THYROID GLAND FROZEN SECTION PEARLS16 [gleaned mostly from this CME]:
- gross exam critical:
- gland capsule: don't call anything benign than goes from intrathyroidal to extrathyroidal (ATC can look amazingly fibrotic or inflammatory.
- nodule encapsulation: FTC & adenoma are completely encapsulated.
- pale foci, especially stellate: papillary thyroid cancer until proven otherwise.
- surgical anatomy:
- right & left lobes: lobectomies may not be shaped so that the pathologist can be absolutely sure if a right or left lobe (we must be very careful as to correct laterality in the path report). The space between the two lobes is bridged by an isthmus "lobe" of variable prominence.
- pyramidal lobe of Lalouette: arises from the superior aspect of the isthmus or anteromedial supra-isthmic aspect of either lobe.
- tubercle of Zuckerkandl: arises from the posterior edge of either lobe about at Berry's ligament (attached to the inferior margin of the cornu of the cricoid cartilage & close to the recurrent laryngeal nerve & inferior thyroid artery), angling as if to project posteriorly around the airway [L11-3439].
- pleomorphic cellularity: think ATC.
- spindled cells: think ATC until proven otherwise.
- squamous ca.: think ATC.
- looks MFH: think ATC.
- elongated follicles: think FVPTC.
- mitoses in thyroid cells: think insular (poorly differentiated) TC.
- lymphocyte filling of follicles: think lymphoma & NOT HT.
- hobnail nuclei: think angiosarcoma of thyroid.
- clear-cell nodule: RCC peculiarly metastasizes to thyroid nodules, but an adenomatoid nodule can be a clear-cell type. ATC has claer cell variant.
- prominent non-HT-pattern inflammation: think ATC.
By palpation or scan, a mass-like abnormality is detected in the
thyroid gland. Early multinodular goiter (hyperplastic or adenomatoid nodules) forms malfunctional
nodules and masses...neoplastic masses are benign (adenoma) or malignant
Here is the ARUP thyroid nodule decision tree & testing algorithm HERE.
These things are discovered by patients themselves, friends, dentists, examining physicians, and during medical imaging of the neck following emergency room visits for injuries requiring head & neck imaging (especially since about 1990). Endocrinologists find them during physical examinations which may include an ultrasound neck exam while in the office. Pre-operative diagnostic "fine needle aspiration" (FNA) cytology samples
are a way to possibly get a pre-operative indication of whether one
is dealing with nodular hyperplasia (malfunction: no more than lobectomy...if
any surgery) vs. nodular Hashimoto's (auto-immune: no more than lobectomy...if
any surgery...unless carcinoma is found) vs. papillary cancer (plan
ahead for total thyroidectomy) vs. follicular cancer (lobectomy only...but
see HCTC, below) vs. follicular adenoma (lobectomy only). "Follicular lesion" masses often cannot be definitely distinguished one from another by the VERY tiny sample from an FNA procedure (especially the office-based ones without any core biopsy). Hence, a concensus meeting of national and international experts formulated the Bethesda recommendations14 for how to handle the various results in reporting FNAs back to your doctor. If a sample did not have sufficient cellularity for diagnosis because it was mostly juice/fluid, that DOES NOT mean that the person performing the FNA procedure did anything wrong. Even an encapsulated cystic papillary thyroid cancer can have mostly cyst fluid and be hard to diagnose [L09-15565 = 3.5mm cancer in a 12mm cystic mass]!
NOTE: when thyroid surgery is performed for an expected benign condition, it is not uncommon to find a small cancer, usually a small papillary cancer [L10-11351].
NOTE: ENT & General Surgery basic Board exam question
that biopsy of a non-thyroid-gland neck mass not clearly in the thyroid that shows any type of thyroid tissue is metastatic thyroid cancer requiring thyroidectomy & neck dissection (it will essentially never be lateral abberant thyroid)! If an error is to be made, it is to be made on the side of removing what is most likely statistically to be a cancer, upon full & final pathology exam.
