Mild chronic inflammatory infiltrates are not uncommon in the parathyroid and are seen in 10–17 % of patients at autopsy. These infiltrates, tend to be sparse, perivascular, predominantly lymphocytic, and do not appear to be associated with parathyroid dysfunction. Given their association with conditions that affect small vessel integrity such as septicemia, septic shock, and myocardial infarction, these inflammatory infiltrates are speculated to be nonspecific in nature [10, 11].

In contrast, true chronic parathyroiditis is a very rare (<0.1 %) inflammatory process of the parathyroid which presents with diffuse parenchymal involvement, germinal center formation, and evidence of immune-mediated parenchymal reaction, either destructive, in the form of fibrosis and atrophy or alternatively proliferative, in the form of hyperplasia [10] (Fig. 30.3). This apparently paradoxical reaction to the inflammation may be dependent on the nature of the auto antibodies involved, whether cytotoxic or stimulatory, analogous to the findings in the much more common Hashimoto’s thyroiditis and Grave’s disease of the thyroid [12, 13]. This putative autoimmune etiology is further supported by the identification of autoantibodies to the parathyroid in a subset of patients with this disease and its association with other autoimmune diseases [9, 12]. Treatment is dependent on clinical presentation.

Granulomatous parathyroiditis is a very rare finding usually associated with sarcoidosis and tuberculosis [10, 14]. It typically presents with hyperparathyroidism and in the company of a parathyroid adenoma or hyperplasia but the association is likely coincidental since the vast majority of parathyroidectomies are performed for hyperparathyroidism as a result of parathyroid parenchymal proliferation. The histologic findings are typical of granulomatous inflammation regardless of site and consist of the presence of giant cells and granulomata with or without central necrosis. The presence or absence of necrosis may be helpful in establishing etiology.

Parathyroid cysts are benign cysts of the parathyroid, and are thought to be of developmental or degenerative origin. They may account for up to 5 % of cystic lesions of the neck. They are usually uni-locular with a well-defined capsule and a smooth internal surface. They are variable in size and are usually filled with a clear fluid but may be bloody. Microscopically, they are lined by either chief cells or an attenuated flattened epithelium and may show adjacent normal or hyperplastic parathyroid tissue [15] (Fig. 30.4). Rarely, they are associated with heterotopic tissues such as salivary gland which supports the argument that they may be developmental in nature [16]. While they may be mistaken for thyroid cysts, the distinction is usually clinical insignificant. Since they can occur anywhere a parathyroid can exist, the differential includes other cysts of the neck and mediastinum, depending on site. If clinically relevant, testing the fluid for parathyroid hormone (PTH) may be helpful [15]. Excision is curative.

Conceptually, parathyroid hyperplasias are expansions of the parenchymal mass of the parathyroid in response to stimuli, either internal (i.e. MEN syndromes) or external (i.e. low serum-free calcium). The expansions may be diffuse or multifocal but they involve all of the parathyroid glands to varying degrees. As such, in parathyroid hyperplasia, there is no “normal” parathyroid. This understanding has been recently complicated by evidence that nodular expansions in hyperplasia are clonal in nature whereas diffuse expansions are polyclonal [20–22]. This finding blurs the distinction between hyperplasias, originally thought to be polyclonal in nature, and adenomas which are by definition monoclonal lesions. However, in any proliferating polyclonal population, individual cells are likely to have different proliferative potentials due to accumulated mutations and over time, nodular monoclonal populations arising from individual hyper-proliferative cells are arguably inevitable, with each nodule in a multi-nodular hyperplasia representing an independent clone.

