Robert J. Paresi, Jr., Chief General Surgery Resident; Gina Cunto-Amesty, Pathology Resident; Darsit Shah, Associate Professor, Departments of Surgery and Pathology, Monmouth Medical Center, Long Branch, NJ
Robert J. Paresi Jr., MD, MPH
Chief General Surgery Resident
Gina Cunto-Amesty, MD
Darsit Shah, MD
Departments of Surgery
Monmouth Medical Center
Long Branch, NJ
Introduction: Thyroid incidentalomas are fairly common and are usually benign; thus, it is not always clear when or how to evaluate these lesions further.
Results and discussion: This paper describes four cases of thyroid incidentalomas found in patients undergoing positron emission tomography (PET) scanning to evaluate previous metastases. All four incidentalomas were malignant. The authors review the literature and discuss the role of PET scanning in evaluating thyroid incidentalomas.
Conclusion: Thyroid incidentalomas observed on the PET scans of patients with known malignancies require further workup. Patients who are undergoing PET scans for cancer screening also have a higher risk of malignancy, and fine-needle aspiration should be undertaken if a thyroid lesion is found. The expense of PET scanning makes it cost-prohibitive for evaluating known thyroid lesions, but for patients whose cytology results are questionable, PET scanning may be more cost-effective than proceeding directly to a thyroidectomy.
Thyroid incidentalomas are previously undiagnosed thyroid lesions discovered on imaging studies conducted for another purpose, such as cancer screening. Incidentalomas have been observed with ultrasonography, computed tomography (CT), magnetic resonance imaging (MRI), and, most recently, positron emission tomography (PET).1-7 On PET scans, incidentalomas show focal uptake of radiotracer; this contrasts with the diffuse uptake of radiotracer observed in cases of thyroiditis. The presence of thyroid nodules in the general population has been reported to range between 19% and 46% on high-resolution ultrasound examinations.8-10 The risk of malignancy for a thyroid incidentaloma found on ultrasonography is low and ranges between 1.5% and 10%.8-10 Further studies are needed to determine the significance of thyroid incidentalomas seen on PET scans.
PET scanning is a functional imaging technique that is being used with increasing frequency to detect metastatic disease and to stage malignancies in cancer patients. The tracer most commonly used in PET scans is 18-fluoro-2-deoxy-D-glucose (FDG). Fluorine-18 (F18) of the FDG molecule crosses the cellular membrane and becomes trapped intracellularly after phosphorylation. Cells that metabolize glucose at a higher rate, such as cancer cells, will accumulate higher concentrations of FDG intracellularly and reveal more activity on PET scans.
In thyroidology, the use of PET scanning has been somewhat limited. It has been used to conduct follow-up examinations or localize disease in patients who experience recurrences of papillary or follicular thyroid cancer, demonstrate an elevated thyroglobulin level, or have negative whole-body radioiodine scans.11,12 In these clinical settings, the malignant cells at the site of recurrence will demonstrate concentration of FDG on PET scans, despite not taking up radioiodine. The use of PET scanning to determine whether a thyroid nodule is malignant has not been recommended. Fine-needle aspiration (FNA) remains the method of choice for diagnosing thyroid nodules. The significance of finding a thyroid incidentaloma on a PET scan remains unclear. We describe four cases of thyroid incidentalomas seen on PET scans. All four were found to be malignant.
Case 1—A 64-year-old woman with a history of breast cancer underwent a PET scan to screen for possible metastatic disease shortly after undergoing a right modified radical mastectomy and breast reconstruction. The PET scan revealed a focus of hypermetabolism (standardized uptake value [SUV] of 4.7) in the lower pole of the left lobe of her thyroid (Figure 1). Ultrasonography showed a 1-cm nodule in the thyroid's left lobe. FNA of the nodule was suspicious for papillary carcinoma, but metastatic breast cancer could not be ruled out. The patient underwent a total thyroidectomy without complications. The final pathology revealed a 1-cm focus of the tall cell variant of papillary carcinoma in the left lobe of the thyroid.
Case 2—A 37-year-old woman who had been treated with chemotherapy and radiation for squamous cell carcinoma of the buccal mucosa presented for a PET scan to stage her malignancy. The PET scan revealed no abnormalities of the oral cavity but did show a tiny focus of hypermetabolism (SUV 4.6) that correlated with the left lobe of the thyroid (Figure 2). A 0.68 x 0.5-cm nodule was observed in the left lobe of the thyroid on follow-up ultrasonography. FNA of the nodule showed papillary carcinoma. The patient underwent an uncomplicated left thyroid lobectomy with isthmus. The final pathology revealed a 0.5 x 0.7-cm focus of papillary carcinoma of the thyroid.
