651 patients with PHP (851 lesions), including 520 with SGD and 131 with MGD (331 lesions). They again found that MIBI was significantly less sensitive (61% versus 97%; p < 0.0001) for MGD than for SGD. They also observed that the sensitivity of the test decreased as the number of lesions increased.

The explanation for the lower sensitivity in MGD is not clear. Lesion location is one possible explanation. The spatial distribution of MGD lesions in the study of Nichols et al.55 was statistically similar to that of SGD lesions and consequently lesion location does not explain the lower sensitivity. Several investigations have found that, in general, MIBI is less sensitive for lesions of lower weight and there are data indicating that MGD lesions are smaller than those of SGD.51,5456 To test the theory that decreased sensitivity in MGD is related to lesion weight, Nichols et al.55 analyzed data for 249 MGD lesions in 111 patients whose lesions were matched by weight, lesion by lesion, to 249 SGD lesions (median weight 260 mg [48 to 13,800 mg] versus median weight 240 mg [48 to 12,600 mg]; Wilcoxon p = 0.68). Despite similar weights for paired SGD and MGD lesions, sensitivity was consistently and significantly lower for MGD lesions (65% versus 94%; p < 0.0001) independent of lesion weight. Thus decreased MIBI sensitivity in MGD does not appear to be related to lesion weight.

Another explanation offered for lower MIBI sensitivity in MGD is related to histology. MGD is usually caused by hyperplasia whereas SGD is usually caused by an adenoma, and MIBI imaging is less sensitive for detecting hyperplastic than adenomatous glands.57 Differentiating hyperplasia from adenoma is not always clear cut and depends on subtle, often ill-defined morphologic criteria. It may be necessary to remove, or at least inspect, other parathyroid glands at the time of surgery. When only one gland is removed, and other glands are not even examined, an increasingly common situation in minimally invasive surgery, it may not be possible to differentiate between the two.58 The problem of establishing unambiguous histologic criteria to diagnose parathyroid disease has resulted in wide variations in the incidence with which relative proportions have been reported of adenomas, hyperplasias, and carcinomas among tissue samples.5962 Histology alone is no longer sufficient to perform this differentiation with a high degree of confidence.58 Finally, it is important to note that MIBI imaging, in addition to being less sensitive for detecting hyperplasic glands, is also significantly less sensitive for detecting double adenomas in MGD.53 This makes the argument that lower sensitivity in MGD is caused by differences in histology even more tenuous.

FIGURE 4.17. There is focally increased MIBI uptake at the level of the thymus in a patient undergoing preoperative parathyroid localization. Although this was interpreted as a mediastinal parathyroid lesion, a benign thymoma was found at surgery. It is important to be cognizant of the fact that MIBI is not specific for parathyroid glands and accumulates in a variety of benign and malignant conditions.

FIGURE 4.18. A: Eutopic 210-mg right inferior parathyroid lesion. On the late pinhole image, there is a small MIBI-avid focus at the lower pole of the right thyroid lobe (arrow ). B: The parallel hole images fail to demonstrate any abnormality.

The explanation for the decreased sensitivity of MIBI in MGD remains elusive. It does not appear to be related to lesion location or weight. Given the limitations inherent in histologic analysis, further studies, at the molecular level, of factors affecting MIBI sensitivity in MGD are warranted.

99mTc-Tetrofosmin

99mTc-tetrofosmin is a lipophilic cationic diphosphine, which like MIBI, was developed for myocardial perfusion imaging. Unlike the intracellular retention of MIBI, which depends primarily on the mitochondrial membrane potential, the intracellular retention of tetrofosmin is dependent on the cell membrane potential.63 Several investigations suggest that tetrofosmin is useful for parathyroid lesion localization.

Hiromatsu et al.64 studied 20 patients with PHP, using double phase tetrofosmin imaging and reported that 19/20 lesions (95% sensitivity) were correctly localized. Tetrofosmin was somewhat more sensitive than ultrasound CT and MRI, all with 85% sensitivity, in this investigation.