Parathyroid hyperplasia may be primary, secondary, or tertiary depending on the mechanism of pathogenesis. In primary parathyroid hyperplasia, there is an increase in the functional cell mass of the parathyroid in the absence of a known stimulus (i.e. low serum calcium). While generally sporadic, 20–30 % of primary parathyroid hyperplasias are associated with familial hyperparathyroidism syndromes and multiple endocrine neoplasia (MEN) syndromes. In particular, in MEN 1 and MEN2A, 90 % and 30–40 % of patients exhibit parathyroid hyperplasia respectively [8, 9]. Secondary parathyroid hyperplasia is a physiologic response to chronic hypocalcemia, most commonly as a result of chronic renal failure, vitamin D abnormalities, or pseudohypoparathyroidism (end organ resistance to parathyroid hormone). Tertiary parathyroid hyperplasia is a failure to return to a euparathyroid state after the clinical stimulus for secondary hyperparathyroidism, e.g. chronic renal failure, is removed [9]. The mechanism for tertiary hyperparathyroidism is uncertain but there is evidence for both an elevation of the “set point” (concentration of half maximum PTH secretion) and a failure of the hyperplastic parathyroids to involute after removal of the stimulus for hyperplasia [23–26].

Regardless of the mechanism of hyperplasia, the gross and histologic features of parathyroid hyperplasia are similar with only minor differences. In general, the parenchymal expansion may be diffuse or nodular (Fig. 30.5) and range from symmetric to markedly asymmetric in both extent and composition across the parathyroid glands (Fig. 30.6). Total parathyroid gland mass can vary widely but is typically between 1 and 3 g [5]. Generally, all four glands are involved to some extent, in contrast to adenomas, which usually involve a single gland. The expansion of the parenchymal mass is usually predominantly composed of chief cells although there may be expansion of other cell types most commonly oxyphils, and pure expansions of other cell types do rarely occur (Fig. 30.7). This expansion of the parenchymal mass, whether nodular, diffuse or mixed, typically involves the entire gland without a discrete rim of residual normal parathyroid; the distinction between normal and hyperplastic tissue may be histologically difficult. The parenchymal cells can be arranged in diverse architectural patterns including solid sheets, cords, acini, and follicles [5, 9] (Fig. 30.8). As with all neuroendocrine lesions, random nuclear atypia may be present and is not a feature of malignancy (Fig. 30.9). However, uniform atypia is not a feature of hyperplasia and should raise the concern for malignancy. Mitotic activity may be present but is generally low (<1 per 10hpf). Atypical mitoses are not seen. Cystic changes, fibrosis, and calcification are uncommon but increase in prevalence with size and duration of the hyperplasia [9]. Hemorrhagic changes may be present, particularly in tertiary hyperplasias, likely due to the extended duration of the hyperplastic process. These changes are most likely degenerative in nature (Figs. 30.10).

Stromal fat is generally absent within nodular proliferations and decreased in the intervening areas of diffuse proliferation. However, given the inherent variability in the normal stromal fat content and distribution across individuals, hypercellularity is a somewhat tenuous indicator of hyperplasia and is most reliable when marked hypercellularity for age is present. Loss of intracellular fat is common and potentially a more reliable indicator of hyperplasia but some parathyroid hyperplasias can have abundant intracytoplasmic fat and interpretation can occasionally be challenging [9].

Although typically well circumscribed by the parathyroid capsule, hyperplasias, particularly those associated with genetic syndromes, may exhibit small nests of parathyroid tissue outside the capsule (“parathyromatosis”) [9, 27]. These nests may be primary as in MEN1 or secondary to prior neck surgery. In the latter case, the parathyromatosis is speculated to arise from auto-implantation of cells spilled during the excision of a hyperplastic or neoplastic parathyroid lesion or inadvertent transection of a normal parathyroid [28, 29]. In primary parathyromatosis, these nests can be distinguished from soft tissue invasion by their rounded contours and lack of a desmoplastic reaction. The situation is complicated in secondary parathyromatosis because the reactive fibrosis associated with prior surgery may mimic desmoplasia and produce angulated contours [9, 28]. However, given the relative rarity of parathyroid carcinoma, other criteria should be applied in the face of prior surgery. In either case, care should be taken to exclude lymphovascular invasion which may have a similar morphology.

