Outline

• Abstract
  
• INTRODUCTION
  
• MATERIALS AND METHODS
  
• RESULTS
  
• DISCUSSION
  
• CONCLUSION
  
• BIBLIOGRAPHY
  

Graphics

• TABLE I. Comparative...
  
• Fig. 1
  

Objectives/Hypothesis: High-resolution ultrasound and sestamibi scanning are regarded as the first-line methods for preoperative localization of parathyroid adenomas. The utility of ultrasound in reoperative cases has been questioned because of concern that scarring will obscure normal tissue planes and vascularity that are critical to identification of an adenoma using this imaging modality. The purposes of the study were to evaluate the ability of high-resolution ultrasound to accurately localize parathyroid adenomas in the reoperative exploration and to identify any factors that influence its accuracy

Study Design: Retrospective chart review at a tertiary care academic medical center.

Methods: All patients seen in referral for parathyroid surgery between May 1994 and September 2002 underwent high-resolution ultrasound as their initial diagnostic test. Patients who subsequently underwent exploration were included in the study. Intraoperative and ultrasound findings were compared.

Results: One hundred forty-two patients were included, 116 without and 26 with prior exploration. The sensitivity and positive predictive value of ultrasound were 86.9% and 89.1%, respectively. These data were not significantly different in patients without (88.2% and 90%) and in patients with (80% and 84.2%) prior thyroid or parathyroid surgery. The overall accuracy was 79% with a false-negative rate of 11.3%. Thyroid nodularity was significantly more common (81.8%) in patients who had a false-positive or false-negative finding on ultrasound than in the total population (61.3%).

Conclusion: High-resolution ultrasound is an accurate method for localizing parathyroid adenomas even in patients who have previously undergone parathyroid exploration. However, the presence of thyroid nodules can interfere with its accuracy.

Repeat operation after failed parathyroid exploration often confronts obscured tissue planes and distorted anatomy and vasculature.9 Therefore, the need for accurate anatomical preoperative localization in patients undergoing a repeat operation for a failed parathyroid exploration is paramount. Sestamibi’s failure to provide exact anatomical localization makes it less desirable than high-resolution US in reoperative cases if the latter is as accurate as in the previously unexplored case. The purposes of the present study were to evaluate the ability of high-resolution US to accurately localize parathyroid adenomas in the reoperative situation and to identify any factors that influence its accuracy.

MATERIALS AND METHODS^
Between May 1994 and September 2002, all patients referred to the Department of Otolaryngology—Head and Neck Surgery at Oregon Health Sciences University (Portland, OR) with primary hyperparathyroidism first underwent high-resolution US and were seen in our clinic for preoperative evaluation. Patients with secondary hyperparathyroidism were excluded from the study. High-resolution US (6–15 MHz) (Acuson Corporation/Siemens, Malvern, PA) was performed by one of the authors (d.b.k.) in all patients. Potential adenomas were identified using gray-scale windows and further analyzed using color and spectral Doppler US to assess the presence of increased vascularity. The US examinations were then prospectively analyzed for the presence or absence of a parathyroid adenoma and for the laterality (right or left side), if present. The interpretation of these examinations did not change with retrospective analysis. All patients who later had transcervical surgical exploration for excision of a parathyroid adenoma were included in the study. In both primary and reoperative cases, unilateral explorations were planned with extension to a bilateral exploration if necessary. Twenty-six patients in the study group had previously had surgical exploration and were referred; the low number of patients reflects the high success rate (98% at the authors’ institution) with initial explorations after localization. Results of the surgical explorations were then retrospectively compared with the initial results of US. None of the surgical explorations was performed in a minimally invasive, single-gland technique.

A finding on US was deemed a true-positive (TP) finding when it correctly predicted the presence and laterality of an adenoma. A false-positive (FP) finding occurred when the US determined that an adenoma was present but none was found surgically or if the US incorrectly lateralized the adenoma. A false-negative (FN) finding was defined as a negative US result when an adenoma was present surgically. Therefore, a true-negative (TN) finding occurred when the US found no adenoma and none was found surgically.

Statistical analysis was performed using Microsoft Excel (Redmond, WA) and XLSTAT (Paris, France). Sensitivity was defined as TP/(TP + FN), and specificity as TN/(TN + FP). Positive predictive value (PPV) was defined as TP/(TP + FP), and negative predictive value (NPV) as TN/(TN + FN). Accuracy was defined as (TP + TN)/(TP + TN + FP + TN). A [chi]² analysis was performed to determine levels of significance (P < .05).

