Advertisement

Discordant findings on preoperative imaging for primary hyperparathyroidism and thyroid disease: Choosing the path to follow

Published:August 27, 2019DOI:https://doi.org/10.1016/j.surg.2019.05.057

      Abstract

      Background

      Preoperative localization of abnormal parathyroid glands in primary hyperparathyroidism is often obtained by sestamibi, and ultrasonography. We aimed to identify which modality is most accurate when laterality of abnormal glands on preoperative imaging is discordant.

      Methods

      A single institution retrospective review identified 112 consecutive patients with primary hyperparathyroidism who underwent successful parathyroidectomy and sestamibi with pertechnetate.

      Results

      Sestamibi with pertechnetate had a sensitivity of 72% and positive predictive value of 90%; ultrasonography had sensitivity of 50% and positive predictive value 80%. Patients with thyroiditis had lesser sensitivity and positive predictive value on sestamibi with pertechnetate (53% and 77%, respectively), in contrast to ultrasonography (54%, 88%, respectively). The sensitivity and positive predictive value of sestamibi with pertechnetate and ultrasonography did not differ in patients with thyroid nodules. Seventeen patients (15%) had discordant laterality on preoperative imaging. In discordant cases, sestamibi with pertechnetate was correct in 53% overall but in only 17% of those with thyroiditis (P = .01), whereas ultrasonography was correct in 26% overall but in 50% of those with thyroiditis (P = .01).

      Conclusion

      Thyroiditis decreased the sensitivity and positive predictive value of sestamibi with pertechnetate in primary hyperparathyroidism. In patients with discordant laterality on preoperative imaging, sestamibi with pertechnetate is the more accurate choice to guide operative planning, although ultrasonography may be a better guide in those with thyroiditis.

      Introduction

      Primary hyperparathyroidism (PHPT) presents most commonly as a single adenoma but is due to multiple gland disease in 10% to 20% of patients.
      • Wilhelm S.M.
      • Wang T.S.
      • Ruan D.T.
      • et al.
      The American Association of Endocrine Surgeons guidelines for definitive management of primary hyperparathyroidism.
      • Salti G.I.
      • Fedorak I.
      • Yashiro T.
      • et al.
      Continuing evolution in the operative management of primary hyperparathyroidism.
      • Vandenbulcke O.
      • Delaere P.
      • Vander Poorten V.
      • Debruyne F.
      Incidence of multiglandular disease in sporadic primary hyperparathyroidism.
      • Haciyanli M.
      • Lal G.
      • Morita E.
      • Duh Q.Y.
      • Kebebew E.
      • Clark O.H.
      Accuracy of preoperative localization studies and intraoperative parathyroid hormone assay in patients with primary hyperparathyroidism and double adenoma.
      With advances in imaging techniques, localization of abnormal glands preoperatively has allowed for a focused, minimally invasive approach in certain patients.
      • Wilhelm S.M.
      • Wang T.S.
      • Ruan D.T.
      • et al.
      The American Association of Endocrine Surgeons guidelines for definitive management of primary hyperparathyroidism.
      Multiple imaging modalities are used commonly for preoperative localization, including cervical ultrasonography (US), sestamibi scintigraphy (MIBI), sestamibi-single photon emission computed tomography (SPECT), and four-dimensional computed tomography (4D-CT). Sensitivity of these modalities varies, with a recent metanalysis reporting the sensitivity of US at 76%, MIBI plus SPECT at 79%, and 4D-CT at 89%.
      • Cheung K.
      • Wang T.S.
      • Farrokhyar F.
      • Roman S.A.
      • Sosa J.A.
      A meta-analysis of preoperative localization techniques for patients with primary hyperparathyroidism.
      In the setting of PHPT, thyroid disease, including thyroid nodules and parenchymal disease such as thyroiditis, has been shown to increase false positive and false negative rates with imaging.
      • Erbil Y.
      • Barbaros U.
      • Yanik B.T.
      • et al.
      Impact of gland morphology and concomitant thyroid nodules on preoperative localization of parathyroid adenomas.
      • Gomez-Ramirez J.
      • Sancho-Insenser J.J.
      • Pereira J.A.
      • Jimeno J.
      • Munne A.
      • Sitges-Serra A.
      Impact of thyroid nodular disease on 99mTc-sestamibi scintigraphy in patients with primary hyperparathyroidism.
      • Isik S.
      • Akbaba G.
      • Berker D.
      • et al.
      Thyroid-related factors that influence preoperative localization of parathyroid adenomas.
      • Hwang S.H.
      • Rhee Y.
      • Yun M.
      • Yoon J.H.
      • Lee J.W.
      • Cho A.
      Usefulness of SPECT/CT in Parathyroid Lesion Detection in Patients with Thyroid Parenchymal (99m)Tc-Sestamibi Retention.
      The addition of Tc-99m pertechnetate thyroid scan is used to highlight thyroid pathology and subtract what is highlighted from MIBI, theoretically leaving only the abnormal parathyroid tissue on MIBI. This approach improves localization of abnormal parathyroid glands in the setting of thyroid disease.
      • Powell D.K.
      • Nwoke F.
      • Goldfarb R.C.
      • Ongseng F.
      Tc-99m sestamibi parathyroid gland scintigraphy: Added value of Tc-99m pertechnetate thyroid imaging for increasing interpretation confidence and avoiding additional testing.
      Imaging modalities, including US, MIBI, SPECT, and pertechnetate, are often complementary and can be used together to aid operative planning.
      • Powell D.K.
      • Nwoke F.
      • Goldfarb R.C.
      • Ongseng F.
      Tc-99m sestamibi parathyroid gland scintigraphy: Added value of Tc-99m pertechnetate thyroid imaging for increasing interpretation confidence and avoiding additional testing.
      • Thanseer N.
      • Bhadada S.K.
      • Sood A.
      • et al.
      Comparative effectiveness of ultrasonography, 99mTc-sestamibi, and 18F-fluorocholine PET/CT in detecting parathyroid adenomas in patients with primary hyperparathyroidism.
      • Lee G.S.
      • McKenzie T.J.
      • Mullan B.P.
      • Farley D.R.
      • Thompson G.B.
      • Richards M.L.
      A multimodal imaging protocol, (123)I/(99)Tc-sestamibi, SPECT, and SPECT/CT, in primary hyperparathyroidism adds limited benefit for preoperative localization.
      But when these imaging modalities are discordant, operative planning can be conflicted. The aim of the present study is to identify which imaging modality is most accurate in discordant cases and how thyroid pathology influences the results.