Wu et al.65 performed double phase tetrofosmin imaging in 40 patients with PHP, including 20 with lesions weighing more than 1.5 g and 20 with lesions weighing between 500 mg and 1,500 mg. Sensitivities were similar: 85% (17/20) for the larger lesions and 80% (16/20) for smaller lesions. They correlated their findings with PgP expression and found that all positive results were PgP expression negative and all false-negative results were PgP expression positive.

Vallejos et al.66 studied 31 patients, 19 with PHP and 12 with SHP, with double phase tetrofosmin imaging. Some patients also underwent pertechnetate thyroid imaging. The sensitivity of the test for SGD was 95% and the sensitivity for MGD was 58%. All abnormal glands were identified on early images. Some lesions were better seen on early images and the authors concluded that, when using tetrofosmin, early imaging might be sufficient.

Ishibashi et al.67 studied double phase tetrofosmin imaging in 26 patients with hyperparathyroidism, including 7 with PHP and 19 with SHP. Tetrofosmin correctly localized all 7 lesions (100% sensitivity) in patients with PHP and 27/37 lesions in patients with SHP (73%).

Gallowitsch et al.68 studied double phase tetrofosmin imaging plus SPECT in 48 patients, including 35 with PHP and 13 with SHP. Among the patients with PHP double phase imaging detected 25 of 36 (69.2% sensitivity) lesions, whereas SPECT detected 34 lesions (94.4% sensitivity). The sensitivity of planar imaging in SHP was 38% whereas the sensitivity of SPECT was 61.5%.

The results of intraindividual comparisons of tetrofosmin and MIBI imaging have been mixed. Froberg et al.69 compared tetrofosmin and MIBI parathyroid imaging using a double phase technique. In their investigation, all nine lesions were correctly identified with MIBI, whereas only two were detected with tetrofosmin. There was a significant increase in lesion to thyroid ratio over time with MIBI, but not with tetrofosmin, and the authors concluded that tetrofosmin is not suitable for double phase parathyroid scintigraphy.

Fjeld et al.70 compared double phase tetrofosmin and MIBI imaging in 16 patients with PHP. Using focally increased activity on early or delayed images as the criterion for a positive study, the sensitivity of both tests was the same: 75%. The authors noted that in six cases, lesion visualization on MIBI images improved on delayed imaging, whereas none of the lesions showed improved visualization over time with tetrofosmin (Fig. 4.19).

Kibar et al.71 compared tetrofosmin with MIBI in 16 patients with PHP. Only patients who had focal abnormalities on MIBI imaging underwent tetrofosmin imaging. The imaging protocol and interpretation was identical for both tracers. Patients underwent double phase imaging at 15 to 20 minutes and again at 2 to 3 hours after injection. A focal area of increased uptake that persisted or increased in intensity from early to late images was considered positive. All 16 lesions were correctly identified with both tracers, although image quality was judged to be slightly better with MIBI.

FIGURE 4.19. Early and late tetrofosmin (UPPER) and early and late MIBI image (LOWER) studies performed on a patient with a 500-mg parathyroid lesion at the lower pole of the left thyroid lobe. On the tetrofosmin study the lesion is seen only on the early image, whereas on the MIBI study the lesion is seen on both the early and late images. Reproduced with permission from Palestro CJ, Tomas MB, Tronco GG. Radionuclide imaging of the parathyroid glands. Sem Nucl Med. 2005;35:266–276.

Wakamatsu et al.72 compared tetrofosmin/pertechnetate and MIBI/pertechnetate subtraction imaging in 28 patients with PHP, including 25 with SGD and 3 with MGD and found no significant differences in sensitivity between the two agents: 63.2% for tetrofosmin versus 68.4% for MIBI. Both agents were less sensitive (47.7% for both) for MGD than for SGD.

