Surgeon-performed ultrasound is superior to 99Tc-sestamibi scanning to localize parathyroid adenomas in patients with primary hyperparathyroidism: results in 516 patients over 10 years

Abstract

Background: Surgeon-performed cervical ultrasound (SUS) and 99Tc-sestamibi scanning (MIBI) are both useful in patients with primary hyperparathyroidism (PHPT). We sought to determine the relative contributions of SUS and MIBI to accurately predict adenoma location.

Study design: We performed a database review of 516 patients undergoing surgery for PHPT between 2001 and 2010. SUS was performed by 1 of 3 endocrine surgeons. MIBI used 2-hour delayed anterior planar and single-photon emission computerized tomography images. Directed parathyroidectomy was performed with extent of surgery governed by intraoperative parathyroid hormone decline of 50%.

Results: SUS accurately localized adenomas in 87% of patients (342/392), and MIBI correctly identified their locations in 76%, 383/503 (p < 0.001). In patients who underwent SUS first, MIBI provided no additional information in 92% (144/156). In patients who underwent MIBI first, 82% of the time (176/214) SUS was unnecessary (p = 0.015). In 32 patients SUS was falsely negative. The reason for these included gland location in either the deep tracheoesophageal groove (n = 9) or the thyrothymic ligament below the clavicle (n = 5), concurrent thyroid goiter (n = 4), or thyroid cancer (n = 1). In 13 cases, the adenoma was located in a normal ultrasound-accessible location but was missed by the preoperative exam. In the 32 ultrasound false-negative cases, MIBI scans were positive in 21 (66%). Of the 516 patients, 7.6% had multigland disease. Persistent disease occurred in 4 patients (1%) and recurrent disease occurred in 6 (1.2%).

Conclusions: When performed by experienced surgeons, SUS is more accurate than MIBI for predicting the location of abnormal parathyroids in PHPT patients. For patients facing first-time surgery for PHPT, we now reserve MIBI for patients with unclear or negative SUS.

Figure 1

Flow diagram of study patients. *One patient had missing records, and 2 patients underwent localization with computerized tomographic scan. SUS, surgeon-performed ultrasound; MIBI, Tc-99m sestamibi scan; HPT, hyperparathyroidism.

Figure 2

Location of image-negative parathyroid adenomas identified at surgery. Adenoma locations were generally assigned to a specific quadrant as shown. Numbers indicate gland location identified at surgery. Numbers in parentheses indicate patients with a false negative SUS and true positive MIBI. Numbers in brackets indicate patients with false negative MIBI and true positive SUS (denominators differ because not all patients undergoing MIBI had an SUS performed). SUS, surgeon-performed ultrasound; MIBI, Tc-99m sestamibi scan; MGD, multiple-gland disease.

Figure 3

Weights of image true positive and false negative parathyroid adenomas. SUS, surgeon-performed ultrasound; MIBI, Tc-99m sestamibi scan. *p value < 0.001; ^p value < 0.001.

Figure 4

Anatomic location and MIBI positivity of patients with false negative SUS results. *Three patients with negative SUS had negative exploration not allowing definitive assignment of false negative status. SUS, surgeon-performed ultrasound; MIBI, Tc-99m sestamibi scan; TE, transesophageal groove.

Figure 5

Concurrent thyroid pathology and MIBI results of patients with false positive SUS results. SUS, surgeon-performed ultrasound; MIBI, Tc-99m sestamibi scan.