      Methods

      After obtaining institutional review board approval, a single institution retrospective cohort study identified 113 consecutive patients from January 2014 through December 2015 with PHPT who underwent successful parathyroidectomy (PTX) and who had MIBIwith pertechnetate (MIBI+P) preoperatively. The use of US was surgeon dependent. All imaging studies were performed in our radiology department preoperatively. Successful PTX was defined as >50% decrease in serum parathyroid hormone levels (PTH) 10 to 20 minutes after all abnormal gland(s) were resected, as compared to baseline intraoperative levels and no documented persistent disease, defined as an normocalcemia within 6 months after PTX. Eight surgeons performed PTXs during this time, with 3 surgeons performing 88% of cases. Variables were collected from imaging reports, operative dictations, and pathology. True positives (TP), false positives (FP), false negatives (FN), and true negatives (TN) were identified for each side of the neck by comparing imaging findings to the intraoperative identification of an abnormal parathyroid gland with histologic confirmation of a parathyroid adenoma or hyperplasia, similar to previously published studies (Table I).
      • Christakis I.
      • Vu T.
      • Chuang H.H.
      • et al.
      The diagnostic accuracy of neck ultrasound, 4D-computed tomography and sestamibi imaging in parathyroid carcinoma.
      • Purcell G.P.
      • Dirbas F.M.
      • Jeffrey R.B.
      • et al.
      Parathyroid localization with high-resolution ultrasound and technetium Tc 99m sestamibi.
      Table IDefinitions of true positive, false positive, and false negative when comparing imaging localization and surgical pathology results
      True laterality
      Determined by operative resection and histopathologic confirmation of adenoma or hyperplasia. Each side of the neck was coded separately.
      True positiveFalse positiveFalse negativeTrue negativeLaterality on imaging
      Abnormal gland on right side only+--+Right
      -++-Left
      ++--Bilateral
      --++No glands
      Abnormal gland on left side only-++-Right
      +--+Left
      ++--Bilateral
      --++No glands
      Abnormal glands bilateral+-+-Right
      +-+-Left
      + +---Bilateral
      --+ +-No glands
      TP; true positive, FP; false positive, FN; false negative, TN; true negative
      Determined by operative resection and histopathologic confirmation of adenoma or hyperplasia. Each side of the neck was coded separately.
      Serum PTH levels were considered abnormal when >65 pg/mL and serum calcium when either total serum calcium was ≥10.5mg/dL or ionized calcium ≥5.5mg/dL. Ionized calcium was used preferentially when both total and ionized levels were available. Normocalcemia was defined as having no documented increase of either total serum and/or ionized calcium. Multiple gland disease was defined as ≥2 abnormal glands with cellular hyperplasia on surgical pathology and single gland disease was defined as 1 abnormal gland confirmed by surgical pathology. Thyroid nodules were identified on US or MIBI+P. Fine needle aspiration (FNA) was performed if nodules were >2 cm or if they exhibited other suspicious findings on US. Thyroidectomy or hemithyroidectomy were recommended after discussion with the patient if evidence of thyroid malignancy or follicular neoplasm was identified by FNA. Evidence of thyroiditis on imaging was defined as heterogenous thyroid parenchyma consistent with thyroiditis on US or impaired uptake on MIBI+P, as reported. Signs of thyroiditis on imaging were not correlated to lab work or surgical pathology. Rather, only imaging findings were used to define patients as having evidence of thyroiditis within this study.
      Statistical analysis was completed using IBM SPSS Statistics (IBM Corp, Armonk, NY). Continuous variables were expressed as either mean ± standard deviation or median (interquartile range, IQR). Variables were analyzed using Kruskal-Wallis and Mann-Whitney U tests.

