Hyperparathyroidism

Advances in Surgery j (2015) j–j

ADVANCES IN SURGERY Hyperparathyroidism What Preoperative Imaging Is Necessary? Rachel R. Kelz, MD, MSCE, FACS*, Douglas L. Fraker, MD, FACS Division of Endocrine & Oncologic Surgery, Hospital of the University of Pennsylvania, 3400 Spruce Street, 4 Silverstein, Philadelphia, PA 19104, USA

Keywords  

Hyperparathyroidism  Preoperative imaging  Parathyroidectomy Ultrasonography

Key points 

Imaging should be used to guide surgical decision making; it should not be used to make or confirm the diagnosis of hyperparathyroidism.



Positive imaging does not exclude the possibility of multigland disease, and, to date, small glands (<300 mg), ectopic glands, and multigland disease remain challenging to detect by all imaging modalities.



It is important that surgeons consider the use of additional intraoperative tools when operating on a patient with primary hyperparathyroidism regardless of the findings of the preoperative evaluation.



Primary hyperparathyroidism is a surgical disease for which there is no approved medical therapy; the morbidity of the operation with current strategies is minimal, and the benefits for the patient are tremendous.

T

he classic 4-gland exploration used to treat primary hyperparathyroidism has been challenged by the routine use of image-guided surgery. The successful use of preoperative localization studies to identify abnormal parathyroid gland(s) [1] has resulted in the frequent use of the minimally invasive approach to parathyroid surgery [2,3]. Surgeons and referring physicians encountering patients with hyperparathyroidism are now challenged to order radiographic studies to localize de novo disease. Testing, in this case, preoperative imaging studies (Box 1), should be considered only when it would alter the clinical decision making of the treating The authors have nothing to disclose.

*Corresponding author. E-mail address: [email protected] 0065-3411/15/$ – see front matter http://dx.doi.org/10.1016/j.yasu.2015.03.011

Ó 2015 Elsevier Inc. All rights reserved.

KELZ & FRAKER

2

Box 1: Imaging studies for preoperative evaluation, de novo primary hyperparathyroidism Noninvasive first-line studies High-resolution neck ultrasonography Technetium 99m sestamibi scan Noninvasive second-line studies MRI Computed tomography (CT) scan Single-photon emission computed tomography/CT scan Four-dimensional CT scana Invasive second-line study Ultrasound-guided jugular venous sampling a

Inappropriate for young patients or those with a history of well differentiated thyroid cancer or kidney dysfunction.

physician. More invasive testing may be appropriate for recurrent or reoperative disease than for de novo disease (Box 2). In the case of a patient presenting with a new diagnosis of hyperparathyroidism (de novo), imaging studies should be ordered by or in consultation with the treating surgeon. The results of the preoperative testing should be used to determine the surgical approach, not to decide on the appropriateness of surgical exploration. The decision to refer to a surgeon and to undergo the surgery should be based on the clinical scenario using evidence-based guidelines and clinical expertise [4]. Box 2: Imaging studies for preoperative evaluation, recurrent or persistent primary hyperparathyroidism Noninvasive High-resolution neck ultrasonography Technetium 99m sestamibi scan MRI Computed tomography (CT) scan Single-photon emission computed tomography/CT scan Four-dimensional CT scan* Invasive Selective venous sampling Angiogram Ultrasound-guided fine-needle aspiration * Inappropriate for young patients or those with a history of well differentiated thyroid cancer or kidney dysfunction.