Conceptually, parathyroid adenomas are benign autonomous monoclonal proliferations of the parathyroid parenchyma. As such they arise in a background of normal parathyroid tissue and are associated with genetic alternations which release them from the normal physiologic control of proliferation. Two genes have been identified which are associated with this uncoupling, the oncogene cyclin D1 (CCND1)/PRAD1 and the tumor suppressor MEN1 [30]. In the case of cyclin D, an inversion has been identified in up to 8 % of parathyroid adenomas which places the cyclin D1 gene adjacent to the 5′ regulatory elements of the parathyroid hormone (PTH) gene resulting in tissue specific upregulation of cyclin D1 expression. Additionally 30–40 % of sporadic parathyroid adenomas overexpress cyclin D1. In the case of MEN1, inactivating mutations have been identified in up to 30 % of sporadic parathyroid adenomas and the gene itself is implicated in the MEN1 syndromes which are strongly associated with parathyroid proliferations [3, 31].

Grossly, parathyroid adenomas tend to be soft, solitary, circumscribed, and tan to red brown in color. They are typically ovoid but may be lobulated and can range greatly in size from as little as 150 mg to greater than 100 g; most fall between 300 mg and 1 g. They may exhibit degenerative changes such as cyst formation, fibrosis, calcification, and hemorrhage. These changes are more common in larger adenomas [5, 8, 9, 32].

Histologically, parathyroid adenomas are markedly hypercellular and composed predominantly of chief cells although oxyphils, transitional oxyphils, and rarely clear cells may be present and occasionally predominate. These parenchymal cells are usually arranged in sheets and cords, although acinar and follicular architectures may be present and are sometimes prominent. These follicles may contain amorphous eosinophilic material reminiscent of colloid. The neoplastic cells are slightly larger than their normal counterpart but this difference is difficult to appreciate without side by side comparison. Stromal fat is typically absent except in the lipoadenoma variant. The lesion is circumscribed and may be thinly encapsulated. A rim of normal or atrophic parathyroid tissue is present in only 60–70 % of cases but when present is a helpful diagnostic discriminator between adenoma and hyperplasia (Fig. 30.11). Random nuclear atypia involving scattered cells and rare giant cells may be present but uniform atypia should not be seen. Mitotic figures are uncommon but when present generally number less than 1/10 hpf although higher mitotic rate are sometimes seen. Atypical mitoses are not seen [5, 8, 9, 32].

Immunohistochemical studies are not usually necessary for diagnosis but on occasion, the differential includes thyroid neoplasms and clear cell neoplasms from other sites. As with all benign proliferations of the parathyroid, parathyroid adenomas are positive for PTH, parafibromin, chromogranin A, cytokeratin 8/18, Gata-3, Bcl-2, RCC, Pax-8, and mdm2. Pertinent negative stains include TTF-1, thyroglobulin, Pax-2, and CD10. [9, 33–39]. Ki-67 typically shows a proliferative fraction of <5 % [37, 40]. Since a subset of parathyroid adenomas express calcitonin [41], if medullary carcinoma is in the differential, concomitant staining for TTF-1 is recommended as medullary carcinoma is positive for TTF-1 whereas parathyroid adenomas are not.

A prime example of the difficulties in reliably distinguishing between adenoma and hyperplasia is lithium induced hyperparathyroidism. Lithium induced hyperparathyroidism occurs in approximately 10 % of patients taking lithium. The mechanism of lithium induced hyperparathyroidism is unknown but it has been speculated lithium competitively binds to CaSRs, effectively increasing the free calcium set-point as well as increasing renal tubule reabsorption of calcium [60]. Since lithium exposure is systemic and withdrawal of lithium is often curative [61], lithium induced hyperparathyroidism should logically be due to hyperplasia [62, 63] analogous to the case of chronic renal failure. However, when surgeries are performed to treat lithium induced hyperparathyroidism, the bulk of the lesions are called adenomas [62, 64, 65]. While it has been speculated that lithium therapy may unmask subclinical primary hyperparathyroidism, this does not explain the observation that cessation of lithium therapy is often curative. It is much more likely that asymmetric parathyroid hyperplasias are being diagnosed as adenomas. This is supported by the high recurrence rate of lithium induced hyperplasias [64]. Given that nodular proliferations in secondary hyperplasias are clonal, it is not surprising that they would be individually indistinguishable from adenomas. The fact that there is low inter-observer concordance between parathyroid adenoma and hyperplasia does not help the situation [66].