      Results

      Patient demographics

      A total of 112 consecutive patients who underwent successful PTX and MIBI+P were identified, 93 (83%) also had US. Within the overall cohort, mean age was 59 ± 15 years, 82% (n = 92) were female, and the average body mass index was 31 ± 8. Overall, 82% (n = 92) of patients had increases in both serum calcium and PTH, 8% (n = 9) had inceased serum PTH but normal serum calcium concentrations, and 10% (n = 11) had increased serum calcium but normal serum PTH levels. Of those with normal serum calcium levels, 3/9 (33%) had low 25-OH vitamin D. Imaging detected 1 or more thyroid nodules in 52% (n = 59) of patients and abnormal thyroid parenchyma consistent with thyroiditis in 22% (n = 25). Overall, only 37% (n = 41) of patients had no thyroid disease, and 12% (n = 13) had both thyroid nodules and signs of thyroiditis. As a result of imaging, 29% (n = 32) underwent FNA of the thyroid nodules, 12% (n = 13) underwent total thyroidectomy, and 11% (n = 12) underwent hemi or partial thyroidectomy. Seven patients (6%) were found subsequently to have concomitant thyroid cancer on final pathology.
      Within our cohort, 31 patients (28%) underwent focused, minimally invasive PTX, while 25 (22%) had unilateral and 56 (50%) had bilateral explorations. Of the 56 bilateral explorations, 13 (23%) were due to a planned, concurrent thyroidectomy and 2 (6%) were due to a planned, concurrent hemithyroidectomy on the contralateral side of the neck, compared to the parathyroid disease. Eight patients (8%) had no abnormal parathyroid glands localized on imaging preoperatively, and in 2 patients (4%) the surgeon was suspicious of bilateral disease based on preoperative imaging. Bilateral exploration was performed in 13/56 (23%) patients due to lack of PTH decrease after gland resection and in 18/56 (32%) because intraoperative appearance of the glands was concerning formultigland disease. After evaluation by pathology, 32 patients (29%) had multigland disease. Median weight of the largest abnormal gland was 375 mg (IQR 245–685 mg). The mean weight of the parathyroid glands in patients who underwent bilateral exploration was less (276 mg, IQR 198–435) compared to those who underwent unilateral or focused PTX (480 mg, IQR 323–847) (P < .001).