IMAGING IN HYPERPARATHYROIDISM

3

The decision to perform a classic 4-gland exploration as opposed to a minimally invasive parathyroidectomy (MIP) varies across surgeons and can be influenced by the patient’s preferences. Parathyroid exploration by an experienced surgeon results in cure rates exceeding 95%, regardless of the surgical approach. Complication rates and long-term cure are believed to be comparable across the 2 groups [5–8]. Because of the excellent cure rates achieved in the setting of negative imaging findings, some experts still criticize the use of extensive preoperative localization studies in patients presenting for an initial operation on the parathyroid glands [9,10]. Moreover, the use of additional intraoperative tools, like intraoperative parathyroid hormone monitoring, may further influence the success of the operation [11]. Preoperative imaging practices should reflect the surgical approach, because multiple localization studies are not necessary, or cost effective, if the surgeon prefers a routine 4-gland exploration. ULTRASONOGRAPHY All patients undergoing parathyroid exploration, regardless of the cause of the disease, should have a preoperative neck ultrasound scan. The presence of a goiter, history of neck irradiation, or history of familial thyroid cancer may influence the decision to perform a simultaneous thyroid operation. The neck ultrasound scan is performed for 2 reasons: (1) to identify the presence of concomitant thyroid disease (and perform fine-needle aspiration, if indicated) and (2) to localize the abnormal parathyroid gland(s) when an MIP is being considered. When a simultaneous thyroid operation is planned, neck ultrasonography can also help by examining the lateral neck for evidence of lymph node metastases in case an occult malignancy is identified. Ultrasonography is the gold standard examination for evaluation of the thyroid gland. Guidelines exist to facilitate the appropriate use of image-guided biopsy when a nodule is identified. All suspicious nodules should be evaluated before neck exploration for hyperparathyroidism to minimize the need for reoperative surgery. Several studies have examined the use of preoperative thyroid imaging in patients referred for hyperparathyroidism, and the identification of occult thyroid disease has been estimated at 30% or more, with approximately 10% representing thyroid malignancy [12–14]. Ultrasonographic findings are influenced by the experience of the operating technician and interpreting physician. However, in experienced hands, the sensitivity is similar for radiologists and surgeons [15]. Moreover, the performance of ultrasonography as a test varies by the indication for the study. Most centers are comfortable with the examination of the thyroid using ultrasonography; however, not all extend the same level of expertise to the identification of parathyroid glands. Some radiologists are less familiar with the use of ultrasonography for the localization of abnormal parathyroid glands. The sensitivity of ultrasonography for the detection of abnormal parathyroid glands varies from 70% to 100% in most modern cohorts [16]. These estimates are often based on the ability to lateralize an abnormal gland, not the ability to localize the gland to the correct side and position (Table 1).

4

Table 1 Ultrasonography test characteristics for localization of abnormal parathyroid gland(s) Number Sensitivity, % Specificity, % of (95% confidence (95% confidence patients interval) interval)

Positive predictive value, %

Negative Accuracy, % predictive (95% confidence Additional value, % interval) findings

Notes

Attie et al [17], 1988 Whelan et al [18], 1989 Kohri et al [19], 1992 Kohri et al [19], 1992 Kohri et al [19], 1992

40 16 28 12 22

73 50 35 77 67

98 75 95 91 100

90

93

71 77 100

Kohri et al [19], 1992 Gilat et al [20], 2005

— — — — Secondary hyperparathyroid patients All patients —

68 77

58 89

94 33

Rodgers et al [21], 2006 Rodgers et al [21], 2006 Starker et al [22], 2011 Starker et al [22], 2011 Agha et al [23], 2012

Laterality Locationa Laterality Locationa —

75 75 87 87 30

57 (47–67) 29 (20–38) 71 48 74

Agha et al [23], 2012

Contrast enhanced

30

Kwon et al [13], 2013

—

105

a b

Location is defined as laterality and position. Values for median (range) in minutes.

