Thyroid Nodule Evaluation: TI-RADS, FNA, and Bethesda Classification

Thyroid Nodule Evaluation: TI-RADS, FNA, and Bethesda Classification

Evidence-based guideline for the workup of thyroid nodules — from ACR TI-RADS sonographic risk stratification through fine-needle aspiration biopsy, Bethesda cytology interpretation, molecular testing, and surgical decision-making.

Otolaryngology Endocrine Surgery Head & Neck Ultrasound Author: gomdori

Executive Summary

Thyroid nodules are extraordinarily common — palpable in 5% of women and 1% of men, detected sonographically in 50–60% of adults — but only 5–10% harbor malignancy. The evaluation framework integrates clinical history, TSH measurement, neck ultrasound with ACR TI-RADS scoring, and selective fine-needle aspiration (FNA). FNA decisions follow size-based thresholds dependent on TI-RADS category: TR5 ≥1.0 cm, TR4 ≥1.5 cm, TR3 ≥2.5 cm. Cytology is reported using the Bethesda System, with malignancy risk ranging from 0–3% (Bethesda II benign) to 97–99% (Bethesda VI malignant). Indeterminate cytology (Bethesda III, IV) often benefits from molecular testing (Afirma GSC, ThyroSeq) to refine surgical decision-making and avoid diagnostic-only lobectomy. Surgical options range from active surveillance (microcarcinomas), thyroid lobectomy (low-risk single lesions), to total thyroidectomy with central neck dissection (high-risk or multifocal disease). Modern overdiagnosis concerns have shifted toward more conservative management of subcentimeter incidentalomas and selected papillary microcarcinomas.

50–60% Sonographic prevalence in adults
5–10% Malignancy rate
85% Papillary thyroid cancer (most common)
≥1.0 cm FNA threshold for TR5 nodules

S1. Epidemiology and Initial Evaluation

Thyroid nodules are among the most common clinical findings in adult primary care and otolaryngology practice. Palpable nodules are present in approximately 5% of women and 1% of men, while high-resolution ultrasound detects nodules in 50–60% of adults — the prevalence rising with age, female sex, iodine deficiency, and radiation exposure history. Despite this remarkable frequency, only 5–10% of nodules are malignant. The diagnostic and management challenge is identifying the small subset of clinically significant cancers while minimizing overdiagnosis, unnecessary surgery, and patient anxiety.

Etiology of thyroid nodules

TypeFrequencyNotes
Colloid (adenomatous) nodule50–60%Most common; benign hyperplastic process
Cystic / mixed cystic-solid15–20%Pure cysts almost always benign; mixed lesions evaluated by solid component
Follicular adenoma10–15%Benign; cytologically indistinguishable from follicular carcinoma without histology
Hashimoto’s thyroiditis with nodularity10–15%Background autoimmune thyroiditis with regenerative nodules
Papillary thyroid carcinoma3–7%~85% of all thyroid cancers; well-differentiated; excellent prognosis
Follicular thyroid carcinoma1–2%~10% of thyroid cancers; hematogenous spread to bone/lung
Medullary thyroid carcinoma1–2%C-cell origin; calcitonin marker; MEN2 association
Anaplastic thyroid carcinoma<1%Rare but lethal; rapidly enlarging mass; median survival <6 months
Lymphoma (primary thyroid)<1%Rare; arises in Hashimoto’s background
Metastasis (renal, lung, breast)<1%Rare; consider in known cancer with new nodule

Risk factors for malignancy

  • History of head/neck radiation (especially childhood) — strongest acquired risk factor; childhood cancer survivors highest risk
  • Family history — papillary carcinoma in first-degree relatives, MEN2 syndromes (RET mutations), familial adenomatous polyposis (cribriform-morular variant), Cowden syndrome (PTEN), DICER1 syndrome
  • Age extremes — children <20 years (lower nodule prevalence but higher malignancy rate) and adults >60 (more aggressive cancers when present)
  • Male sex — although nodules less common, more likely malignant when present
  • Rapid growth — clinically significant change over weeks-to-months
  • Hoarseness or vocal changes — may indicate recurrent laryngeal nerve involvement
  • Dysphagia or dyspnea — extrathyroidal extension or large goiter
  • Cervical lymphadenopathy — particularly central compartment or jugular chain
  • Fixation to surrounding tissues — hard nodule that does not move with swallowing