      Performance of imaging modalities

      Preoperatively, 92% (n = 103) of patients had findings consistent with an abnormal parathyroid gland(s) localized on either US or MIBI+P. Nine patients (8%) had no gland localized on either imaging modality. No gland was localized on US in 25 patients (22%) and on MIBI+P in 10 patients (9%). US was not performed in 19 patients (17%). Of those without an US, an abnormal gland was localized on MIBI+P in 18 (95%) and MIBI+P identified the correct anatomic laterality in 16 (84%) patients.
      Overall, MIBI+P had a better sensitivity (Sn) and positive predictive value (PPV) (72% and 90%, respectively) compared to US (50% and 80%, respectively) (Table II). In patients with single gland disease (n = 81, 72%), Sn and PPV improved to 78% and 88% with MIBI+P and 59% and 75% with US. Both imaging modalities were poorly sensitive in multigland disease, but highly specific (Sn 51%, PPV 97% with MIBI+P; Sn 39%, PPV 86% with US). MIBI+P and US performed similarly in patients with thyroid nodules as compared to those without nodules. In patients with thyroiditis, however, MIBI+P performed considerably worse when compared to patients without thyroiditis (Sn 53%, specificity (Sp) 72%, PPV 77% vs Sn 71%, Sp 93%, PPV 94%). Conversely, US had a better Sn and PPV in the presence of thyroiditis (Sn 54%, PPV 88% vs Sn 49%, PPV 78%).
      Table IISensitivity and PPV of imaging modalities in detecting correct laterality of abnormal parathyroid gland
      Imaging ModalityTPFPFNTNSensitivitySpecificityPPVNPV
      All patients (N = 112, each with 2 lateralities)
       US
      Not all patients underwent US. No true negatives were present because laterality was determined per patient rather than per gland.
      5714585750%80%80%50%
       MIBI+P9210368672%90%90%70%
      Single gland disease (n = 80, each with 2 lateralities)
       US
      Not all patients underwent US. No true negatives were present because laterality was determined per patient rather than per gland.
      3813265259%80%75%67%
       MIBI+P639187078%89%88%80%
      Multiple gland disease (n = 32, each with 2 lateralities)
       US
      Not all patients underwent US. No true negatives were present because laterality was determined per patient rather than per gland.
      19130639%86%95%17%
       MIBI+P29128651%86%97%18%
      Thyroiditis (n = 25, each with 2 lateralities)
       US
      Not all patients underwent US. No true negatives were present because laterality was determined per patient rather than per gland.
      152131654%89%88%55%
       MIBI+P175151353%72%77%46%
      Thyroid nodules (n = 59, each with 2 lateralities)
       US
      Not all patients underwent US. No true negatives were present because laterality was determined per patient rather than per gland.
      339343449%79%79%50%
       MIBI+P485234268%89%91%65%
      FP, false positive; NPV, negative predictive value; PPV, positive predictive value; TN, true negative; TP, true positive.
      Not all patients underwent US. No true negatives were present because laterality was determined per patient rather than per gland.

      Negative imaging results and operative findings