79 91 10

91 59 78 87 68

— — — — —

92 98

64 —

74 —

94 (88–99) 86 (82–90) — — —

— — 87 82 —

— — — — —

— — — — —

100

—

—

—

—

93 (88–98)

—

97 (94–100) —

— Not limited by nodular thyroid disease — — — — Cost $56 Time 10 (8–15)b Cost $112 Time 15 (13–20)b —

91 (85–96)

KELZ & FRAKER

Author, year

IMAGING IN HYPERPARATHYROIDISM

5

Normal parathyroid glands are seldom visualized on ultrasonographic examination, because of the small size and relatively hypoechoic characteristic (Fig. 1). The most common reasons for false-positive examination results include enlarged lymph nodes and exophytic thyroid nodules [13]. Parathyroid glands are difficult to see on preoperative ultrasonography when they are located behind the thyroid gland or in ectopic locations. Furthermore, substernal visualization is not possible, so true thymic parathyroid adenomas are never shown on ultrasonography. Because ultrasonography cannot image behind the airfield trachea, superior parathyroid adenomas located in the tracheoesophageal groove are frequently missed by ultrasonography as well. This location is particularly challenging for patients with a large body habitus. When concomitant thyroid disease exists, the sensitivity of ultrasonography has been reported to decrease to 47% to 84% [2,20]. For patients with persistent or recurrent hyperparathyroidism, the sensitivity is reported between 36% and 63% [24,25]. Overall, high-resolution ultrasonography, a cost-effective and noninvasive test, is helpful for operative planning. SESTAMIBI SCAN The most common nuclear test used for the preoperative localization of abnormal glands(s) is the sestamibi scan. The use of nuclear scanning was popularized for the detection of abnormal parathyroid glands in the 1980s. The most common nuclear study used to identify the parathyroid gland(s) is the technetium 99m (Tc 99m) sestamibi scan [26]. The study uses Tc 99m– labeled methoxyisobutyl isonitrile (sestamibi). The mitochondria in the parathyroid cells sequester the tracer. In hyperfunctioning parathyroid glands, the washout of the tracer is delayed, resulting in the ability to identify the parathyroid gland(s) as distinct from the thyroid gland. The 2 most common techniques to differentiate between the thyroid and parathyroid gland(s) are the dual-phase or double-phase, single-contrast method (washout) and the dual-tracer or double-tracer, single-phase method

Fig. 1. Ultrasonographic images of right inferior parathyroid adenoma. (A) Without color flow; (B) with color flow. The blue arrow indicates the parathyroid adenoma. CCA, common carotid artery; SUBC, subclavian artery; THY, thyroid.

6

KELZ & FRAKER

(subtraction). The washout method relies on the strong uptake and longer retention of the tracer by the parathyroid gland when compared with the thyroid. Initial images are acquired, and then, delayed images are taken to capitalize on the kinetics of sestamibi sequestration. The presence of concomitant thyroid disease is a major limitation to the washout method. This imaging method does not allow the visualization of the thyroid alone. The dual-phase method achieves better results when combined with an anatomic scan such as concurrent ultrasonography or single-photon emission computed tomography (SPECT) (Tables 2 and 3). The subtraction method uses 2 agents: one to highlight the thyroid and the other the parathyroid glands. To highlight the thyroid gland by promoting uptake of a tagged isotope specific for the thyroid gland, thyroid hormone replacement is typically stopped 2 to 3 weeks in advance, adding to the morbidity of the procedure. Using digital subtraction, the parathyroid glands are more easily illuminated. Historically, the patient had to lie still to make sure that the images were well superimposed, thus making this method less comfortable for the patient. Now, the agents can be given simultaneously, so the inconvenience of immobility has been curtailed. The sestamibi scan has a reported sensitivity of 30% to 92%. Sestamibi is less accurate in multigland disease, when concomitant thyroid disease exists, and for glands that are located superiorly. Metabolic conditions can also affect the test performance. The sensitivity can be affected by vitamin D deficiency and thyroid suppression. Most patients with primary hyperparathyroidism undergo sestamibi imaging before surgery. The added usefulness of sestamibi scanning after ultrasonographic examination is not clear. The location in which ultrasonographic examination is most limited includes inferior parathyroid adenomas in the thymus and superior parathyroid adenomas in the tracheoesophageal groove behind the trachea. In these 2 areas, sestamibi scanning is particularly helpful, with high sensitivity, because they are some distance from the thyroid gland and not obscured by thyroid disease. One study suggested that sestamibi after ultrasonography added additional information in only 8% of patients [29]. These results may not be generalizable. Moreover, the additional information gained may dramatically influence the surgical approach when an ectopic gland is identified. It is unclear how many hyperparathyroid patients undergo sestamibi imaging overall, because many are not referred for surgery if an abnormal gland cannot be localized. This strategy may be a mistake because most cases can be cured by surgical exploration, regardless of the imaging findings. The role of sestamibi scanning in patients with secondary and tertiary hyperparathyroidism is less clear. Some advocate for sestamibi scanning to identify supernumerary glands and ectopic glands preoperatively. The dual-tracer method has a reported sensitivity of 30% to 90% in this population [19,30]. A recent meta-analysis found the pooled sensitivity to be 58% (95% confidence interval [CI], 52%–65%)and the pooled specificity to be 93% (95% CI, 85%– 100%) [30]. Because of the known limitations, especially in detecting small