- anomalous nodules of anomalous lobe configuration
- ectopic thyroid tissue (anywhere along the developmental line of the thyroglossal duct (and some of the primitive anlage can be "dragged" even into the mediastinum by such as the developing heart); & ectopic follicles can be in cervical lymph nodes3.
- struma ovarii: thyroid tissue in an ovarian dermoid (such can be in dermoids in other locations).
- ultimobranchial body rests: small solid cell nests/clusters (p63 and TG positive8).
- thyroglossal duct: no lymphocyte accumulations (as happens in branchial cleft cyst); various epithelial linings, foci of thyroid in wall.
- intrathyroidal: most are degenerations of hyperplastic nodules; others have various linings; branchial cleft/pouch inclusion cyst [LMC-04-6880].
- hyperplastic nodule: (normal thyroid weight averages 15g for females & 18 for males) for various etiological reasons, the gland is trying to enlarge and "compensate" for insufficient levels of
circulating thyroid hormone; the multinodular thyroid..."goiter" when gland usually has increased weight...can have adenomatoid nodules of many and variable types of adenoma-like
histological pattern (have seen a goitrous nodule in a gland lobe of 2.3 total grams [L07-8888]):
- follicular with no/incomplete
capsule: "Nodules not showing these features [of true adenoma, below], even if they are solitary within a normal gland or in a gland showing evidence of microscopic nodular goiter, are called adenomatous ["adenomatoid"...adenoma-like] or hyperplastic nodules (if solitary within microscopic multinodularity, the term 'dominant hyperplastic nodule' may be used)2". This applies to any follicular nodule, be it a hyperplastic or adenomatoid Hurthle cell nodule [L-06-8455; L10-10700] or otherwise.
- usually have: (1) same cellularity both within & outside of nodule; (2) multiple nodules or background of follicular enlargement; (3) degenerative or metaplastic change in nodule associated usually with hyperplasia.
- multinodular hyperplasia: the term "goiter" is
used when the gland is actually enlarged; "goitrous change" can happen in a gland of normal weight range3[LMC-05-1523],
non-enlarged nodularity. But, even before "nodularity" there may be diffuse enlargement due to larger follicle size and accumulation of increased colloid. A nodule can have a fetal-type microfollicularity plus fatty metaplasia confined within it and "look" an awful lot like an intrathyroidal parathyroid nodule [L10-8004 parathyroid is TTF-1 negative] [adenomata can rarely be multiple9]. We had a case of not pre-operatively diagnosed nontoxic Hashimoto's goiter with diffuse Hurthle cell change [L11-13727].
- papillary hyperplasia: in hyperthyroid gland, especially if autoimmune8.
- papillary hyperplastic nodule: in young & usually female; solitary & 20% are hot, 10-15% of cases being in hyperthyroid gland & is 2-3 cm with capsule, often cystic & cells have oncocytic cytology with dark round nuclei8.
- adenoma (benign): has a usually thin (but well developed9)...to naked eye...& always complete16 capsule &
almost always solitary...usually scan cold or cool, sometimes warm, but rarely "hot"...a toxic adenoma.
****True adenoma is all of: (1) completely encapsulated (tends to be thick8), (2) adenoma does not
infiltrate capsule, (3) is follicular, (4) and cells usually look different than cells of surrounding normal thyroid2,8. (5)
Lacks nuclear changes of PTC. Within the adenoma, (6) the cytohistological pattern is the same3 (but some point out that both hyperplastic
& adenoma can vary within nodule due to degeneration and metaplasia9). (7) It is always "clonal"2 (but hyperplastic can be
- follicular adenoma, usual (normofollicular; simple).
- colloid adenoma (macrofollicular).
- fetal adenoma (microfollicular): this variant is of very small follicles sometimes amongst almost solid sheets of follicular cells & this gives a cut surface gross character seeming like an undifferentiated malignancy, especially when large[L07-5519].
