As mentioned previously, ultrasound is usually the first tool for localizing parathyroid lesions, as it is cost-effective, noninvasive, widely available, and radiation-free (1, 3, 33). However, its sensitivity is highly variable, ranging from 57% to 84% (34, 35), with a nadir of 47% when thyroid goiter coexists (35).

When located in typical sites, enlarged parathyroids glands are usually easily recognized as homogenous hypoechoic solid masses, with oval or oblong shape and peripheral vascularity, often with a typical extrathyroidal feeding vessel (8, 11, 36, 37). However, ultrasound is well-known to be an operator-dependent technique, and when lesions are located within the thyroid capsule, these features may be indistinguishable from those of thyroid nodules, making the differential diagnosis between IPAs and thyroid nodules challenging, or even impossible, sonographically (1, 3, 11, 33).

As a result, the sensitivity of ultrasonography for detecting IPAs is significantly reduced with a variable range from 29% to 67% (1, 5, 29, 32, 38, 39).

Yabuta et al. found that the most common ultrasound features of IPAs included a hypoechoic solid mass with smooth borders, a regular tumor shape and a characteristic hyperechoic line on the ventral surface of the adenoma, produced by the thin capsule between the parathyroid gland and the thyroid parenchyma (14).

In a recent study, Haciyanli et al. (21) proposed several criteria to distinguish IPAs from thyroid nodules:

Nevertheless, the differential diagnosis may be troublesome. Since IPAs usually mimic thyroid nodules with intermediate-high risk ultrasound characteristics (40), this can lead to unnecessary FNAb and potentially to misinterpretation of cytological results.

The main localization method for parathyroid lesions is 99mTc-MIBI scan, with a sensitivity of up to 88%, when studying a single-gland disease (41). However, when detecting IPAs, some studies report significantly lower sensitivity, around 60% (1).

This technique does have certain limitation, such as an unsatisfactory specificity, the ability of some thyroid nodules to uptake 99mTc-MIBI (6, 11) and poor spatial resolution (33).

Therefore, functional imaging cannot reliably distinguish between a thyroid tissue and a hyperplastic parathyroid gland, as normal thyroid gland or thyroid nodule also uptake this radioisotope (14). The most common cause of false positives is solid thyroid nodules (42), including toxic thyroid adenomas and oncocytic tumors (3). Additionally, delayed tracer washout has been observed in thyroid carcinoma (42), and 99mTc-MIBI uptake may strengthen the indication for FNAb, alongside factors such as nodule size and ultrasound risk level (10).

On the other hand, false negatives can occur in the presence of smaller adenomas or concomitant thyroid multiglandular disease (1). Therefore, a negative 99mTc-MIBI scan, especially in patients with a history of hyperparathyroidism, should not exclude the diagnosis. When a false negative is suspected, newer nuclear imaging modalities such as PET-TC with ¹⁸F-fluorocholine, ¹¹C‐methionine or ¹¹C-choline, have proven useful as second-line methods (1, 7).

¹¹C-methionine is retained in the parathyroid adenomas due to its involvement in the synthesis of parathyroid hormone precursors (43). The reported sensitivity of ¹¹C-methionine PET/CT is 72–86% (44), but a recent retrospective study (43) demonstrated high sensitivity (98%) and specificity (93%) for ¹¹C-methionine PET/CT. This technique appears to be particularly useful in cases of suspected ectopic parathyroid adenoma, prior unsuccessful surgery or inconclusive previous imaging (7).

In recent years, both ¹¹C-choline and ¹⁸F-fluorocholine PET/CT have gained attention as functional imaging techniques for parathyroid lesions (7).

The pooled sensitivity of ¹¹C-choline PET/CT is reported at 86% (34), although Noltes et al. observed a higher sensitivity of 97% in patients with primary hyperparathyroidism who underwent surgery following non-conclusive first-line imaging (44).

A major limitation of both ¹¹C-methionine and ¹¹C-choline imaging is the complex and costly production process of tracers, necessitating an on-site cyclotron due to their short half-lives (7, 45).