      US was negative in 25 patients (22%), MIBI+P was negative in 9 (8%) patients, and 9 (8%) had no abnormal gland detected on either US or MIBI+P (Table III). Multigland disease was found in 4/9 (44%) of those with negative US and MIBI+P. Thyroid nodules were identified in 15/25 (60%) of patients with a negative US and in 6/9 (67%) of those with negative US and negative MIBI+P. Bilateral exploration was performed in 89% of patients with both negative US and MIBI+P, 70% of those with a negative MIBI+P, and 56% of those with a negative US.
      Table IIIRate of negative preoperative imaging and patient characteristics
      No abnormal gland detected on:Total n (%)SGD n (%)MGD n (%)Thyroid nodule n (%)Thyroiditis n (%)Type of surgery completed:
      Focused n (%)Unilateral n (%)Bilateral n (%)
      US only25 (22%)19 (76%)6 (24%)15 (60%)6 (24%)5 (20%)6 (24%)14 (56%)
      MIBI+P only9 (8%)6 (67%)3 (33%)4 (44%)3 (33%)2 (22%)1 (11%)6 (67%)
      Both US and MIBI+P9 (8%)5 (56%)4 (44%)6 (67%)3 (33%)0 (0%)1
      Bilateral exploration was planned, but dissection was stopped after unilateral dissection due to recurrent laryngeal nerve traction injury. IOPTH confirmed removal of the single abnormal gland.
      (11%)
      8 (89%)
      MGD, multiple gland disease; SGD, single gland disease;.
      Bilateral exploration was planned, but dissection was stopped after unilateral dissection due to recurrent laryngeal nerve traction injury. IOPTH confirmed removal of the single abnormal gland.

      Concordant imaging results

      Of patients who had abnormal glands localized on both MIBI+P and US preoperatively (n = 51), both imaging modalities localized to the same side of the neck in 34 patients (67%) (Table IV). In this concordant group, 10/34 (29%) had multigland disease. In patients with concordant imaging and single gland disease, correct laterality was found in 23/24 (92%). Overall, 20/34 (59%) were able to undergo unilateral or focused exploration, and 14/34 (41%) had bilateral exploration. There was no difference in the rate of thyroid nodules or thyroiditis between patients with correct and incorrect lateralization; incorrect laterality, however, was associated with older age (P = .002).
      Table IVConcordant imaging patient (n = 34) characteristics
      All concordant patients (n = 34)
      This includes only patients who had glands with the same laterality identified on MIBI+P and US. Excluded are patients with discordant findings (n = 17), no gland on US (n = 25), no gland on MIBI+P (n = 10), no US or MIBI+P (n = 9), or no US performed (n = 19). One patient had no gland on MIBI+P and no US was performed.
      Correct laterality (n = 25)Incorrect laterality (n = 9)P value
      Age58 ± 1555 ± 1569 ± 10.002
      Female sex28 (82%)21 (84%)7 (78%).675
      BMI27 ± 527 ± 628 ± 5.818
      Known thyroid disease11 (32%)8 (32%)3 (33%).942
      Thyroid nodule on imaging20 (59%)15 (60%)5 (56%).816
      Thyroiditis on imaging6 (18%)4 (16%)2 (22%).675
      Extent of operation<.001
       Unilateral or focused20 (59%)20 (80%)0 (0%)
       Bilateral14 (41%)5 (20%)9 (100%)
      Result of surgical pathology<.001
       Single gland disease24 (71%)23 (92%)1 (11%)
       Multiple gland disease10 (29%)2 (8%)8 (89%)
      Largest abnormal gland, milligram (median, IQR)430, 267–900510, 289–900370, 276–650.673
      BMI, body mass index.
      This includes only patients who had glands with the same laterality identified on MIBI+P and US. Excluded are patients with discordant findings (n = 17), no gland on US (n = 25), no gland on MIBI+P (n = 10), no US or MIBI+P (n = 9), or no US performed (n = 19). One patient had no gland on MIBI+P and no US was performed.