Initial workup

TestPurposeIndications
TSHDetermine thyroid function statusAll patients with thyroid nodule
Free T4 / Free T3Confirm hyper- or hypothyroidism if TSH abnormalAbnormal TSH
Anti-TPO, anti-TG antibodiesDiagnose Hashimoto’s thyroiditisHypothyroidism, multinodular gland, suspected autoimmune background
CalcitoninScreen for medullary thyroid carcinomaRoutine use controversial; recommended if family history MEN2, multinodular goiter undergoing surgery
Thyroid ultrasound (high-frequency probe)Characterize nodule(s), assess lymph nodes, plan FNAAll thyroid nodules
I-123 thyroid scintigraphyDistinguish hot vs cold nodule when TSH suppressedSubnormal TSH only — not for euthyroid or hypothyroid
Cross-sectional imaging (CT/MRI)Substernal extension, retropharyngeal, large goiterSelected cases; avoid iodinated contrast if RAI planned
FNA biopsyHistologic diagnosisPer TI-RADS size thresholds (S2)

Algorithm — initial workup

Thyroid Nodule Initial Workup Algorithm Decision tree from initial discovery through TSH, ultrasound, and FNA biopsy. Thyroid Nodule Workup Algorithm Thyroid nodule detected Measure TSH + Ultrasound TSH LOW → I-123 scintigraphy TSH NORMAL/HIGH → Apply TI-RADS, FNA per size “Hot” nodule Rarely malignant (1–4%) Manage hyperfunction “Cold” nodule Apply TI-RADS pathway FNA per size threshold FNA based on TI-RADS TR5≥1.0 cm, TR4≥1.5, TR3≥2.5 → Bethesda → Surgery vs surveillance
Thyroid Nodule — Thyroid Anatomy Thyroid gland anatomy with nodule locations Thyroid Nodule — Thyroid Anatomy Trachea Right Lobe Left Lobe Isthmus Bethesda System (FNA Cytology) Bethesda I: Non-diagnostic Bethesda II: Benign Bethesda III: AUS/FLUS Bethesda IV: Follicular neoplasm Bethesda V: Suspicious Bethesda VI: Malignant
Thyroid Nodule — Thyroid Anatomy

S2. ACR TI-RADS Sonographic Risk Stratification

The American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS), introduced in 2017, standardized sonographic risk stratification of thyroid nodules. It assigns points across 5 sonographic categories — composition, echogenicity, shape, margin, and echogenic foci — and totals them to derive a TI-RADS level (TR1 through TR5). The TR level combined with nodule size determines whether FNA, follow-up ultrasound, or no action is recommended.

The 5 sonographic categories — point assignment

CategoryFindingPoints
CompositionCystic or almost completely cystic0
Spongiform0
Mixed cystic and solid1
Solid or almost completely solid2
EchogenicityAnechoic0
Hyperechoic or isoechoic1
Hypoechoic2
Very hypoechoic3
Cannot determine1 (default)
ShapeWider-than-tall0
Taller-than-wide (anteroposterior > transverse)3
MarginSmooth0
Ill-defined0
Lobulated or irregular2
Extrathyroidal extension3
Echogenic fociNone or large comet-tail artifacts0
Macrocalcifications1
Peripheral (rim) calcifications2
Punctate echogenic foci (microcalcifications)3

TR level assignment and FNA thresholds

TR LevelTotal PointsCategoryFNA ThresholdFollow-up
TR10BenignNoneNone required
TR22Not suspiciousNoneNone required
TR33Mildly suspicious≥2.5 cmFollow at 1, 3, 5 years if <2.5 cm
TR44–6Moderately suspicious≥1.5 cmFollow at 1, 2, 3, 5 years if <1.5 cm
TR5≥7Highly suspicious≥1.0 cmFollow annually × 5 years if <1.0 cm