Sensitivity, % (95% confidence interval)

Specificity, % (95% confidence interval)

Positive predictive value, %

Negative predictive value, %

Accuracy, % (95% confidence interval)

Author, year

Technique

Number of patients

Attie et al [17], 1988 Kohri et al [19], 1992 Kohri et al [19], 1992 Kohri et al [19], 1992 Kohri et al [19], 1992 Agha et al [23], 2012

Subtraction Subtraction Subtraction Subtraction Subtraction Washout

62 24 12 20a 62b 30

59 40 53 34 37 81

98 92 94 100 93 —

94 67 80 100 88 —

88 80 81 6 52 —

88 77 80 36 61 —

Kwon et al [13], 2013

Washout

105

92 (87–97)

—

97 (94–100)

—

90 (84–95)

Additional outcomes — — — — — Cost $448 Time 180 (150–200)c —

IMAGING IN HYPERPARATHYROIDISM

Table 2 Scintigraphy test characteristics for localization of abnormal parathyroid gland(s)

a

Only patients with secondary hyperparathyroidism. b All patients. c Values are median (range) in minutes.

7

8

Table 3 Tc 99m–Sestamibi-SPECT test characteristics for localization of abnormal parathyroid gland(s) Author, year

Condition

Number of Sensitivity, % (95% Specificity, % (95% Positive predictive Negative predictive patients Information confidence interval) confidence interval) value, % value, %

Rodgers et al [21], 2006 Rodgers et al [21], 2006 Witteveen et al [27], 2011 Witteveen et al [27], 2011 Starker et al [22], 2011 Starker et al [22], 2011 Ciappuccini et al [28], 2012

— — Primary Reoperative — — —

75 75 23 19 87 87 54

a

Laterality Locationa Laterality Laterality Laterality Locationa Locationa

65 (55–75) 33 (24–42) 67 33 62 40 92 (80–98)

88 (80–96) 83 (79–87) 100 80 — — 83 (36–100)

— — 100 63 84 79 —

— — 75 53 — — —

Location is defined as laterality and position.

KELZ & FRAKER

IMAGING IN HYPERPARATHYROIDISM

9

glands and along with frequent thyroid disease in the renal cohort, the usefulness of sestamibi in this cohort requires additional study. Sestamibi is often used after failed parathyroid exploration and in the case of recurrent disease. Before reoperation, imaging should be obtained, because the scar tissue increases the surgical risk. Because nuclear scanning alone does not provide detailed anatomic information, sestamibi scans are typically combined with cross-sectional imaging for patients with recurrent or persistent disease. SINGLE-PHOTON EMISSION COMPUTED TOMOGRAPHY SPECT images can be superimposed on sestamibi images to permit the use of 2 localizing agents in the detection of a parathyroid adenoma. Occasionally, multigland disease can be seen using SPECT (Fig. 2). The cross-sectional images enhance understanding of the anatomic location of the adenoma when compared with the scintigraphic images from the sestamibi scan. Using this technique, the reported sensitivity of the study is generally increased compared with scintigraphy alone, and in 1 report, the sensitivity increased from 54% for sestamibi alone to 79% when combined with SPECT [2]. The test does not perform so well in the reoperative setting when compared with the de novo assessment (see Table 3) [27].