      Discordant imaging results

      A total of 17 patients (15% of all patients) had discordant laterality reported on US and MIBI+P (Table V). Thyroid nodules were present in 11 (65%) and thyroiditis in 5 (35%). Multigland disease was found in 29%, 41% of whom underwent bilateral exploration. In this group of patients, MIBI+P was correct in 53% of patients but only in 17% of those with thyroiditis (P = .01). In contrast, US was correct in 26% of patients, but this increased to 50% of those with thyroiditis (P = .01). The Figure illustrates imaging findings in a patient with thyroiditis in whom MIBI+P and US identified glands with opposite laterality; true laterality of the abnormal gland after resection was consistent with laterality on US.
      Table VDiscordant imaging (n = 17) patient characteristics
      All discordant patients (n = 17)US correct (n = 5)MIBI+P correct (n = 10)Neither correct (n = 2)P value
      Age59 ± 1160 ± 1257 ± 1162 ± 16.855
      Female sex15 (88%)5 (33%)8 (53%)2 (13%).316
      BMI30 ± 830 ± 631 ± 925 ± 5.401
      Known thyroid disease6 (35%)2 (33%)2 (33%)2 (33%).069
      Thyroid nodule on imaging11 (65%)2 (18%)8 (73%)1 (13%).279
      Thyroiditis on imaging6 (35%)3 (50%)1 (17%)2 (33%).012
      Extent of operation.130
       Unilateral or focused10 (59%)3 (30%)7 (70%)0 (0%)
       Bilateral7 (41%)2 (29%)3 (43%)2 (29%)
      Result of surgical pathology.025
       Single gland disease12 (71%)3 (25%)9 (75%)0 (0%)
       Multiple gland disease5 (29%)2 (40%)1 (20%)2 (40%)
      Largest abnormal gland, milligram (median, IQR)400, 234–725400, 174–490440, 250–1250228, 125–228.363
      Figure thumbnail gr1
      FigureUltrasonography (US) and MIBI+P in a patient with discordant imaging. (A) MIBI+P localizes the adenoma in left neck (arrows) after 90 minutes. (B) USof the left neck reveals a small normal parathyroid gland (small arrow) and demonstrates heterogenous parenchyma consistent with thyroiditis (large arrow). (C) US of the right neck reveals a large parathyroid adenoma (small arrow) and heterogenous parenchyma (large arrow).

      Surgical variances in discordant patients

      Surgical variances in patients with discordant preoperative US and MIBI+P are detailed in Table VI. When MIBI+P localized to a single laterality and US demonstrated bilateral parathyroid disease (n = 9), surgeons consistently started the procedure on the side localized on MIBI+P. Similarly, when MIBI+P indicated bilateral disease but US localized to a single laterality (n = 3), surgeons started on the side localized by US. When US and MIBI+P identified abnormal glands with opposite laterality (n = 5), the side a surgeon started on did not consistently follow an imaging modality. Interestingly, the surgeons started on the right side in all 5 cases. In the 2 patients without thyroiditis, MIBI+P identified the laterality of the abnormal gland correctly. In the 3 patients with thyroiditis, US identified unilateral disease correctly in 1 patient and 2 patients had bilateral disease.
      Table VIDiscordant imaging and operative results
      Patient IDLaterality on USLaterality on MIBI+PLaterality surgeon started onTrue laterality of abnormal glandsThyroid nodule presentLaterality of thyroid noduleUS or MIBI+P consistent with thyroiditis
      US bilateral glands, MIBI unilateral gland (n = 9)
       1BilateralLeftLeftLeftYesBilateralYes
       2BilateralLeftLeftBilateralYesLeftNo
       3BilateralLeftLeftBilateralNoN/ANo
       4BilateralLeftLeftLeftYesBilateralNo
       5BilateralLeftLeftLeftYesRightNo
       6BilateralLeftLeftLeftYesBilateralNo
       7BilateralRightRightRightYesBilateralNo
       8BilateralLeftLeftLeftYesBilateralNo
       9BilateralRightRightRightNoN/ANo
      MIBI bilateral glands, US unilateral gland (n = 3)
       10LeftBilateralLeftLeftNoN/AYes
       11RightBilateralRightBilateralYesBilateralNo
       12LeftBilateralLeftLeftYesBilateralYes
      Both MIBI and US unilateral, opposite laterality (n = 5)
       13RightLeftRightLeftNoN/ANo
       14LeftRightRightRightYesRightNo
       15RightLeftRightRightNoN/AYes
       16RightLeftRightBilateralNoN/AYes
       17LeftRightRightBilateralYesRightYes