If we accept that we cannot reliably distinguish between adenomas and asymmetric hyperplasias histologically, where does this leave us? While the current management, i.e. minimally invasive parathyroidectomy, appears to be effective in the short term (5 years or less), the higher recurrence rates seen in cases of “double adenoma” and lithium induce hyperparathyroidism suggest that we should treat them as hyperplasias with bilateral exploration and possibly subtotal parathyroidectomy.

Conceptually, an atypical parathyroid adenoma is a neoplasm which shows some histologic features worrisome for malignancy (Figs. 30.15 and 30.16) but does not have diagnostic malignant features such as infiltrative growth (capsular or soft tissue invasion), lymphovascular invasion, perineural invasion, or metastasis. These worrisome features include a thickened capsule, irregular capsular contours, entrapped intracapsular nests, trabecular growth pattern (Fig. 30.17), internal fibrosis, increased mitotic activity (>1 per HPF but <5 per HPF), atypical mitoses, coagulative necrosis, uniform atypia, and prominent nucleoli [5, 7–9]. Immunohistochemical studies also show a profile which is intermediate between adenoma and carcinoma [35, 37, 38, 67, 68]. Since molecular profiling studies do not support a progression from adenoma to carcinoma [17–19], this category most likely represents a heterogenous mix of adenomas and low-risk carcinomas which have overlapping histologic features rather than a true intermediate entity. Long-term follow-up in patients with atypical adenomas is quite scant but what little data there is suggests that these lesions are indolent with only rare cases of subsequent identification of overt malignancy [68–70]. As such, while patients require more aggressive follow-up for recurrence and surveillance for metastasis, the clinical course will most likely be benign.

Conceptually, parathyroid carcinoma is a malignant neoplasm arising from the parenchymal cells of the parathyroid which has both locally invasive and metastatic potential. Parathyroid carcinoma arises de novo rather than as a malignant transformation of existing adenomas [17–19]. It is a very rare malignancy and usually presents with marked hypercalcemia but may be nonfunctional [71]. Aberrant expression of various oncogenes and tumor suppressor genes have been identified in parathyroid carcinoma [72] but the HRPT2 gene is thought to play the primary role in carcinogenesis [35, 38, 73–78].

Grossly, parathyroid carcinomas tend to be somewhat larger than their benign counterparts but there is significant overlap. They typically presents as solitary white, firm, irregular, adherent masses, often described as difficult to dissect from surrounding tissues [5, 8, 9, 79]. This is in contrast to hyperplasias and adenomas which are tan to red brown, soft, well circumscribed, and easily excised.

The histologic diagnosis of parathyroid carcinoma is challenging. Histologic evaluation does not correlate well with clinical behavior [80, 81]. Likely factors producing this inconsistency include the inexperience on the part of the pathologist and surgeon due to the rarity of this disease; it is generally low-grade cytology; the locally aggressive nature of parathyromatosis, whether benign or malignant in nature; diagnostic dilemmas associated with prior fine needle biopsy or previous neck surgery; and the evolution of diagnostic criteria for malignancy [82].