Fig. 2. Two abnormal parathyroid glands identified on SPECT in the right paratracheal region (white arrows).

10

KELZ & FRAKER

MAGNETIC RESONANCE IMAGING The sensitivity of MRI varies from 50% to 82% in published studies (Table 4) MRI is not widely used in the imaging algorithm for de novo primary hyperparathyroidism. MRI can be used to identify abnormal parathyroid glands in the setting of persistent or recurrent disease. As a cross-sectional imaging modality, MRI is often preferred to computed tomography (CT) in patients with normal renal function, because of the lack of ionizing radiation exposure. The hyperfunctioning parathyroid gland is generally hypointense on unenhanced T1-weighted images and hyperintense on contrast-enhanced T2weighted images and T1-weighted images. MRI is expensive relative to some of the other studies and also depends on the radiologist’s level of expertise in the identification of parathyroid gland(s). MRI scans do not typically distinguish cervical lymph nodes from similar sized parathyroid adenomas, both of which are located in the same area inferolateral to the thyroid gland. Ultrasonography, on the other hand, distinguishes lymph node anatomy better by looking at the architecture, such as a fatty hilum, as well as the echogenicity of the parathyroids. Sestamibi scans distinguish lymph nodes from parathyroid glands by physiologic uptake of the isotope. COMPUTED TOMOGRAPHY Traditional CT scans rely on the vascularity of the parathyroid glands for identification of abnormal glands. Therefore, the use of contrast material increases the yield of the study. The addition of contrast and exposure to ionizing radiation makes the risk of the procedure less desirable than the alternative imaging modalities for patients with de novo hyperparathyroidism. Thin-cut scans are a helpful adjunct positive venous sampling after a failed exploration or suspected ectopic gland. CT scans can also be helpful in visualizing ectopic glands in the chest (Fig. 3). However, in some centers, interpreting physicians have reported that streak artifact and adenopathy make ectopic substernal glands difficult to identify. As with ultrasonography, the results are dependent on the radiologist’s experience. FOUR-DIMENSIONAL COMPUTED TOMOGRAPHY SCAN More recently four-dimensional (4D) CT scans have been used to identify the location of the parathyroid glands in patients with de novo hyperparathyroidism as well as in those patients presenting for recurrent or persistent disease. The sensitivity of the study ranges from 70% to 94% (Table 5). This study uses time as the fourth dimension in addition to the three-dimensional reconstructions. Over time, the rapid uptake and washout of the contrast material from the hyperfunctioning parathyroid gland(s) add additional information that is not readily available when using the more traditional protocols for CT of the neck. The 4D CT scan is more sensitive and specific in locating the correct quadrant for the abnormal gland than ultrasonography or sestamibi alone. However, its performance in multigland disease is inferior to that in single-gland disease. The study requires exposure to low-dose ionizing

Author, year

Sensitivity, % Specificity, % Positive Negative Specified patient Number of (95% confidence (95% confidence predictive predictive Additional population patients interval) interval) value, % value, % Accuracy, % outcomes

Attie et al [17], 1988 Whelan et al [18], 1989 Kohri et al [19], 1992 Kohri et al [19], 1992 Kohri et al [19], 1992 Agha et al [23], 2012

— — Primary Secondary All —

a

60 16 12 23 35 30

82 65 73 54 57 71

97 66 90 50 86

90 — 79 96 92

94 — 88 5 41

93 66 85 54 64

IMAGING IN HYPERPARATHYROIDISM

Table 4 MRI test characteristics for localization of abnormal parathyroid gland(s)

— — — — — Cost $672 Time 60 (50–70)a

Values are median (range) in minutes.