RESULTS^
One hundred fifty-one patients with hyperparathyroidism were evaluated and ultimately underwent neck exploration. Nine patients were excluded because of the presence of parathyroid hyperplasia found at operation, leaving for analysis 142 patients in all. Twenty-six patients were referred after a failed exploration. In the overall group, there were 115 women and 27 men, with an age range of 20 to 95 years; 23 women and 3 men made up the reoperative group (age range, 30–86 y).

Results were tabulated for the total group in addition to separate analyses of the larger initial surgery group and the reoperative group (Table I). In the total group, US had a sensitivity and specificity of 86% and 35%, respectively. The PPV was 89%, whereas the NPV was 30%. The computed accuracy of high-resolution US was 79%.

In the reoperative group, the sensitivity (80%), specificity (50%), PPV (84.2%), and NPV (42.9%) were all comparable to those for the overall group. Accuracy in this group was also similar (73.1%). The [chi]² analysis showed no significant difference in any of these statistics (P > .05 in all five comparisons). Of these 26 previously failed explorations, four patients had undergone two previous explorations (the remaining patients had just one exploration), with US accurately identifying an adenoma in three of those four cases. Furthermore, 7 of the 26 patients had a hemithyroidectomy or total thyroidectomy as part of their previous neck surgery. Ultrasound was able to accurately localize an adenoma in five of the seven cases.

Further analysis showed that there were 22 inaccurate US examinations in the group of patients who had not had surgery previously (FN + FP). Of these 22 examinations, 18 patients (81.8%) had nodular thyroid glands. In the total group, 87 of 142 patients had the same degree of thyroid nodularity (61.3%). Comparison of these two groups showed a significant difference (P = .048), demonstrating that patients with inaccurate examinations were more likely to have concomitant thyroid nodules.

Of the 142 total examinations, there were 16 false-negative (11.3%) and 13 false-positive (9.2%) findings. The examinations with false-positive findings included US results that inaccurately lateralized a present adenoma. Four of the 26 reoperative cases had a mediastinal adenoma. Ultrasound failed to localize the adenoma in all four cases; further imaging modalities failed to localize an adenoma in two of these cases. Nonetheless, the adenomas were found on surgical exploration.

DISCUSSION^
Treatment for primary hyperparathyroidism has undergone an evolution from a standard bilateral neck exploration with four-gland identification and removal of an adenoma to unilateral and/or single-gland, minimally invasive techniques. As a result, the current focus in preoperative imaging for parathyroid surgery centers on delineation of the method with best accuracy and cost-effectiveness. Although US necessitates an experienced ultrasonographer, it provides three-dimensional anatomical localization in a noninvasive, relatively inexpensive method that avoids the need for ionizing radiation.1 However, it is limited by its ability to pick up only adenomas that reside in the neck. In contrast, sestamibi scintigraphy uses nuclear isotopes and ionizing radiation to provide a more generalized location of the adenoma (left vs. right side, neck vs. chest, and two-dimensional location relative to the thyroid) without an exact definition of surrounding relationships (Fig. 1A). The advantage of this technique centers on its wide availability, its relatively straightforward methodology and image interpretation, and its ability to identify adenomas outside the neck.

At the authors’ institution, the algorithm for parathyroid adenoma in the reoperative case includes, in order, high-resolution US, sestamibi scanning, CT, and, if needed, selective venous sampling for parathyroid hormone. The ultrasonographer provides one of three possible results: findings consistent with the presence of an adenoma, findings that suggest but cannot confirm the presence of an adenoma, or no adenoma noted. If the finding on US examination is categorically positive, a second imaging modality that corroborates the US is needed. If the finding on US examination is questionably positive or negative, testing continues until two independent imaging modalities localize the adenoma. The four tests listed are complementary and rely on different factors in their ability to identify an adenoma.

Ultrasound examination has the ability to examine vascularity in addition to position. Sestamibi relies on the differential washout of the radionucleotide tracer in thyroid versus parathyroid, which is, theoretically, relatively unaffected by scarring. Computed tomography can identify adenomas in both the neck and chest, using their vascularity as demonstrated by enhancement with contrast relative to the surrounding tissues.11 It is particularly useful for adenomas that are paraesophageal and in the thymus low in the neck. Venous sampling, although not anatomically precise, can help determine the side on which an adenoma is located and may localize its relative position in the neck versus the chest. This diagnostic approach has resulted in a high success rate.