Aigner et al.73 compared late tetrofosmin and MIBI imaging in 10 patients with PHP, and reported that, although MIBI image contrast was somewhat better, all 10 lesions were detected with both tracers. Tetrofosmin showed a slower elimination from the parathyroid adenomas than MIBI in 6 of the 10 cases.

Although it is easier to prepare, offers a somewhat lower radiation exposure to the patient, and probably has comparable sensitivity when performed and interpreted appropriately, tetrofosmin has never enjoyed the popularity of MIBI for parathyroid imaging.

PET Tracers

A variety of positron-emitting tracers have been used to localize parathyroid lesions, including 18F-fluorodopa, 11C-methionine, and 18F-fluorodeoxyglucose.

11C-Methionine (MET PET)

Beggs and Hain74 performed MET PET on 51 patients with hyperparathyroidism. Final diagnoses were based on surgery or clinical follow-up. Sensitivity of the test was 83% for adenomas. None of the hyperplastic parathyroid glands were detected.

Otto et al.75 compared MET PET and MIBI SPECT in 30 patients with hyperparathyroidism. Sixteen patients had PHP, twelve patients had SHP, and two patients had parathyroid carcinoma. Imaging was performed 10 and 40 minutes after injection. Sensitivity of MET PET was 94% for SGD (17/18) and 69% (25/36) for MGD. Sensitivity of MIBI was 50% (9/18) for SGD and 47% (17/36) for MGD. The highest SUVparathyroid/SUVcervical soft tissue ratio was at 10 minutes, and the highest SUVparathyroid/SUVthyroid ratio was at 40 minutes post injection. In three patients lesions were clearly identified only on the 40-minute post-injection images.

Sundin et al.76 performed MET PET on 34 patients with 37 parathyroid lesions, including 32 with PHP and 2 with SHP. MET PET sensitivity, on a per lesion basis, was 81% (31/37). There were no significant differences in sensitivity based on lesion type.

Weber et al.77 performed MET PET/CT on 33 patients with PHP, including 30 with SGD and 3 with MGD. Sensitivity of the test was 83% (25/30) for SGD and 67% for MGD. There was a significant correlation between true-positive results and lesion weight (2.42 ± 4.05 g) and diameter (2 ± 1.18 cm); false-negative lesions were significantly smaller (0.98 ± 0.54 cm) and less massive (0.5 ± 0.38 g) than true-positive ones.

Tang et al.78 compared MET PET/CT with MIBI SPECT in 30 patients with 46 parathyroid lesions: 24 adenomatous and 22 hyperplastic glands. For adenomatous glands the sensitivity of MET PET/CT was 92% versus 95% for MIBI SPECT. For hyperplastic glands the sensitivity of MET PET/CT was 68% versus 59% for MIBI SPECT.

Caldarella et al.79 performed a meta-analysis on the diagnostic performance of MET PET imaging of parathyroid adenomas. Using patient-based sensitivity, rather than lesion-based sensitivity, they reported that the sensitivity ranged from 43% to 95% with a pooled estimate of 81% (95% CI: 74% to 86%) and concluded that the diagnostic performance of MET PET is comparable to that of MIBI.

18F-Fluorodeoxyglucose (FDG PET)

Melon et al.80 performed FDG PET on seven patients with PHP and reported that only 2/9 parathyroid lesions were identified. Neumann et al.81 studied 17 patients with PHP and 22 parathyroid lesions, including 14 with SGD and 3 with MGD. FDG PET correctly localized 13/14 SGD lesions (93% sensitivity) and 6/8 MGD lesions (75% sensitivity). There were three false-positive results including two follicular thyroid adenomas; the etiology of the third was unknown.