In contrast, ¹⁸F-fluorocholine PET/CT offers high sensitivity (90–96%) (44) and has the advantage of a longer half-life compared to ¹¹C-choline PET/CT and ¹¹C-methionine PET/CT, removing the need for on-site production (7, 43, 44). Additionally, thanks to a more amenable positron range, ¹⁸F-fluorocholine PET/CT offers higher spatial resolution than 99mTc-MIBI scan and ¹¹C-methionine PET/CT, enabling the detection of smaller or ectopic parathyroid adenomas, including IPAs (1).

False-negative findings with choline PET may occur in smaller adenomas or in adenomas with a low number of oxyphilic cells, and IPAs due to the masking effect of thyroid uptake (45). False-positive findings are often due to thyroid nodules that also may take up the radiotracer (45). It has been suggested that ¹⁸F-fluorocholine PET/CT may be particularly useful in the study of thyroid nodules with indeterminate cytology to estimating the risk of malignancy, due to the high negative predictive value (96%) (46). Thus, the presence of a ¹⁸F-fluorocholine-avid thyroid nodule could favor the decision to perform FNAb.

Recent guidelines do not specifically address the potential presence of IPAs (47). As noted in recent studies, ¹⁸F-fluorocholine-PET/CT has shown high accuracy in detecting benign parathyroid lesions, particularly when other imaging techniques are negative or discordant (48). However, data on IPAs remain limited.

Therefore, given the limited available data and the limitations of the various techniques discussed above, no single imaging modality is currently considered superior for differential diagnosis.

For all the reasons outlined above, in case of IPAs, inadvertent sampling of PTT by FNAb is a common pitfall in the diagnostic process of IPAs, particularly when a patient’s history of PHPT is not provided (49).

The prevalence of inadvertently sampling unsuspected PTT, presenting as thyroid lesions, has been reported to be up to 0.4% (6, 9, 11). Although cytomorphology alone may sometimes allow for a correct diagnosis, the ability of routine FNAb to distinguish between thyroid tissue and PTT is limited due to cytomorphological overlap, especially when parathyroid adenomas are located within the thyroid gland. In these cases, the sensitivity rages from 40% to 86% (6, 11, 37, 49).

Parathyroid cells share similar cytomorphological features with thyroid follicular cells. Consequently, PTT in FNAb sample can easily be confused with papillary thyroid carcinoma, Hürthle cell thyroid neoplasms and lymphocytic thyroiditis (19, 50). This is due to the presence of papillary fragments, epithelial cells arranged in microfollicular pattern, colloid-like material, anisokaryosis and the presence of oxyphil cells and naked nuclei of chief cells (50, 51), all of which may mimic neoplastic thyroid lesions (51).

Three different types of parathyroid cells have been described (52):

In these cases, a thorough medical history is essential. Known history of PHPT in the presence of a suspicious lesion justifies the request for specific ancillary tests that have been shown to be useful in distinguishing between parathyroid and thyroid lesions.

However, as in our case, the patient may be asymptomatic or may present with inconsistent symptoms of PHTP, which can lead to diagnostic delays (1). Moreover, calcium-phosphorus metabolism tests are not routinely required as part of the preliminary evaluations prior to FNAb.

Indeed, several cases of IPAs with cytological diagnosis of thyroid lesions have been reported in the literature, where the use of ancillary techniques allowed for the correct diagnosis on cytology specimens (9, 11, 13–15).

Therefore, in cases of high clinical suspicion for HPTH (e.g. a known history of HPTH), in the presence of a cytologically indeterminate intrathyroidal lesion and inconclusive imaging, the literature suggests that the use of ancillary techniques on FNAb specimens may be useful, as described below.

Firstly, the integration of cytological analysis with liquid-based preparation, in addition to the conventional smear, appears to be useful in the differential diagnosis (49). Common cytological features of parathyroid cells observed in liquid-based preparation include cells with small, dark nuclei, moderate to scant lacy cytoplasm, and, in some cases, oncocytic differentiation, clustered into small groups (49).