Case 3—A 60-year-old man with a history of non-Hodgkin's lymphoma presented for a PET scan to assess his stage IV lymphoma after completing chemotherapy. The PET scan revealed two foci of hypermetabolism in the patient's pelvis and a focus of hypermetabolism (SUV 3.3) in the right lobe of the thyroid (Figure 3). The areas of hypermetabolism in the pelvis correlated with active lymphoma. Ultrasonography confirmed a 2.2 x 2.3-cm complex cystic nodule in the right lobe of the thyroid. FNA of the nodule revealed only necrotic cellular debris. A core biopsy of the same nodule showed pathologic features suggestive of lymphoma, but other neoplastic processes could not be ruled out. The patient underwent a right thyroid lobectomy without complications. The final pathologic diagnosis of the nodule was large cell lymphoma.
Case 4—An 80-year-old woman with a history of Hodgkin's lymphoma underwent a PET scan after completing radiation treatment to evaluate her stage IV lymphoma. The PET scan revealed a focus of hypermetabolism (SUV 11.6) in the right lobe of the thyroid (Figure 4). Ultrasonography showed a 1.2-cm nodule in the right lobe of the thyroid. FNA of the nodule showed atypical cells that were highly suspicious for papillary carcinoma. The patient underwent an uncomplicated right thyroid lobectomy with isthmus. Pathology examination revealed a 0.9-cm focus of the tall cell variant of papillary carcinoma.
FNA, with or without image guidance, has become the diagnostic test of choice for evaluating thyroid nodules.
When a thyroid nodule is discovered on physical examination or incidentally during a diagnostic study, the immediate concern is the risk of cancer. A workup is needed to rule out malignancy, and FNA, with or without image guidance, has become the diagnostic test of choice for evaluating thyroid nodules. The use of FNA has resulted in better screening of patients, and the rate of malignancy identified on final pathology has increased from 14% to 47%.13 The role of PET scanning in evaluating thyroid nodules has yet to be defined. FDG-PET scanning has a reported sensitivity between 75% and 90% and a specificity of 90% in detecting thyroid malignancies.14,15
To determine the usefulness of PET scanning in evaluating thyroid nodules for malignancy, we must first define which patients could benefit from its use. Patients who have a history of malignancy are a select group and must be evaluated differently than the general population. The literature and the cases presented in our paper indicate that thyroid incidentalomas found on PET scans in cancer patients have an increased risk of malignancy and warrant further investigation.
Cohen and associates conducted a retrospective review of 4,525 patients who underwent PET scanning as part of a staging protocol for a known malignancy, during which 102 (2.3%) were identified as having a thyroid incidentaloma.3 Thyroid biopsy was performed for 15 of the patients, who were the focus of this study. PET scanning demonstrated diffuse uptake for 1 patient, who was then excluded from the study and later determined to have Hashimoto's thyroiditis. Of the 14 remaining patients, 7 (50%) were found to have thyroid cancer. Histopathological data were available for only 15 patients, which could suggest selection bias; nevertheless, the study's identification of a 50% rate of malignancy indicates further workup is warranted when a thyroid nodule is discovered in a patient with a previous malignancy.
All 4 patients in our report underwent thyroid ultrasonography and FNA of their thyroid incidentalomas as part of their workup. Clinical suspicion was warranted because 3 patients had papillary cancer at the area of hypermetabolism and 1 patient had metastatic large cell lymphoma. Peyton and associates reported 5 cases of papillary carcinoma discovered on PET scans done as part of a metastatic workup.6 These authors recommend further investigation of thyroid incidentalomas observed on PET scans in patients with other malignancies.
Focal thyroid incidentalomas seen on PET scanning in healthy patients appear to pose a high enough risk of cancer to warrant FNA of the lesions.