Neumann et al.82 subsequently compared FDG PET and double phase MIBI SPECT in 21 patients with PHP and found that FDG PET was significantly more sensitive than MIBI SPECT (86% versus 43%; p< 0.001). There was no significant difference in lesion weight between true-positive and false-negative FDG PET studies, but false-negative lesion weights were significantly lower than those of true-positive lesions ( p = 0.001) for MIBI SPECT. MIBI SPECT, however, was more specific than FDG PET (90% versus 78%; p = 0.063).

SUMMARY

MIBI currently is the radionuclide procedure of choice for preoperative parathyroid lesion localization. Although reasonably good results can be obtained with planar imaging, the use of pinhole collimation, thyroid subtraction imaging, and SPECT or SPECT/CT improves the accuracy of the test. Because the parathyroid glands can be located anywhere between the mandibular angle and the base of the heart, this entire region must be included in the field of view, regardless of the protocol used. MIBI is more sensitive for detecting lesions in SGD than lesions in MGD and the results of the test must be used together with rapid intraoperative PTH assay to ensure that all offending lesions have been removed.

REFERENCES

1. Palestro CJ, Tomas MB, Tronco GG. Radionuclide imaging of the parathyroid glands. Semin Nucl Med. 2005;35:266–276.

2. Garvie NW. Imaging the parathyroids. In: Peters AM, ed. Nuclear Medicine in Radiologic Diagnosis. London: Martin Dunitz; 2003:681–694.

3. Marcocci C, Cetani F. Primary hyperparathyroidism. N Engl J Med. 2011;365: 2389–2397.

4. Mihai R, Farndon JR. Parathyroid disease and calcium metabolism. Br J Anaesth. 2000;85:29–43.

5. Timio M. Cardiotoxiciity of parathyroid hormone in chronic renal failure. Ital J Mineral Electrol Metab. 1995;9:119–124.

6. Krusback AJ, Wilson SD, Lawson TL, et al. Prospective comparison of radionuclide, computed tomographic, sonographic, and magnetic resonance localization of parathyroid tumors. Surgery. 1989;106:639–646.

7. Clark OH, Okerlund MD, Moss AA. Localization studies in patients with persistent or recurrent hyperparathyroidism. Surgery. 1985;98:1083–1094.

8. Rodgers SE, Hunter GJ, Hamberg LM, et al. Improved preoperative planning for directed parathyroidectomy with 4-dimensional computed tomography. Surgery. 2006;140:932–941.

9. Stucken EZ, Kutler DI, Moquete R, et al. Localization of small parathyroid adenomas using modified 4-dimensional computed tomography/ultrasound. Otolaryngol Head Neck Surg. 2012;146:33–39.

10. Basso LV, Keeling C, Goris ML. Parathyroid imaging: Use of dual isotope scintigraphy for the localization of adenoma before surgery. Clin Nucl Med. 1992; 17:380–383.

11. Sandrock D, Merino MJ, Norton JA, et al. Parathyroid imaging by Tc/Tl scintigraphy. Eur J Nucl Med. 1990;16:607–613.

12. Hauty M, Swartz K, McClung M, et al. Technetium-thallium scintiscanning for localization of parathyroid adenoma and hyperplasia. A reappraisal. Am J Surg. 1987;153:479–486.

13. Coakley AJ, Kettle AG, Wells CP, et al. 99mTc-sestamibi—a new agent for parathyroid imaging. Nucl Med Commun. 1989;10:791–794.

14. Chiu ML, Kronauge JF, Piwnica WD. Effect of mitochondrial and plasma membrane potential on accumulation of hexakis (2-methoxyisobutylisonitrile) technetium(I) in cultured mouse fibroblasts. J Nucl Med. 1990;31:1646–1653.

15. Hetrakul N, Civelek AC, Stagg CA, et al. In vitro accumulation of technetium-99m-sestamibi in human parathyroid mitochondria. Surgery. 2001;130:1011–1018.

16. Melloul M, Paz A, Koren R, et al. 99m Tc-MIBI scintigraphy of parathyroid adenomas and its relation to tumour size and oxyphil cell abundance. Eur J Nucl Med. 2001;28:209–213.