The results reported in the present study with preoperative localization for patients who are presenting for initial adenoma excision fall within the reported rates in the literature.1,2,8 However, what is more important is the fact that there is no statistical difference in the ability of US examination to identify adenomas in patients who have failed previous explorations when compared with the total group. Thus, high-resolution US is a reliable first-line test in the preoperative planning of a repeat parathyroid exploration. The failure of US to localize adenomas in this setting may be a result of the anatomical location of parathyroid adenomas. However, in the current algorithm, failure of an US examination to adequately localize an adenoma would lead to other imaging modalities. In the total population, some authors have documented a low occurrence of ectopic parathyroid adenomas (<2%).2 However, a 1996 study by Shen et al.9 demonstrated that 50% of adenomas in reoperative patients were intrathymic or were located elsewhere in the mediastinum, and another study demonstrated 15 of 16 ectopic adenomas in reoperative cases.10 In the current study, 15% of reoperative adenomas were mediastinal in location. Nevertheless, the sensitivity and PPV in the present study’s reoperative group were similar to those for the overall group.

The finding that a significant correlation exists between an inaccurate US examination and the synchronous presence of thyroid nodularity (either solitary or in multinodular goiter: 82% of inaccurate examinations demonstrate thyroid nodularity) is important because it may help establish the overall significance of a negative or positive finding on US examination. In this setting, further testing for confirmation of adenoma position may be needed.

CONCLUSION^
High-resolution US is an effective tool for the preoperative localization of a parathyroid adenoma even in the reoperative setting. It is a first-line test that, in experienced hands, can provide an exact three-dimensional picture of adenoma location. However, the presence of thyroid nodularity can interfere with its accuracy (both false-positive and false-negative findings) and may dictate the need for additional preoperative imaging before exploration is undertaken.

BIBLIOGRAPHY^
1. Ulanovski D, Feinmesser R, Cohen M, Sulkes J, Dudkiewics M, Shpitzer T. Preoperative evaluation of patients with parathyroid adenoma: role of high-resolution ultrasonography. Head Neck 2002;24:1–5. [Context Link]

2. Reeder SB, Desser TS, Weigel RJ, Jeffrey RB. Sonography in primary hyperparathyroidism: review with emphasis on scanning technique. J Ultrasound Med 2002;21:539–552. Bibliographic Links ExternalResolverBasic [Context Link]

3. Haber RS, Kim CK, Inabnet WB. Ultrasonography for preoperative localization of enlarged parathyroid glands in primary hyperparathyroidism: comparison with (99m)technetium sestamibi scintigraphy. Clin Endocrinol 2002;57:241–249. Bibliographic Links ExternalResolverBasic [Context Link]

4. Miller P, Kindred A, Kosoy D, et al. Preoperative sestamibi localization combined with intraoperative parathyroid hormone assay predicts successful focused unilateral neck exploration during surgery for primary hyperparathyroidism. Am Surg 2003;69:82–85. Bibliographic Links ExternalResolverBasic [Context Link]

5. Bergenfelz A, Lindblom P, Tibblin S, Westerdahl J. Unilateral versus bilateral neck exploration for primary hyperparathyroidism. Ann Surg 2002;236:543–551. Ovid Full Text Bibliographic Links ExternalResolverBasic [Context Link]

6. Smit PC, Rinkes IH, van Dalen A, van Vroonhoven TJ. Direct, minimally invasive adenomectomy for primary hyperparathyroidism: an alternative to conventional neck exploration? Ann Surg 2000;231:559–565. Ovid Full Text Bibliographic Links ExternalResolverBasic [Context Link]

7. van Dalen A, Smit CP, van Vroonhaven TJ, Burger H, de Lange EE. Minimally invasive surgery for solitary parathyroid adenomas in patients with primary hyperparathyroidism: role of US with supplemental CT. Radiology 2001;220:631–639. Bibliographic Links ExternalResolverBasic [Context Link]

8. Koslin DB, Adams J, Andersen P, Everts E, Cohen J. Preoperative evaluation of patients with primary hyperparathyroidism: role of high-resolution ultrasound. Laryngoscope 1997;107:1249–1253. Ovid Full Text Bibliographic Links ExternalResolverBasic [Context Link]

9. Shen W, Duren M, Morita E, et al. Reoperation for persistent or recurrent primary hyperparathyroidism. Arch Surg 1996;131:861–867. Ovid Full Text Bibliographic Links ExternalResolverBasic [Context Link]

10. Feingold DL, Alexander HR, Chen CC, et al. Ultrasound and sestamibi scan as the only preoperative imaging tests in reoperation for parathyroid adenomas. Surgery 2000;128:1103–1109. Ovid Full Text Bibliographic Links ExternalResolverBasic [Context Link]

11. Gross ND, Weissman JL, Veenker E, Cohen JI. The diagnostic utility of computed tomography for preoperative localization in surgery for hyperparathyroidism. Laryngoscope 2004;114:227–231. [Context Link]

Key Words: Parathyroid adenoma; reoperative; ultrasound