Suspicious lymph nodes warrant FNA at any size

Any cervical lymph node with sonographic features suggestive of metastasis (cystic change, microcalcifications, hypervascularity, loss of fatty hilum, hyperechoic foci, rounded shape) should undergo FNA regardless of size — especially in a patient with a suspicious thyroid nodule. FNA of the suspicious node and washout for thyroglobulin (Tg) measurement greatly enhances sensitivity for metastasis detection. Tg in the FNA washout fluid is highly specific for thyroid origin even in cytologically equivocal samples.

High-risk sonographic features (memory aids)

FeatureWhy it matters
Microcalcifications (punctate echogenic foci)Psammoma bodies of papillary carcinoma; specificity 80–95%
Markedly hypoechoic (darker than strap muscles)Cellular density of tumor; specificity 75–90%
Taller-than-wide (AP > transverse)Growth pattern across normal tissue planes; specificity 85–95%
Irregular/lobulated/spiculated marginInfiltrative growth; specificity 75–90%
Extrathyroidal extensionCapsular breach; high specificity for malignancy
Suspicious lymphadenopathy (cystic, microcalcifications, vascular)Metastatic disease until proven otherwise

Low-risk / benign features

  • Pure cyst (anechoic, no solid component) — <1% malignancy
  • Spongiform appearance (≥50% small cystic spaces) — <3% malignancy
  • “Hot” nodule on radioiodine scan — 1–4% malignancy
  • Background of multinodular goiter (lower risk per individual nodule, but sample largest/most suspicious)
  • Comet-tail artifact (large) — colloid; benign
Thyroid Nodule — Comparison Comparison table: Thyroid Nodule — Comparison TR1 TR2 TR3 TR4 TR5 Description Benign (0 pts); no FNA needed Not suspicious (2 pts); no FNA Mildly suspicious (3 pts); FNA if ≥2.5cm Moderately suspicious (4-6 pts); FNA if ≥1.5cm Highly suspicious (≥7 pts); FNA if ≥1cm
Thyroid Nodule — Comparison

S3. FNA Biopsy and Bethesda Classification

Fine-needle aspiration technique

Ultrasound-guided FNA is the gold standard for cytologic diagnosis. Free-hand palpation-guided FNA is acceptable for large palpable nodules but has higher non-diagnostic rates. Standard technique:

  • 25–27 gauge needle (smaller gauge = less blood contamination, fewer artifacts)
  • 2–6 passes per nodule with capillary technique or syringe-aspiration; smear preparation per local pathology preference
  • Rapid on-site evaluation (ROSE) by cytopathology improves adequacy rates
  • Sample multiple areas of the nodule (especially solid components of mixed lesions)
  • Local anesthesia not routinely required; minimal patient discomfort
  • Complications rare: hematoma (small, self-limited), infection (extremely rare), transient pain, vasovagal response
  • Hold antiplatelet/anticoagulant per local protocol (often not required given small needle gauge)

The Bethesda System for Reporting Thyroid Cytopathology

CategoryDiagnosisRisk of MalignancyManagement
INon-diagnostic / unsatisfactory5–10%Repeat FNA with US guidance in 4–12 weeks; consider on-site cytology
IIBenign (colloid nodule, Hashimoto’s, etc.)0–3%Sonographic surveillance per TI-RADS schedule
IIIAtypia of undetermined significance / FLUS10–30%Repeat FNA, molecular testing, or diagnostic lobectomy
IVFollicular neoplasm / suspicious for FN25–40%Molecular testing or diagnostic lobectomy
VSuspicious for malignancy50–75%Lobectomy or near-total thyroidectomy
VIMalignant97–99%Lobectomy or total thyroidectomy ± central neck dissection

Detail on each Bethesda category

Bethesda I — Non-diagnostic

Insufficient cellularity (typically <6 groups of well-preserved follicular cells). Causes: cyst fluid only, blood-only aspirate, scant cellularity. Repeat FNA in 4–12 weeks. If second non-diagnostic, consider diagnostic lobectomy especially if sonographically suspicious. ~50% become diagnostic on repeat.