17. Taillefer R, Boucher Y, Potvin C, et al. Detection and localization of parathyroid adenomas in patients with hyperparathyroidism using a single radionuclide imaging procedure with technetium-99m-sestamibi (double-phase study). J Nucl Med. 1992;33:1801–1807.

18. Neumann DR, Esselstyn CB Jr, Go RT, et al. Comparison of double phase 99mTc-sestamibi with 123I-99mTc-sestamibi subtraction SPECT in hyperparathyroidism. AJR Am J Roentgenol. 1997;169:1671–1674.

19. Nichols KJ, Tomas MB, Tronco GG, et al. Optimizing the accuracy of preoperative parathyroid lesion localization. Radiology. 2008;248:221–232.

20. Wei JP, Burke GJ, Mansberger AR Jr. Prospective evaluation of the efficacy of technetium 99m sestamibi and iodine 123 radionuclide imaging of abnormal parathyroid glands. Surgery. 1992;112:1111–1116.

21. Casas AT, Burke GJ, Sathyanarayana, et al. Prospective comparison of technetium-99m-sestamibi/iodine-123 radionuclide scan versus high-resolution ultrasonography for the preoperative localization of abnormal parathyroid glands in patients with previously unoperated primary hyperparathyroidism. Am J Surg. 1993;166:369–373.

22. Wakamatsu H, Noguchi S, Yamashita H, et al. Parathyroid scintigraphy with 99mTc-MIBI and 123I subtraction: A comparison with magnetic resonance imaging and ultrasonography. Nucl Med Commun. 2003;24:755–762.

23. Chen CC, Skamulis MC, Fraker DL, et al. Technetium-99m-sestamibi imaging before reoperation for primary hyperparathyroidism. J Nucl Med. 1995;3:2186–2191.

24. Hindie E, Melliere D, Jeanguillaume C, et al. Parathyroid imaging using simultaneous double-window recording of technetium-99m-sestamibi and iodine-123. J Nucl Med. 1998;39:1100–1105.

25. Caveny SA, Klingensmith WC 3rd, Martin WE, et al. Parathyroid imaging: The importance of dual-radiopharmaceutical simultaneous acquisition with 99mTc-sestamibi and 123I. J Nucl Med Technol. 2012;40:104–110.

26. Wei JP, Burke GJ, Mansberger AR Jr. Preoperative imaging of abnormal parathyroid glands in patients with hyperparathyroid disease using combination Tc-99m-pertechnetate and Tc-99m-sestamibi radionuclide scans. Ann Surg. 1994;219:568–573.

27. Chen CC, Holder LE, Scovili WA, et al. Comparison of parathyroid imaging with technetium-99m-pertechnetate/sestamibi subtraction, double-phase technetium-99m-sestamibi and technetium-99m-sestamibi SPECT. J Nucl Med. 1997;38: 834–839.

28. Leslie WD, Dupont JO, Bybel1 B, et al. Parathyroid 99mTc-sestamibi scintigraphy: Dual-tracer subtraction is superior to double-phase washout. Eur J Nucl Med. 2002;29:1566–1570.

29. Rubello D, Saladini G, Casara D, et al. Parathyroid imaging with pertechnetate plus perchlorate/MIBI subtraction scintigraphy: A fast and effective technique. Clin Nucl Med. 2000;25:527–531.

30. Billotey C, Sarfati E, Aurengo A, et al. Advantages of SPECT in technetium-99m sestamibi parathyroid scintigraphy. J Nucl Med. 1996;37:1773–1778.

31. Martínez-Rodríguez I, Banzo I, Quirce R, et al. Early planar and early SPECT Tc-99m sestamibi imaging can it replace the dual-phase technique for the localization of parathyroid adenomas by omitting the delayed phase? Clin Nucl Med. 2011;36:749–753.