      Discussion

      The operative approach of focused PTX with preoperative imaging and intraoperative PTH monitoring is well established in the treatment of PHPT.
      • Wilhelm S.M.
      • Wang T.S.
      • Ruan D.T.
      • et al.
      The American Association of Endocrine Surgeons guidelines for definitive management of primary hyperparathyroidism.
      • Ishii H.
      • Mihai R.
      • Watkinson J.C.
      • Kim D.S.
      Systematic review of cure and recurrence rates following minimally invasive parathyroidectomy.
      Preoperative imaging is performed most commonly with MIBI, and the additional of single-photon emission computed tomography (SPECT-CT) and pertechnetate has increased its sensitivity and specificity.
      • Cheung K.
      • Wang T.S.
      • Farrokhyar F.
      • Roman S.A.
      • Sosa J.A.
      A meta-analysis of preoperative localization techniques for patients with primary hyperparathyroidism.
      • Powell D.K.
      • Nwoke F.
      • Goldfarb R.C.
      • Ongseng F.
      Tc-99m sestamibi parathyroid gland scintigraphy: Added value of Tc-99m pertechnetate thyroid imaging for increasing interpretation confidence and avoiding additional testing.
      • Wang T.S.
      • Cheung K.
      • Farrokhyar F.
      • Roman S.A.
      • Sosa J.A.
      Would scan, but which scan? A cost-utility analysis to optimize preoperative imaging for primary hyperparathyroidism.
      Adding thyroid US to MIBI allows for an additional modality to identify abnormal parathyroid glands and concomitant thyroid pathology, which may require operative intervention.
      • Milas M.
      • Mensah A.
      • Alghoul M.
      • et al.
      The impact of office neck ultrasonography on reducing unnecessary thyroid surgery in patients undergoing parathyroidectomy.
      • Adler J.T.
      • Chen H.
      • Schaefer S.
      • Sippel R.S.
      Does routine use of ultrasound result in additional thyroid procedures in patients with primary hyperparathyroidism?.
      Using 2 imaging modalities, however, can lead to discordant results, impacting the operative planning and intraoperative decision-making. Our study showed that the presence of thyroiditis on either US or MIBI+P negatively impacts the Sn and PPV of MIBI+P. In addition, our results indicate that in patients with discordant imaging results and thyroiditis, US is correct more often.
      The strengths of our study include having a cohort of patients with complex imaging results and detailed correlation with intraoperative findings, as well as the inclusion of patients with negative imaging and multigland disease. Though 92% of our patients had preoperative localization of abnormal parathyroid glands, only 50% had focused or unilateral PTX. In those with negative imaging, those with a negative MIBI+P had a much greater rate of bilateral exploration than those with a negative US and positive MIBI+P. Interestingly, even in patients with concordant imaging findings, there was a 29% incidence of multigland disease and a 41% bilateral exploration rate. Other studies show 96% sensitivity when US and MIBI are concordant in single gland disease but only a 17% sensitivity with concordant imaging in multigland disease .
      • Adler J.T.
      • Chen H.
      • Schaefer S.
      • Sippel R.S.
      Does routine use of ultrasound result in additional thyroid procedures in patients with primary hyperparathyroidism?.
      • Arici C.
      • Cheah W.K.
      • Ituarte P.H.
      • et al.
      Can localization studies be used to direct focused parathyroid operations?.
      In our cohort, 15% of cases had discordant findings on MIBI+P and US which affected the operative planning. The accuracy of MIBI+P was decreased in patients who had signs of thyroiditis on either US or MIBI+P, whereas thyroid nodules had no effect on the performance of either study type. Thyroiditis is common and has been reported to be as great as 27% in women and 7% of men,
      • Arici C.
      • Cheah W.K.
      • Ituarte P.H.
      • et al.
      Can localization studies be used to direct focused parathyroid operations?.
      and therefore, understanding how it effects imaging studies used commonly in parathyroid surgery is important. Previous literature has demonstrated that the interpretation of parathyroid imaging is more difficult in the setting of autoimmune thyroiditis.
      • Erbil Y.
      • Barbaros U.
      • Yanik B.T.
      • et al.
      Impact of gland morphology and concomitant thyroid nodules on preoperative localization of parathyroid adenomas.
      • Hwang S.H.
      • Rhee Y.
      • Yun M.
      • Yoon J.H.
      • Lee J.W.
      • Cho A.
      Usefulness of SPECT/CT in Parathyroid Lesion Detection in Patients with Thyroid Parenchymal (99m)Tc-Sestamibi Retention.
      • Woods A.M.
      • Bolster A.A.
      • Han S.
      • et al.
      Dual-isotope subtraction SPECT-CT in parathyroid localization.
      Hwang et al demonstrated increased MIBI retention in the setting of autoimmune thyroid disease.
      • Hwang S.H.
      • Rhee Y.
      • Yun M.
      • Yoon J.H.
      • Lee J.W.
      • Cho A.
      Usefulness of SPECT/CT in Parathyroid Lesion Detection in Patients with Thyroid Parenchymal (99m)Tc-Sestamibi Retention.
      Increased retention of MIBI may lead to impaired detection of abnormal parathyroid tissue and account for the decreased performance of MIBI+P in patients with thyroiditis. Although we did not examine other imaging modalities in the present study, the utility of other imaging modalities such as 4D-CT in the setting of thyroid disease may also be altered.
      • Sho S.
      • Yuen A.D.
      • Yeh M.W.
      • Livhits M.J.
      • Sepahdari A.R.
      Factors associated with discordance between preoperative parathyroid 4-dimensional computed tomographic scans and intraoperative findings during parathyroidectomy.
      Our study shows that an additional benefit of multimodal preoperative imaging with US is the identification of thyroiditis, which can aid surgeons in choosing which imaging modality to follow when planning the operative approach.
      Limitations of our study include its retrospective nature, which limits our insight into operative planning based on the imaging modalities obtained. As a single institutional review, these findings may not be generalizable, especially given that results of imaging studies and specifically US are operator dependent. Our US was performed by US technologists and interpreted by radiologists, and therefore, these results may not apply to US at other institutions. Additionally, our study cohort represents that of a tertiary referral center with a considerably greater rate of multigland disease than other studies have described previously. Our small number of patients with discordant imaging may also limit the ability to detect statistical differences.
      In conclusion, thyroiditis decreased the performance of MIBI+P, but thyroid nodules had no effect on the performance of either MIBI+P or US. When MIBI+P and US had discordant laterality, MIBI+P was most often correct, except in the case of thyroiditis when US was correct more often. This observation suggests that while MIBI+P is the more accurate choice to guide operative planning when imaging is discordant, US may be a better guide in those patients with thyroiditis.