Bethesda II — Benign

Adequate sample with benign cytology (colloid nodule, Hashimoto’s, granulomatous thyroiditis, cyst). Approximately 70% of FNA results. Sonographic surveillance per TI-RADS schedule. Repeat FNA if growth (≥20% in 2 dimensions) or new suspicious features.

Bethesda III — Atypia of Undetermined Significance / FLUS

Indeterminate category with cytologic atypia not meeting threshold for follicular neoplasm or suspicion for malignancy. Heterogeneous category. Options: repeat FNA (40–60% benign on repeat), molecular testing (Afirma GSC negative result allows surveillance), or diagnostic lobectomy.

Bethesda IV — Follicular Neoplasm / Suspicious for FN

Microfollicular pattern; cannot distinguish follicular adenoma from follicular carcinoma without histologic evaluation of capsular and vascular invasion. ~25–40% are malignant. Options: molecular testing or diagnostic lobectomy. The Hürthle cell variant (now called oncocytic neoplasm) carries similar management.

Bethesda V — Suspicious for Malignancy

Cytologic features suggestive but not diagnostic of malignancy (e.g., suspicious for papillary carcinoma but lacking 2 of 3 cardinal features: nuclear grooves, intranuclear pseudoinclusions, optically clear nuclei). Surgical management: lobectomy if low-risk, total thyroidectomy if high-risk features.

Bethesda VI — Malignant

Diagnostic of papillary carcinoma, medullary carcinoma, anaplastic carcinoma, lymphoma, or metastasis. Surgical management based on tumor type, size, extent, and ATA risk stratification.

When to refer to endocrine surgery / otolaryngology

Refer to endocrine surgery or head and neck oncologic surgery for: Bethesda V or VI cytology; Bethesda III/IV with positive molecular testing; suspicious lymphadenopathy on imaging; nodule >4 cm regardless of cytology (lobectomy considered due to false-negative FNA rates); compressive symptoms (dysphagia, dyspnea, hoarseness); substernal extension; recurrent nodule after prior surgery; high suspicion despite benign cytology (clinical-cytologic discordance). High-volume surgeons (>25 thyroid surgeries/year) achieve significantly better outcomes including lower rates of recurrent laryngeal nerve injury and hypoparathyroidism.

Thyroid Nodule — Procedure Steps Surgical stepper: Thyroid Nodule — Procedure Steps Thyroid Nodule — Procedure Steps 1 Initial workup TSH; thyroid US with TI-RADS scoring; calcitonin if MTC suspected 2 FNA criteria (ACR TI-RADS) TR3 ≥2.5cm; TR4 ≥1.5cm; TR5 ≥1cm; consider lower threshold for cervical LN abnormality 3 Bethesda I-II I (non-diagnostic): repeat FNA; II (benign): observation, repeat US in 1-2y 4 Bethesda III-IV III (atypia of undetermined significance, AUS): repeat FNA OR molecular testing (Afirma, ThyroSeq); 5 Bethesda V-VI V (suspicious for malignancy): lobectomy or total thyroidectomy; VI (malignant): surgery 6 Surveillance Benign nodule: US at 1-2y, then 3-5y if stable; growth >50% volume = repeat FNA
Thyroid Nodule — Procedure Steps

S4. Molecular Testing and Surgical Decision-Making

Molecular testing for indeterminate cytology

Bethesda III (AUS/FLUS) and IV (follicular neoplasm) cytology results have intermediate malignancy risk (10–40%) — too high to ignore, too low to mandate surgery in all patients. Historically, diagnostic lobectomy was performed for all such patients, with 60–80% revealing benign histology — meaning many patients underwent unnecessary surgery. Molecular testing has substantially refined this decision-making.