Histologically, parathyroid carcinomas tend to have low cytologic grade and are composed of the same parenchymal cell types as parathyroid adenomas and hyperplasias. The histologic criteria for parathyroid carcinoma were first proposed by Schantz and Castleman [83] and included the presence of a thickened fibrous capsule or fibrous intra-lesional trabeculae; trabecular or rosette-like architecture; mitotic activity and capsular or vascular invasion. Later, additional features were noted including lobular architecture, irregular capsular contours, entrapped intracapsular nests, atypical mitoses, coagulative necrosis, uniform atypia, perineural invasion, and prominent nucleoli [5, 8, 9]. However, many of these features have also been identified in atypical adenomas and rarely, typical adenomas. While all of these characteristics are more common in parathyroid carcinomas and should prompt careful evaluation, the diagnosis of parathyroid carcinoma should rest on unequivocal malignant characteristics such as capsular invasion, lymphovascular invasion, perineural invasion, soft tissue invasion, or distant metastasis [84] (Fig. 30.18).

Because parathyroid carcinoma has proven to be such a difficult histologic diagnosis, significant research has focused on ancillary studies, in particular, the HRPT2 gene. Molecular studies suggest that up to 77 % of aggressive parathyroid carcinomas have mutations in HRPT2 [74, 75]. Unfortunately, 20 % of mutations lie outside mutational hotspots in exons 1, 2 and 7, and would require total gene sequencing to identify. Furthermore, many mutations are large deletions and some cases of HRPT2 downregulation appear to be epigenetic [82], neither of which can be identified by sequencing. These factors make the routine HRPT2 sequencing in parathyroid lesions impractical.

This has led to interest in parafibromin , the protein product of HRPT2. Immunohistochemical (IHC) studies of parafibromin in parathyroid carcinoma have been a mixed bag with some authors showing that loss/reduction of parafibromin nuclear staining is highly sensitive and specific for parathyroid carcinoma [38] while others have had considerably less promising results [82]. These discrepancies can be partially attributed to differences in selection criteria (behavior vs. histology); differences in assay protocols (different primary antibodies and retrieval methods); differences in the criteria for scoring reactivity; and finally, sample size. The results do suggest that loss of parafibromin is a feature of atypical adenomas and carcinomas but not typical adenomas. To improve sensitivity, parafibromin has been incorporated into IHC panels with other proteins associated with parathyroid carcinoma such as APC, Rb, Galectin-3, PGP9.5, and Ki-67, with promising results [40, 67, 85]. Loss of parafibromin also appears to independently predict recurrence/metastasis [68, 86]. Overall, these findings support the use of parafibromin IHC studies in the evaluation of parathyroid carcinomas and atypical adenomas. Given the technically difficult nature of the assay, testing should be done in a lab with significant experience and consistent results.

No standard staging system exists for parathyroid carcinoma. However, a number of histologic features have been identified as negative prognostic indicators including vascular invasion, lymph node metastasis, invasion into surrounding organs, and distant metastasis [87–89]. Based on these indicators Talat and Schulte [88] have proposed two classification systems, a classic TNM system and a binary high- and low-risk classification system. Both systems appear to be robust on retrospective Kaplan Meyer survival analysis [87, 88]. Immunophenotypic markers have also been correlated with survival/recurrence, including parafibromin, CaSR, and PGP9.5 [68, 86] and may ultimately need to be incorporated into any standard prognostic algorithm.

Metastatic disease to the parathyroids is uncommon but has been demonstrated in up to 12 % of patients with widespread metastatic disease at autopsy. This rather high percentage represents an end stage process; metastatic disease to the parathyroid as an initial presenting diagnosis is much rarer (<0.5 % of all parathyroidectomies). The most common primary sites in descending order of prevalence are: breast, melanoma, hematopoietic malignancies, lung, prostate, and kidney [92]. These metastases may present in one or more glands and are not usually associated with hyperparathyroidism. Since most of these metastatic lesions are high grade and present in late stages of the disease, diagnosis of metastatic disease is not usually challenging, although as previously noted, metastatic renal cell carcinoma can be mistaken for water-clear cell adenoma/carcinoma.