      Funding/Support

      A.C. Beck was supported by the National Institutes of Health grant T32CA148062.

      Conflict of interest/Disclosure

      No authors have conflicts of interest or other disclosures.

      References

        • Wilhelm S.M.
        • Wang T.S.
        • Ruan D.T.
        • et al.
        The American Association of Endocrine Surgeons guidelines for definitive management of primary hyperparathyroidism.
        JAMA Surg. 2016; 151: 959-968
        • Salti G.I.
        • Fedorak I.
        • Yashiro T.
        • et al.
        Continuing evolution in the operative management of primary hyperparathyroidism.
        Arch Surg. 1992; 127 (discussion 6–7): 831-836
        • Vandenbulcke O.
        • Delaere P.
        • Vander Poorten V.
        • Debruyne F.
        Incidence of multiglandular disease in sporadic primary hyperparathyroidism.
        B-ENT. 2014; 10: 1-6
        • Haciyanli M.
        • Lal G.
        • Morita E.
        • Duh Q.Y.
        • Kebebew E.
        • Clark O.H.
        Accuracy of preoperative localization studies and intraoperative parathyroid hormone assay in patients with primary hyperparathyroidism and double adenoma.
        J Am Coll Surg. 2003; 197: 739-746
        • Cheung K.
        • Wang T.S.
        • Farrokhyar F.
        • Roman S.A.
        • Sosa J.A.
        A meta-analysis of preoperative localization techniques for patients with primary hyperparathyroidism.
        Ann Surg Oncol. 2012; 19: 577-583
        • Erbil Y.
        • Barbaros U.
        • Yanik B.T.
        • et al.
        Impact of gland morphology and concomitant thyroid nodules on preoperative localization of parathyroid adenomas.
        Laryngoscope. 2006; 116: 580-585
        • Gomez-Ramirez J.
        • Sancho-Insenser J.J.
        • Pereira J.A.
        • Jimeno J.
        • Munne A.
        • Sitges-Serra A.
        Impact of thyroid nodular disease on 99mTc-sestamibi scintigraphy in patients with primary hyperparathyroidism.
        Langenbecks Arch Surg. 2010; 395: 929-933
        • Isik S.
        • Akbaba G.
        • Berker D.
        • et al.
        Thyroid-related factors that influence preoperative localization of parathyroid adenomas.
        Endocr Pract. 2012; 18: 26-33
        • Hwang S.H.
        • Rhee Y.
        • Yun M.
        • Yoon J.H.
        • Lee J.W.
        • Cho A.
        Usefulness of SPECT/CT in Parathyroid Lesion Detection in Patients with Thyroid Parenchymal (99m)Tc-Sestamibi Retention.
        Nucl Med Mol Imaging. 2017; 51: 32-39
        • Powell D.K.
        • Nwoke F.
        • Goldfarb R.C.
        • Ongseng F.
        Tc-99m sestamibi parathyroid gland scintigraphy: Added value of Tc-99m pertechnetate thyroid imaging for increasing interpretation confidence and avoiding additional testing.
        Clin Imaging. 2013; 37: 475-479
        • Thanseer N.
        • Bhadada S.K.
        • Sood A.
        • et al.
        Comparative effectiveness of ultrasonography, 99mTc-sestamibi, and 18F-fluorocholine PET/CT in detecting parathyroid adenomas in patients with primary hyperparathyroidism.
        Clin Nucl Med. 2017; 42: e491-e497
        • Lee G.S.
        • McKenzie T.J.
        • Mullan B.P.
        • Farley D.R.
        • Thompson G.B.
        • Richards M.L.
        A multimodal imaging protocol, (123)I/(99)Tc-sestamibi, SPECT, and SPECT/CT, in primary hyperparathyroidism adds limited benefit for preoperative localization.
        World J Surg. 2016; 40: 589-594
        • Christakis I.
        • Vu T.
        • Chuang H.H.
        • et al.
        The diagnostic accuracy of neck ultrasound, 4D-computed tomography and sestamibi imaging in parathyroid carcinoma.
        Eur J Radiol. 2017; 95: 82-88
        • Purcell G.P.
        • Dirbas F.M.
        • Jeffrey R.B.
        • et al.
        Parathyroid localization with high-resolution ultrasound and technetium Tc 99m sestamibi.
        Arch Surg. 1999; 134 (discussion 8–30): 824-828
        • Ishii H.
        • Mihai R.
        • Watkinson J.C.
        • Kim D.S.
        Systematic review of cure and recurrence rates following minimally invasive parathyroidectomy.
        BJS Open. 2018; 2: 364-370
        • Wang T.S.
        • Cheung K.
        • Farrokhyar F.
        • Roman S.A.
        • Sosa J.A.
        Would scan, but which scan? A cost-utility analysis to optimize preoperative imaging for primary hyperparathyroidism.
        Surgery. 2011; 150: 1286-1294
        • Milas M.
        • Mensah A.
        • Alghoul M.
        • et al.
        The impact of office neck ultrasonography on reducing unnecessary thyroid surgery in patients undergoing parathyroidectomy.
        Thyroid. 2005; 15: 1055-1059
        • Adler J.T.
        • Chen H.
        • Schaefer S.
        • Sippel R.S.
        Does routine use of ultrasound result in additional thyroid procedures in patients with primary hyperparathyroidism?.
        J Am Coll Surg. 2010; 211: 536-539
        • Arici C.
        • Cheah W.K.
        • Ituarte P.H.
        • et al.
        Can localization studies be used to direct focused parathyroid operations?.
        Surgery. 2001; 129: 720-729
        • Woods A.M.
        • Bolster A.A.
        • Han S.
        • et al.
        Dual-isotope subtraction SPECT-CT in parathyroid localization.
        Nucl Med Commun. 2017; 38: 1047-1054
        • Sho S.
        • Yuen A.D.
        • Yeh M.W.
        • Livhits M.J.
        • Sepahdari A.R.
        Factors associated with discordance between preoperative parathyroid 4-dimensional computed tomographic scans and intraoperative findings during parathyroidectomy.
        JAMA Surg. 2017; 152: 1141-1147