32. Lorberboym M, Minski I, Macadziob S, et al. Incremental diagnostic value of preoperative 99mTc-MIBI SPECT in patients with a parathyroid adenoma. J Nucl Med. 2003;44:904–908.

33. Lorberboym M, Ezri T, Schachter PP. Preoperative technetium Tc 99m sestamibi SPECT imaging in the management of primary hyperparathyroidism in patients with concomitant multinodular goiter. Arch Surg. 2005;140:656–660.

34. Jorna FH, Jager PL, Que Th, et al. Value of 123I-subtraction and single-photon emission computed tomography in addition to planar 99mTc-MIBI scintigraphy before parathyroid surgery. Surg Today. 2007;37:1033–1041.

35. Civelek AC, Ozalp E, Donovan P, et al. Prospective evaluation of delayed technetium-99m sestamibi SPECT scintigraphy for preoperative localization of primary hyperparathyroidism. Surgery. 2002;131:149–157.

36. Moka D, Voth E, Dietlein M, et al. Technetium 99m-MIBI-SPECT: A highly sensitive diagnostic tool for localization of parathyroid adenomas. Surgery. 2000; 128:29–35.

37. Perez-Monte JE, Brown ML, Shah AN, et al. Parathyroid adenomas: Accurate detection and localization with Tc-99m sestamibi SPECT. Radiology. 1996;201: 85–91.

38. Thomas DL, Bartel T, Menda Y, et al. Single photon emission computed tomography (SPECT) should be routinely performed for the detection of parathyroid abnormalities utilizing technetium-99m sestamibi parathyroid scintigraphy. Clin Nucl Med. 2009;34:651–655.

39. Patel CN, Salahudeen HM, Lansdown M, et al. Clinical utility of ultrasound and 99mTc sestamibi SPECT/CT for preoperative localization of parathyroid adenoma in patients with primary hyperparathyroidism. Clin Radiol. 2010;65:278–287.

40. Serra A, Bolasco P, Satta L, et al. Role of SPECT/CT in the preoperative assessment of hyperparathyroid patients. Radiol Med. 2006;111:999–1008.

41. Ciappuccini R, Morera J, Pascal P, et al. Dual-phase 99mTc sestamibi scintigraphy with neck and thorax SPECT/CT in primary hyperparathyroidism. A single-institution experience. Clin Nucl Med. 2012;37:223–228.

42. Gayed, IW, Kim EE, Broussard WF, et al. The value of 99mTc-sestamibi SPECT/CT over conventional SPECT in the evaluation of parathyroid adenomas or hyperplasia. J Nucl Med. 2005;46:248–252.

43. Pata G, Casella C, Besuzio S, et al. Clinical appraisal of 99m technetium-sestamibi SPECT/CT compared to conventional SPECT in patients with primary hyperparathyroidism and concomitant nodular goiter. Thyroid. 2010;20:1121–1127.

44. Krausz Y, Bettman L, Guralnik L, et al. Technetium-99m-MIBI SPECT/CT in primary hyperparathyroidism. World J Surg. 2006;30:76–83.

45. Oksuz MO, Dittmann H, Wicke C, et al. Accuracy of parathyroid imaging: A comparison of planar scintigraphy, SPECT, SPECT-CT, and C-11 methionine PET for the detection of parathyroid adenomas and glandular hyperplasia. Diagn Interv Radiol. 2011;17:297–307.

46. Lavely WC, Goetze S, Friedman KP, et al. Comparison of SPECT/CT, SPECT, and planar imaging with single- and dual-phase 99mTc-sestamibi parathyroid scintigraphy. J Nucl Med. 2007;48:1084–1089.

47. Kipper MS, LaBarbera JJ, Krohn LD, et al. Localization of a parathyroid adenoma by the addition of pinhole imaging to Tc-99m sestamibi dual-phase scintigraphy. Report of a case and review of experience. Clin Nucl Med. 1992;22:73–75.