- trabecular/solid (embryonal): columns of small closely packed follicle cells with minimal follicle formation.
- hyalinizing trabecular adenoma (often in Hashimoto's
cases [LMC-03-947])(HTA): can have psammoma bodies & lots of nuclear grooves & may be able to see the distinctive "cytoplasmic yellow body" in paranuclear location...yellow & refractile1.
- Hurthle cell adenoma (HCA): Hurthle cells (oxyphil cells; oncocytic cells) are larger
pink thyroid follicle cells stuffed with mitochondria.
Fine needle aspirates having such cells could reflect goiter
nodules, small foci of Hurthle cell change in other adenomata or nodules, Hashimoto's thyroiditis, HCA, or HCC (see below). When
Hurthle cells on smears, the most specific position one
can take is that there is generic evidence of a Hurthle
cell neoplasm/tumor. Of Hurthle cell neoplasms/tumors,
about 30% turn out to be HCC & 70% HCA. So, one must
first remove the thyroid lobe for careful differentiation
between HCC (malignant) & HCA (benign)...frozen section
is not recommended (one must assume it is cancer until
proven otherwise by lobectomy). If is HCA, this is all
the surgery needed. If it is HCC, see below. Beware misdiagnosing HCA/HCC when it is really an oncocytic follicular variant of papillary thyroid ca. or an oncocytic tall-cell variant of papillary thyroid ca.
- follicular adenoma with papillary architecture1: FA with some papillations or pseudopapillations.
- clear cell adenoma: signet ring vs. mucin-producing vs. lipid-rich1.
- atypical adenoma: pronounced cell. prolferation & irregular architecture1. Can even suggest sarcoma & need many sections to prove capsule not invaded3.
- adenoma with bizarre nuclei: only oddity is clusters of huge dark nuclei1.
- adenolipoma: adenoma with fatty metaplasia1.
- adenochondroma: adenoma with chondroid metaplasia1.
- black adenoma: adenoma with black pigment assoc. with minocyclin therapy.
- diffuse enlargement:
- Hashimoto's thyroiditis (struma lymphomatosa): plasma cells, lymphoid follicles with germinal centers; but, as seen in the coroner's autopsy population, Hashimotos inflammation most often does not cause an enlarged gland. It can have an associated encephalopathy which may manifest as memory problems to muscle twitching (fasciculation...and more ominous causes here) to seizures.
- granulomatous thyroiditis: typically involves only one area of gland.
- Riedel's struma: "invasive" fibrous thyroiditis.
- chronic nonspecific thyroiditis: focal lymphocytic infiltration, even with fibrosis, is fairly common but can even get diffuse enough to be concerning for lymphoma or small cell carcinoma3.
- nodular goiter: usually visibly multinodular but nodularity can be microscopic.
- hamartomatous adiposity.
- coincidental curiosities:
- melanosis thyroidii: some dark pigment in most thyroid follicular cells but can be so pronounced as to cause "black thyroid". Due to an accumulation of combined mix of colloid & organelles (ambilysosomes)17; and accelerated by minocycline (Rx for acne)3[LMC-05-1523]. Background follicular gland and a hyperplastic nodule may have pigment, yet a follicular variant of PTC did not [L13-6526-7924]. This is different than the melanin pigment rarely seen in medullary thyroid cancer (such a case had a black mass).
MALIGNANT: the detection of cancer...predominately papillary cancer...depends on how thorough the histological study (33% of thyroid glands serially sliced every
2mm and totally processed histologically will disclose a small cancer [101 consecutive Finnish autopsies10]). TC behavior is according to the worst differentiated area in a tumor (if is FTC but has a focus of[prognostic
medullary thyroid carcinoma (MTC): cancer of C cells in solid clusters surrounded by dense stroma; serum calcitonin follow-up; imaging
exams may show characteristic coarse calcification [LMC-01-5521];
watch out for MEN syndromes. DDX includes intrathyroidal parganglioma (TTF-1 negative). Large cells with basophilic or amphophilic cytoplasm and neuroendocrine type nuclei. They can be pigmented18.