11

12

KELZ & FRAKER

Fig. 3. Sestamibi scanning showed an ectopic gland in the chest. Follow-up CT scan showed the intrathymic adenoma seen here (blue arrows). (A) Coronal image and (B) axial image. The patient underwent a curative transcervical thymectomy.

radiation and iodinated contrast material, and therefore is not recommended for young patients or those with well-differentiated thyroid cancer or impaired kidney function [22]. This study is not yet widely available for use.

VENOUS SAMPLING Venous sampling is invasive. For bilateral jugular venous sampling, blood is aspirated from bilateral jugular veins to measure parathyroid hormone levels. The gradient is used to determine the side of the suspected adenoma. Bilateral jugular venous sampling can be performed intraoperatively before incision in the case of nonlocalization by imaging or under direct visualization to lateralize the hyperfunctioning gland in the case of a missing adenoma. In 1 series, preoperative aspiration after failed imaging had a sensitivity of 52% and a positive predictive value of 65% [33]. More recently, bilateral jugular vein sampling as an ultrasound-guided office-based procedure has been reported to facilitate a minimally invasive approach after inconclusive ultrasonographic findings. In the case of inconclusive ultrasonographic results, the office-based procedure has been shown in a single study to have an accuracy of 81% [34]. Historically, selective venous sampling for parathyroid hormone levels was considered a helpful adjunct after failed neck exploration. In this invasive test, using catheters inserted through the deep venous system, successive venous samples are obtained from the neck and upper chest veins for parathyroid hormone levels. The samples direct the attention of the radiologist to an area of increased suspicion based on a spike in the parathyroid hormone (PTH) level. An angiogram can then be performed to identify the adenoma, because of its increased vascularity. Alternatively, thin-cut cross-sectional imaging can be performed across the area of suspicion to identify the affected parathyroid gland.

Author, year

Technique

Number of patients

Rodgers et al [21], 2006 Rodgers et al [21], 2006 Starker et al [22], 2011 Starker et al [22], 2011 Hunter et al [31], 2012 Hunter et al [31], 2012 Kelly et al [32], 2014 Kelly et al [32], 2014

Laterality Locationa Laterality Locationa Laterality Locationa De novo Reoperative

75 75 87 87 143 134 90 36

Accuracy, %

Sensitivity, % (95% confidence interval)

Specificity, % (95% confidence interval)

Positive predictive value, %

— — — — 94 87 78 84

88 (81–95) 70 (59–81) 94 86 — — — —

88(80–96) 89 (85–93) — — — — — —

— — 94 84 — — 86 73

IMAGING IN HYPERPARATHYROIDISM

Table 5 Four-dimensional CT scan test characteristics for localization of abnormal parathyroid gland(s)

a

Location is defined as laterality and position.