Park et al. compared the results of conventional smear and liquid-based preparation using ThinPrep and SurePath methods. Common features of parathyroid lesions observed in both conventional smear and liquid-based preparation included microfollicular structure, small round-to-oval nuclei with lymphocyte-like chromatin, and naked nuclei in the background. Specimens prepared using both ThinPrep and SurePath methods showed higher nuclear detail and better defined cytoplasm (53).

Moreover, bubbly or vacuolated cytoplasm was only observed in liquid-based preparation. Other findings included oxyphilic parathyroid cells, which may resemble Hürthle cells of the thyroid gland, although Hürthle cells tend to have larger sizes and plumper cytoplasm than oxyphilic parathyroid cells, and white blood cells in the background, which can be useful for differential diagnosis (53), as the nuclear size of parathyroid cells is similar to or smaller than that of inflammatory cells, while the nuclear size of follicular cells is larger (53).

However, the cytological features of parathyroid cells are various, and the liquid-based preparation may not be available at all centers.

Another additional method is the measurement of PTH in the washout fluid from FNAb, which has been shown to be helpful when hyperfunctioning parathyroid glands are suspected bat cannot be clearly identified via ultrasound examination, FNAb or 99Tc-MIBI scan (3, 13). Cansu et al. demonstrated that PTH-washout is superior to cytological examination for the localization of parathyroid adenomas in patients with negative imaging results (3). Furthermore, Abdelghani et al. showed that PTH-washout is highly sensitive and specific for the identification of PTT sampling in FNAb samples (54). In a recent retrospective study, PTH-washout reached 100% sensitivity and specificity (55). However, a key limitation of this technique is that a defined PTH-washout cut-off value for PTT has not yet been established. Various cut-off values have been proposed: Maser et al.’s suggestion that a PTH level above the normal reference range is diagnostic, while Abdelghani et al., and other specialists, proposed that PTH levels in the washout fluid should exceed serum levels, thus defining a washout to serum PTH ratio greater than 1 (3, 49). It has also been reported that a PTH washout value in FNAb specimens greater than 245 pg/mL strongly correlates with PTT (11). Another major limitation of this method is the labile nature of PTH, which requires specific preservation and transport conditions for the assay, limiting its widespread use (13).

Another ancillary study is the use of immunocytochemical staining on FNAb specimens, but this is only possible if cell block material is available. Staining for PTH and thyroglobulin has proven useful in distinguishing between PTT and thyroid tissue (13, 19, 49). Sardana et al. observed high specificity and sensitivity for PTH immunostain, with values of 100% and 85.7% respectively (6). False negatives can occur if there is insufficient hormone storage within individual cells (49). Therefore, thyroglobulin staining serves as a control to exclude the presence of thyroid tissue (49). Other immunocytochemical markers of neuroendocrine differentiation, such as chromogranin and synaptophysin, may also support the diagnosis (50, 56). The use of GATA-3 immunostaining has been proposed as a marker of parathyroid tissue, although it is less specific than PTH, as its positivity can be observed in other tumors as well as in lymphocytes (6). However, immunocytochemical staining is not routinely performed, and not all FNAb specimens are suitable for cell blocks preparation, as this technique requires a minimal level of cellularity (6). Some authors (6, 57, 58) have performed immunocytochemical staining on air-dried cytology smears or liquid-based cytology slides, but this approach is not yet considered standard practice.

Finally, emerging techniques, such as next-generation sequencing (NGS), enable the simultaneous identification of a broad spectrum of genetic alterations, requiring only minimal nucleic acid samples extracted from FNAb specimens. Molecular testing programs, such as the gene expression classifier (GEC) test (commercially known as Afirma, Veracyte, South San Francisco, Calif.) and Multi-Gene ThyroSeq Next-Generation, have been shown to be useful in confirming a parathyroid origin (13, 59, 60), but they remain costly and are not always available.