- differentiated thyroid carcinoma (DTC):
Treatment controversial with no less than lobectomy for minimally invasive FTC and PTC; many prefer total/subtotal surgery with postop radio-ablation and TBG serum monitoring. More radical
surgery with more agressive & extensive cancer...radical neck for medullary. The key job of frozen section is to identify differentiated thyroid cancer (DTC) from worse malignancies .
- follicular thyroid carcinoma (FTC): "...characterized by follicle formation & lacks nuclear features of papillary carcinoma
Goes to extrathyroidal after =/> 20mm size and lungs after =/>30-40mm6.
- key parameters defined:
- encapsulation: refers to pathologist's estimate that the usually-completely (but rarely incompletely) encompassing fibrous rim is not a pseudocapsule (see above).
- capsule invasion: it is not invasion (it may be pseudoinvasion due to focal absence/incompleteness of capsule into which the nodule bulges) unless the diagnosing pathologist thinks it is invasion; & authorities tend to require that it be through-and-through invasion of capsule9...not just intrusion into (but not penetrating through) the fibrous peritumoral capsule (see [#1] below; this requires thorough sampling of the capsular zone.
- vascular invasion: it is not vascular invasion unless the diagnosing pathologist thinks it is vascular invasion. Authorities tend to require that a case be marked as positive for vascular invasion only (1) when tumor is in vessel within or beyond capsule (and with that penetrating twig of tumor being in continuity...skip foci don't count) and (2) tumor plugs free in a vascular space (not endothelialized or fibrin-coated or attached to vessel wall) either don't count or must be truly convincing.
- ordinary follicular carcinoma: encapsulation usually =/> 1.5mm.2 in thickness.
Four situational types:
- [#1] encapsulated & minimally capsular invasive (when invades into & does not penetrate through the capsule2), is called "follicular tumor of uncertain malignant potential"ED Williams, 2000, in 9.
- [#2] encapsulated & "minimally invasive" (when invades capsule but does not reach extrathyroidal tissue or reaches extrathyroidal but without vascular invasion being found9).
- [#3] encapsulated angioinvasive (an older study showed increased risk for early mets if 5 or more vessels invaded; 30-50% risk of distant mets at 10 years when vascular invasion2).
- [#4] widely invasive9: "To qualify as widely invasive, a tumor must show capsular and vascular invasion & penetrate beyond the confines of the thyroid proper". This kind may over-run and obliterate the tumor capsule.
- Hurthle cell thyroid carcinoma (HCC): see HCA, above; HCC only
3% of all types of thyroid cancers. Synonymes: oxyphil ca., oncocytic ca., eosinophilic cell ca., and Askenazy cell ca. Hurthle cell
cancer reportedly behaves in a more aggressive
fashion than other well-differentiated thyroid
cancers, with a tendency to higher incidence
of metastasis and a lower survival rate. This
is truer for the lesions that are clearly demonstrated
to be malignant and in patients who are considered
to be at high risk based on such factors as age,
tumor size, invasiveness, and the presence of
metastasis. Widely invasive tumors behave more
aggressively. Treatment likely to be total or
near-total thyroidectomy (by way of a second
surgery) and follow-up I131 ablation. And, since some go to nodes, may need node dissection2. When oncocytic but with less differentiated (follicular or papillary) histology, but not truly "poorly diferentiated", and considered an intermediate grade HCC; if poorly granulated small cells (& p53 positive), have worse behavior 12. Beware misdiagnosing HCA/HCC when it is really an oncocytic follicular variant of papillary thyroid ca. or an oncocytic tall-cell variant of papillary thyroid ca.
- Hyalinizing trabecular thyroid carcinoma (HTTC): see HTA, above. Some feel it is a variant of PTC1.