48. Tomas MB, Pugliese PV, Tronco GG, et al. Pinhole versus parallel-hole collimators for parathyroid imaging: An intraindividual comparison. J Nucl Med Technol. 2008;36:189–194.

49. Sun SS, Shiau YC, Lin CC, et al. Correlation between P-glycoprotein (P-gp) expression in parathyroid and Tc-99m MIBI parathyroid image findings. Nucl Med Biol. 2001;28:929–933.

50. Torregrosa JV, Fernandez-Cruz L, Canalupo A, et al. 99mTc-sestamibi scintigraphy and cell cycle in parathyroid glands of secondary hyperparathyroidism. World J Surg. 2000;24:1386–1390.

51. Bhatnagar A, Vezza PR, Bryan JA, et al. Technetium-99m-sestamibi parathyroid scintigraphy: Effect of P-glycoprotein, histology and tumor size on detectability. J Nucl Med. 1998;39:1617–1620.

52. Bergson EJ, Snyter LA, Dubner S, et al. Sestamibi scans and intra-operative PTH measurement in the treatment of primary hyperparathyroidism due to multiple gland disease. Arch Otolaryngol Head Neck Surg. 2004;130:87–91.

53. Milas M, Wagner K, Easley KA, et al. Double adenomas revisited: Nonuniform distribution favors enlarged superior parathyroids (fourth pouch disease). Surgery. 2003;134:995–1004.

54. Chiu B, Sturgeon C, Angelos P. What is the link between nonlocalizing sestamibi scans, multigland disease, and persistent hypercalcemia? A study of 401 consecutive patients undergoing parathyroidectomy. Surgery. 2006;140:418–422.

55. Nichols KJ, Tomas MB, Tronco GG, et al. Sestamibi parathyroid scintigraphy in multigland disease. Nucl Med Commun. 2012;33:43–50.

56. Berber E, Parikh RT, Ballem N, et al. Factors contributing to negative parathyroid localization: An analysis of 1000 patients. Surgery. 2008;144:74–79.

57. O’Doherty MJ, Kettle AG, Wells P, et al. Parathyroid imaging with technetium-99m-sestamibi: Preoperative localization and tissue uptake studies. J Nucl Med. 1992;33:313–318.

58. Johnson SJ. Changing clinicopathological practice in parathyroid disease. Histopathology. 2010;56:835–851.

59. Jennings T, DeLellis R, Lloyd R, et al. Recommendations for the reporting of parathyroid specimens. The Association of Directors of Anatomic and Surgical Pathology. 2004. www.adasp.org.

60. DeLellis RA. Tumors of the Parathyroid Gland. Fascicle 6, Third Series, Atlas of Tumor Pathology. Washington, DC: Armed Forces Institute of Pathology; 1993.

61. Bondeson AG, Bondeson L, Ljungberg O, et al. Fat staining in parathyroid disease-diagnostic value and impact on surgical strategy: Clinicopathologic analysis of 191 caseses. Hum Pathol. 1985;16:1255–1263.

62. Ghandur-Mnaymneh L, Kimura N. The parathyroid adenoma: A histopathologic definition with a study of 172 cases of primary hyperparathyroidism. Am J Pathol. 1984;115:70–83.

63. Arbab AS, Koizumi K, Arki T. Uptake of technetium-99m tetrofosmin, technetium-99m MIBI and thallium 201 in tumor cell lines. J Nucl Med. 1996;37: 551–556.

64. Hiromatsu Y, Ishibashi M, Nishida H, et al. Technetium-99m tetrofosmin parathyroid imaging in patients with primary hyperparathyroidism. Intern Med. 2000;39:101–106.

65. Wu HS, Liu YC, Kao A, et al. Using technetium 99m tetrofosmin parathyroid imaging to detect parathyroid adenoma and its relation to P-glycoprotein expression. Surgery. 2002;132:456–460.