13

14

KELZ & FRAKER

Because of the invasive nature of the study, it is not used in the case of de novo hyperparathyroidism. SUMMARY Imaging should be used to guide surgical decision making. Imaging should not be used to make or confirm the diagnosis of hyperparathyroidism. Positive imaging does not exclude the possibility of multigland disease. Small glands (<300 mg), ectopic glands, and multigland disease remain challenging to detect by all imaging modalities. These are the same challenges that surgeons face during neck exploration. Therefore, it is important that surgeons consider the use of additional intraoperative tools when operating on a patient with primary hyperparathyroidism, regardless of the findings of the preoperative evaluation. Parathyroid surgery can now be performed safely using a 4-gland approach or minimally invasive technique. The imaging obtained in the preoperative setting should be determined by the clinical condition and planned operative strategy. Local talent and resources play a role in the test ordering strategies, given the important role of the interpreting physician in the test performance. Ultrasonography should always be performed and reviewed before surgery for the treatment of hyperparathyroidism to exclude simultaneous thyroid disease necessitating concomitant thyroid surgery. Additional studies should be ordered judiciously to avoid unnecessary costs. The use of intraoperative aids in conjunction with preoperative testing results yields exceptional cure rates. One important consideration regarding imaging for hyperparathyroidism is that negative results of imaging should never influence referral for potential surgical cure of this disease. Primary hyperparathyroidism is a surgical disease for which there is no approved medical therapy; the morbidity of the operation with current strategies is minimal, and the benefits for the patient are tremendous. A recent large series reported that the cure rate at 6 months in nonimaged patients was greater than 98% and not significantly different from patients who had positive imaging results [10]. Imaging studies can facilitate minimally invasive surgery, such as allowing the operation to be performed under conscious sedation, but should never determine who is referred for surgery, because the results are equivalent in experienced hands. References [1] Okerlund MD, Sheldon K, Corpuz S, et al. New method with high sensitivity and specificity for localization of abnormal parathyroid glands. Ann Surg 1984;200(3):381–8. [2] Mohebati A, Shaha AR. Imaging techniques in parathyroid surgery primary hyperparathyroidism. Am J Otolaryngol 2012;33(4):457–68. ˚ kerstro [3] Udelsman R, A ¨ m G, Biagini C, et al. The surgical management of asymptomatic primary hyperparathyroidism: proceedings of the Fourth International Workshop. J Clin Endocrinol Metab 2014;99:3595–606. [4] Bilezikian JP, Brandi ML, Eastell R, et al. Guidelines for the management of asymptomatic primary hyperparathyroidism: summary statement from the Fourth International Workshop. J Clin Endocrinol Metab 2014;99:3561–9.