- clear cell carcinoma: see clear cell adenomata types, above .
- squamous cell carcinoma.
- mucoepidermoid carcinoma
- sclerosing mucoepidermoid with eosinophilia (sclerosing squamid tumor with eosinophilia): arise in Hashimoto's & striking infiltrate of eosinophils.
- mucinous carcinoma: lacks squamous component.
- carcinoma showing thymus-like epithelium (CASTLE): may be ectopic thymic ca.
- papillary thyroid carcinoma (PTC): Diagnosis depends on NUCLEAR cytopathology; about half have fibrosis & half have psamomma bodies1. Tend K903 positive1. Goes to nodes after =/> 5mm size and lungs after =/>20mm6. Even when less than 1mm in diameter, 6% have positive nodes13! Total thyroidectomy for bilateral (routine path exam finds 20% bilateral & other cancers have independent clonal origins6) & multifocal [L-04-10718,
bilat. & about 42 tumors] and I131 ablation; surgical reduction of residual thyroid that acts as a sink for the radioactive iodine & can deplete what would be going after metastatic cells;
then serial, periodic follow-up with serum thyroglobulin (TGB or Tg) testing [LMC-01-6488].
Some 19% of even the very small cancers fail I131 therapy & must be retreated.
When molecular testing on a thyroid neoplasm is positive for the BRAF V600E mutation, then the neoplasm is (1) PTC and (2) mutation-positives have a poorer prognosis [ARUP files].
- ordinary classical papillary: ovoid nuclei, overlapping nuclei, crowed & often non-polar nuclei, prominent nuclear
grooves (due to redundant nuclear membrane), nuclear chromatin clearing (don't be fooled by poor-fixation nuclear "bloating" or the clear nuclei at FS which are artifact), intra-nuclear cytoplasmic intrusive pseudo-inclusions (visible on frozen section and cytopreps), psamomma bodies (almost never in benign thyroid except HTA, above. About 45% have some focal squamous metaplasia3.
- classical papillary microcarcinoma: 1.0CM or less in maximum size; 50% have node mets at surgery6.
- occult papillary: so small as to be an incidental finding in thyroid tissue removed for other reasons...may be in as many as a third of glands
- Warthin's-like node mets: we have seen cystic & lymphoid-rich mets remindful of Warthin's tumor (L07-5276); so, think of met. PTC when think of Warthin's in
a tumor not within the interior of a salivary gland (and note this ENT caveat NOTE, above).
- encapsulated PTC: are hot on scan; produce pale vacuolated colloid; go to nodes & almost never distant mets or death1.
- well-differentiated tumor of uncertain malignant potential (WDT-UMP): may look follicular but has PTC cytopathology changes quantitatively unconvincing; small % go to nodes & almost never distant mets or death1.
- follicular variant (FVPTC): nuclear holes (cytoplasmic pseudo-inclusions) on FS [LMC-02-5355] & from no to partial to complete capsulation2.
- solid variant: particularly common in children.
- macrofollicular variant.
- diffuse (multinodular) variant.
- encapsulated follicular variant (Lindsay's tumor).
- diffuse sclerosing variant of PTC: dominant fibrosis & lots of psammoma bodies.
- oncocytic (oxyphilic) variant of PTC: a rare Hurthle cell or oncocytic variant of PTC
- tall cell or columnar cell variant of PTC: this variant has very scant...even absent...PTC-type nuclear changes & is of columnar-like cells (at least twice as tall as wide) with oxyphilic cytoplasm on papillations & tends to present in older patients and as a large tumor (>5cm.) & with extrathyroidal spread & vascular invasion and poorer prognosis [L16-3901]. Prognosis better when heavy lymphoplasmacytic infiltrate of tumor stroma11. Warthin's salivary gland tumor might provoke thoughts of TC-PTC [L07-5103; L11-9711].