Bloom and colleagues reported on 19 patients who underwent a PET scan before thyroid surgery.5 Solitary thyroid nodules were seen in 12 patients and multinodular goiters in 7 patients. Four of the 12 solitary nodules were determined to be malignant, and 3 of the 4 malignancies showed areas of intensely increased FDG accumulation. Of the 15 benign nodules, 9 showed areas of increased FDG uptake on PET scans; thus, 3 of 12 patients (25%) whose preoperative PET scans demonstrated focal uptake had thyroid cancer. Although this study also has a selection bias, the authors did note that the dose uptake ratio of malignant lesions was significantly higher than the benign ones (10.8 vs 3.7, respectively).5 Bloom and colleagues' 1993 paper was one of the first to distinguish between the uptake of FDG by benign and malignant nodules on PET scans. Three of the four cancer cases described in our paper had SUVs lower than 5. The fourth cancer case demonstrated a particularly ominous progression of disease. This was illustrated through PET scanning, which showed an increase in SUV from 6.2 to 11.6 over 4 months in the area of the right lobe of the thyroid gland, pathologically identified as a 0.9-cm focus of the tall cell variant of papillary cancer. Thyroid lesions that demonstrate an SUV greater than 2.5 on PET scans appear to have a significant likelihood of malignancy, suggesting that higher SUVs correlate to greater risk of malignant disease.5
PET scanning may play a role in evaluating thyroid incidentalomas found during cancer screening. Kang and associates begin to explore this in their study of 1,330 patients who underwent PET scans; 331 patients were undergoing cancer screening and the remaining 999 were undergoing evaluation for metastatic disease. In this study, 29 patients (2.2%) showed focal (n=21) or diffuse (n=8) uptake on PET scanning. The two groups demonstrated similar rates of thyroid incidentalomas: 1.9% (19 of 999 patients) in the metastatic evaluation group and 3% (10 of 331 patients) in the cancer screening group (=0.27). No statistically significant difference was found between the two groups for focal incidentalomas (13 of 999 vs 8 of 331; =0.20). In the metastatic evaluation group, 3 of the 13 focal incidentalomas were identified as cancer, and 1 of the 8 focal incidentalomas found among patients in the cancer screening group was cancer. Only 3% of patients in the cancer screening group had focal incidentalomas, of which only 1 had cancer, suggesting that PET scanning is not useful as a screening test for thyroid cancer in the general population. Focal thyroid incidentalomas seen on PET scanning in healthy patients, however, appear to pose a high enough risk of cancer to warrant FNA of the lesions.
The cost and limited availability of PET scanning prohibits its use as a screening test in the management of thyroid nodules. The total cost of a PET scan can range from $1,400 to $7,800 per scan (economies of scale account for the range in costs).16 Palpation-guided FNA costs between $100 and $200 and an ultrasound-guided FNA ranges between $120 and $220. Using PET scans to evaluate thyroid nodules constitutes a 7- to 35-fold increase in cost and does not provide a tissue diagnosis.17,18 FNA, therefore, remains the tool of choice for evaluating thyroid nodules because of its low cost and diagnostic accuracy.
PET scanning may help determine which follicular lesions are at high risk of malignancy and should be excised.
PET scans may be useful in patients with inconclusive cytology results. FNA biopsy is inconclusive in up to 20% of patients with solitary nodules, and most of these patients subsequently undergo hemithyroidectomy. A study by de Geus-Oei and associates demonstrated the possible utility of PET scans for this patient population.19 In this study, 44 patients scheduled for a hemithyroidectomy following inconclusive FNA cytology results were administered PET scans pre- operatively to determine which patients likely had malignant thyroid nodules. All 44 patients underwent hemithyroidectomy regardless of the PET scan findings. Subsequent pathology results demonstrated that PET scanning had been 100% accurate in identifying preoperatively which patients had benign versus cancerous nodules. Therefore, the negative predictive value of
PET scans in this study was 100%. The use of this preoperative screening method would have reduced the number of negative hemithyroidectomies by 66%. Using PET scanning to evaluate patients with inconclusive cytology results could eliminate unnecessary thyroidectomies, justifying the expense of the scanning.
Follicular lesions remain a diagnostic pitfall of FNA because it is difficult to discriminate a well-differentiated thyroid cancer from a follicular adenoma. PET scanning may help determine which follicular lesions are at high risk of malignancy and should be excised. A recent study suggests that PET scans have limited value for selecting surgical candidates among patients with cytologically diagnosed follicular neoplasms, but further studies are needed to confirm this finding.20
PET/CT scan coregistration technology, a newer imaging modality, has shown some promise in evaluating patients with recurrent or metastatic thyroid cancer. However, its usefulness for examining thyroid nodules has yet to be determined.
Based on the cases we have managed and the published literature, focal thyroid incidentalomas identified during PET scanning of patients who have a history of cancer warrant further workup because these patients have an increased risk of malignancy. The use of PET scanning to screen for thyroid incidentalomas in healthy individuals is questionable. For individuals whose thyroid incidentalomas are identified during PET scanning as part of the cancer screening process, there is enough risk of malignancy to warrant FNA of the lesion. When cytology results are inconclusive, PET scanning may be a useful modality for distinguishing between benign and malignant disease. More studies are needed to determine the role PET scanning should play in evaluating thyroid incidentalomas.
The authors have no relationship with any commercial entity that might represent a conflict of interest with the content of this article and attest that the data meet the requirements for informed consent and for the Institutional Review Boards.