IMAGING IN HYPERPARATHYROIDISM

15

[5] Fraker DL, Harsono H, Lewis R. Minimally invasive parathyroidectomy: benefits and requirements of localization, diagnosis, and intraoperative PTH monitoring. Long-term results. World J Surg 2009;33:2256–65. [6] Udelsman R, Lin Z, Donovan P. The superiority of minimally invasive parathyroidectomy based on 1650 consecutive patients with primary hyperparathyroidism. Ann Surg 2011;253(3):585–91. [7] Hodin R, Angelos P, Carty S, et al. No need to abandon unilateral parathyroid surgery. J Am Coll Surg 2012;215(2):297. [8] Norman J, Lopez J, Politz D. Abandoning unilateral parathyroidectomy: why we reversed our position after 15,000 parathyroid operations. J Am Coll Surg 2012;214:260–9. [9] Doppman JL, Miller DL. Localization of parathyroid tumors in patients with asymptomatic hyperparathyroidism and no previous surgery. J Bone Miner Res 1991;6(suppl2):S153–8. [10] Wachtel H, Bartlett EK, Kelz RR, et al. Primary hyperparathyroidism with negative imaging: a significant clinical problem. Ann Surg 2014;260:474–82. [11] Wharry LI, Yip L, Armstrong MJ, et al. The final intraoperative parathyroid hormone level: how low should it go? World J Surg 2014;38(3):558–63. [12] Ryan S, Courtney D, Timon C. Co-existent thyroid disease in patients treated for primary hyperparathyroidism: implications for clinical management. Eur Arch Otorhinolaryngol 2015;272(2):419–23. [13] Kwon JH, Kim E, Lee HS, et al. Neck ultrasonography as preoperative localization of primary hyperparathyroidism with an additional role of detecting thyroid malignancy. Eur J Radiol 2013;82:e17–21. [14] Sloan DA, Davenport DL, Eldridge RJ, et al. Surgeon-driven thyroid interrogation of patients presenting with primary hyperparathyroidism. J Am Coll Surg 2014;218(4):674–83. [15] Van Husen R, Kim LT. Accuracy of surgeon-performed ultrasound in parathyroid localization. World J Surg 2004;28:1122–6. [16] Solbiati L, Osti V, Cova L, et al. Ultrasound of the thyroid, parathyroid glands, and neck lymph nodes. Eur Radiol 2001;11:2411–24. [17] Attie JN, Rumancik WM, Moskowitz GW, et al. Preoperative localization of parathyroid tumors. Am J Surg 1988;156:323–6. [18] Whelan PJ, Rotstein LE, Rosen IB, et al. Do we really need another localizing technique for parathyroid glands? Am J Surg 1989;158:382–4. [19] Kohri K, Ishikawa Y, Kodma M, et al. Comparison of imaging methods for localization of parathyroid tumors. Am J Surg 1992;164:140–5. [20] Gilat H, Cohen M, Feinmesser R, et al. Minimally invasive procedure for resection of a parathyroid adenoma: the role of preoperative high resolution ultrasonography. J Clin Ultrasound 2005;33(6):283–7. [21] Rodgers SE, Hunter GJ, Hamberg LM, et al. Improved preoperative planning for directed parathyroidectomy with 4-dimensional computed tomography. Surgery 2006;140: 932–41. [22] Starker LF, Mahajan A, Bjo ¨ rklund P, et al. 4D parathyroid CT as the initial localization study for patients with de novo primary hyperparathyroidism. Ann Surg Oncol 2011;18:1723–8. [23] Agha A, Hornung M, Rennert J, et al. Contrast-enhanced ultrasonography for localization of pathologic glands in patients with primary hyperparathyroidism. Surgery 2012;151: 580–6. [24] Rodriguez JM, Tezelman S, Siperstein AE, et al. Localization procedures in patients with persistent or recurrent hyperparathyroidism. Arch Surg 1994;129:870–5. [25] Powell AC, Alexander HR, Chang R, et al. Reoperation for parathyroid adenoma: a contemporary experience. Surgery 2009;146:1144–55. [26] Kettle AG, O’Doherty MJ. Parathyroid imaging: how good is it and how should it be done? Semin Nucl Med 2006;36:206–11. [27] Witteveen JE, Kievit J, Stokkel MP, et al. Limitations of Tc99m-MIBI-SPECT imaging scans in persistent primary hyperparathyroidism. World J Surg 2011;35:128–39.

16

KELZ & FRAKER

[28] 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(3):223–8. [29] Untch BR, Adam MA, Scheri RP, et al. Surgeon-performed ultrasound is superior to 99Tcsestamibi scanning to localize parathyroid adenomas in patients with primary hyperparathyroidism: results in 516 patients over 10 years. J Am Coll Surg 2011;212:522–9. [30] Caldarella C, Treglia G, Pontecorvi A, et al. Diagnostic performance of planar scintigraphy using 99m Tc-MIBI in patients with secondary hyperparathyroidism: a meta-analysis. Ann Nucl Med 2012;26:794–803. [31] Hunter GJ, Schellingerhout D, Vu TH, et al. Accuracy of 4-dimensional CT for the localization of abnormal parathyroid glands in patients with primary hyperparathyroidism. Radiology 2012;264(3):789–95. [32] Kelly HR, Hamberg LM, Hunter GJ. 4D-CT for preoperative localization of abnormal parathyroid glands in patients with hyperparathyroidism: accuracy and ability to stratify patients by unilateral versus bilateral disease in surgery-naı¨ve and re-exploration patients. AJNR Am J Neuroradiol 2014;35:176–81. [33] Alvarado R, Meyer-Rochow G, Sywak M, et al. Bilateral internal jugular venous sampling for parathyroid hormone determination in patients with nonlocalizing hyperparathyroidism. World J Surg 2010;34:1299–303. [34] Carneiro-Pla D. Effectiveness of ‘‘office’’-based, ultrasound-guided differential jugular venous sampling (DJVS) of parathormone in patients with primary hyperparathyroidism. Surgery 2009;146:1014–20.