- cribriform-morular variant of PTC:
- PTC with exuberant nodular fasciitis-like stroma.
- adenoacanthomatous FTC or PTC: when very extensive squamous metaplasia.
- Hybrid thyroid carcinoma (HTC): Combined features of FTC cancer and PTC2...mixed FTC & PTC [L08-1819]...turns out is considered a PTC by 201016.
- lymph node biopsy or dissection:
- benign inclusion: [LMC-02-82].
- mixed medullary & follicular.
- mixed medullary and papillary.
- small cell:
intermediate (poorly differentiated) thyroid carcinoma: otherwise known as an "insular carcinoma".
anaplastic thyroid carcinoma (ATC or UTC): can look sarcomatous on frozen section; may be giant cell type; can look sclerotic as if with a few large cells here & there but they are Ki67 pos. & ATC likely IHC neg. for thyroid markers; there is a clear-cell var. of ATC.
- calcitonin positive: probably true small-cell medullary.
- calcitonin negative: very agressive & probably poorest differentiated C-cell cancer (better considered to be an undifferentiated small cell ca.).
lymphoma: may arise in a Hashimoto background.
sarcoma: remember spindle-cell or giant-cell anaplastic carcinoma.
- squamous cell ca.: rare but quite malignant3 & likely an ATC16.
- Rosai J, Rosai AND Ackerman's Surgical Pathology, 9th Ed., p. 515-594, 2004.
- LiVolsi V, 9th Annual Seminar in Pathology, April/May 2002, Pittsburgh (EBS).
- Meissner WA & Warren S.,Thyroid Fascicle, (AFIP), 1969.
- Chan JKC, "Strict Criteria Should Be Applied in the Diagnosis of Encapsulated Follicular Variant of Papillary Thyroid Carcinoma", AJCP 117:16-18, 2002.
- Renshaw AA & Gould EW, "Why There Is the Tendency to 'Overdiagnose' the Follicular Variant of Papillary Thyroid Carcinoma", AJCP 117:19-21, 2002.
- Mazzaferri EL, "Managing Small Thyroid Cancers", JAMA 295(18):2179-2181, 2006.
- Veinot JP & Ghadially FN, "Melanosis Thyroidii", Ultrastructural Pathology 22:401-406, 1998.
- LiVolsi V, 13th Annual Seminar in Pathology, 6 May 2006, Pittsburgh (EBS).
- Saul Suster, "Thyroid Tumors With a Follicular Growth Pattern, Problems In Differential Diagnosis", Arch. Path. & Lab. Med. 130(7):984-988, July 2006.
- H. Gilbert Welch, M.D., M.P.H.,
Should I Be Tested for Cancer?
Maybe Not and Here's Why, March 2004.
- Ozaki O, et al, "Papillary Variant...", AJSP 20(6):695-698, 1996.
- Papotti M, et al, "Poorly Differentiated...", AJSP 20(6):686-694, 1996.
- Kuffner HA @ U. Pittsburgh, Oct. 2007 American Thyroid Association meeting in New York, abstract #184 presented 10/5/2007 (MedScape Medical News).
- Cibas E.S. and Ali S.Z. The Bethesda System for Reporting Thyroid
Cytopathology. Am J Clin Pathol 2009;132:658-665.
- Bethesda criteria posted by Rex Pathology Lab HERE.
- Saul Suster, "Thyroid Pathology Session 11/15", "Oncologic Pathology: A Review and Update of Current Diagnostic Problems", ASCP CME in Charleston, S. C. , 15-17 Nov. 2010.
- Veinot JP & Ghadially FN., "Melanosis thyroidi", Ultrastruct Pathol., 22(5):401-6, Sept-Oct. 1998. HERE.
- Singh K, et. al., "Melanotic medullary carcinoma of thyroid – report of a rare case with brief review of literature", Diagnostic Pathology, 3:2, 11 January 2008, HERE.
(posted 9 Feb. 2002; latest addition 10 April