66. Vallejos V, Martin-Comin J, Gonzalez MT, et al. The usefulness of Tc-99m tetrofosmin scintigraphy in the diagnosis and localization of hyperfunctioning parathyroid glands. Clin Nucl Med. 1999;24:959–964.

67. Ishibashi M, Nishida H, Kumabe T, et al. Tc99m tetrofosmin. A new diagnostic tracer for parathyroid imaging. Clin Nucl Med. 1995;20:902–905.

68. Gallowitsch HJ, Mikosch P, Kresnik E, et al. Technetium 99m tetrofosmin parathyroid imaging. Results with double-phase study and SPECT in primary and secondary hyperparathyroidism. Invest Radiol. 1997;32:459–465.

69. Froberg AC, Valkema R, Bonjer HJ, et al. 99mTc-tetrofosmin or 99mTc-sestamibi for double-phase parathyroid scintigraphy? Eur J Nucl Med. 2003;30: 193–196.

70. Fjeld JG, Erichsen K, Pfeffer PF, et al. Technetium-99m-tetrofosmin for parathyroid scintigraphy: A comparison with sestamibi. J Nucl Med. 1998;38:831–834.

71. Kibar M, Tetiker T, Yuce S, et al. Double phase technetium-99m-tetrofosmin scintigraphy in the detection and localization of parathyroid adenomas: A comparative study with technetium-99m-sestamibi scintigraphy. Turk J Nucl Med. 1996;5:201–207

72. Wakamatsu H, Noguchi S, Yamashita H, et al. Technetium-99m tetrofosmin for parathyroid scintigraphy: A direct comparison with 99mTc-MIBI, 201Tl, MRI and US. Eur J Nucl Med. 2001;28:1817–1827.

73. Aigner RM, Fueger GF, Nicoletti R. Parathyroid scintigraphy: Comparison of technetium-99m methoxyisobutylisonitrile and technetium-99m tetrofosmin studies. Eur J Nucl Med. 1996;23:693–696.

74. Beggs AD, Hain SF. Localization of parathyroid adenomas using 11C-methionine positron emission tomography. Nucl Med Commun. 2005;26:133–136.

75. Otto D, Boerner AR, Hofmann M, et al. Pre-operative localisation of hyperfunctional parathyroid tissue with 11C-methionine PET. Eur J Nucl Med Mol Imaging. 2004;31:1405–1412.

76. Sundin A, Johansson C, Hellman P, et al. PET and parathyroid L-[carbon-11]methionine accumulation in hyperparathyroidism. J Nucl Med. 1996;37:1766–1770.

77. Weber T, Cammerer G, Schick C, et al. C-11 methionine positron emission tomography/computed tomography localizes parathyroid adenomas in primary hyperparathyroidism. Horm Metab Res. 2010;42:209–214.

78. Tang BNT, Moreno-Reyesa R, Blockleta D, et al. Accurate pre-operative localization of pathological parathyroid glands using 11C-methionine PET/CT. Contrast Media Mol Imaging. 2008;3:157–163.

79. Caldarella C, Treglia G, Isgro MA, et al. Diagnostic performance of positron emission tomography using 11C-methionine in patients with suspected parathyroid adenoma: A meta-analysis. Endocrine. 2013;43:78–83.

80. Melon P, Luxen A, Hamoir E, et al. Fluorine-18-fluorodeoxyglucose positron emission tomography for preoperative parathyroid imaging in primary hyperparathyroidism. Eur J Nucl Med. 1995;22:556–558.

81. Neumann DR, Esselstyn CB Jr., MacIntyre WJ, et al. Primary hyperparathyroidism: Preoperative parathyroid imaging with regional body FDG PET. Radiology. 1994;192:509–512.

82. Neumann DR, Esselstyn CB Jr., MacIntyre WJ, et al. Comparison of FDG-PET and sestamibi-SPECT in primary hyperparathyroidism. J Nucl Med. 1996;37: 1809–1815.