TestMethodologyPerformanceUse
Afirma Genomic Sequencing Classifier (GSC)RNA-seq gene expression classifier; 10,196 genes analyzed; benign/suspicious binary callNPV ~96%, PPV ~50% in Bethesda III/IVRule-out test; benign result allows surveillance
ThyroSeq v3NGS-based mutation + gene fusion + CNV + gene expression panel; 112 genesNPV ~97%, PPV ~66%Both rule-out and rule-in; identifies actionable mutations (BRAF, RAS, RET/PTC, PAX8/PPARG)
RosettaGX RevealmicroRNA classifier (24 miRNAs)NPV ~92%, PPV ~50%Used on cytology smears (does not require fresh tissue)
BRAF V600E mutationTargeted PCR/NGS for single mutation~99% PPV for papillary carcinoma when positive; low sensitivityConfirmatory rule-in test; predicts more aggressive behavior

Insurance and cost considerations

Molecular testing costs $3000–5000 per specimen. Most commercial insurers and Medicare cover Afirma GSC and ThyroSeq for Bethesda III/IV with appropriate documentation. Cost-effectiveness analyses generally favor molecular testing when avoidance of unnecessary surgery is factored in. Discussion with patient should include test characteristics, downstream implications of results, and out-of-pocket cost.

Surgical decision-making — lobectomy vs total thyroidectomy

The 2015 ATA guidelines significantly broadened indications for thyroid lobectomy (vs total thyroidectomy) in low-risk thyroid cancer. Key principles:

ScenarioRecommended Surgery
Bethesda VI papillary carcinoma, <1 cm, no extrathyroidal extension, no lymph nodes, no high-risk featuresLobectomy or active surveillance (selected cases)
Bethesda VI papillary carcinoma, 1–4 cm, intrathyroidal, no nodes, no high-risk featuresLobectomy (preferred) or total thyroidectomy
Bethesda VI, >4 cm OR extrathyroidal extension OR clinical N1 diseaseTotal thyroidectomy ± central neck dissection
Bethesda VI with distant metastases, high-grade histology, or anaplastic featuresTotal thyroidectomy + therapeutic lymph node dissection + RAI
Medullary thyroid carcinomaTotal thyroidectomy + central neck dissection (and lateral if N1)
Bethesda V (suspicious)Lobectomy first; complete to total if histology malignant and high-risk
Bethesda IV with molecular high-risk mutation (BRAF, TERT)Lobectomy if intrathyroidal <4 cm; consider total thyroidectomy if extensive
Bethesda IV with molecular low-risk (RAS) or benign resultLobectomy; surveillance after benign histology

Active surveillance for papillary microcarcinoma

Pioneered at Kuma Hospital (Japan) and validated at Memorial Sloan Kettering, active surveillance (AS) is an alternative to immediate surgery for selected low-risk papillary microcarcinomas (≤1 cm, no extrathyroidal extension, no lymph node involvement, no aggressive cytology features). Outcomes from large series:

  • Tumor growth (≥3 mm in any dimension): 5–10% over 5 years
  • New lymph node metastasis: 1–4% over 5 years
  • Disease-specific survival: ~100% even when surgery is delayed
  • Patients can transition to surgery at any time without compromised outcomes
  • Avoid surveillance in young patients (<40 years), pregnant women, those with Hashimoto’s background, family history of thyroid cancer, or extrathyroidal location near critical structures

Surgical complications

ComplicationLobectomyTotal thyroidectomy
Recurrent laryngeal nerve injury (transient)1–3%3–5%
Recurrent laryngeal nerve injury (permanent)<0.5%1–2%
External branch superior laryngeal nerve injury<1%1–2%
Transient hypoparathyroidism0% (only one lobe removed)20–30%
Permanent hypoparathyroidism0%1–4%
Hematoma requiring re-operation<1%1–2%
Infection<1%<1%
Need for thyroid hormone replacement20–30% (TSH >5 post-op)100% (lifelong)
Rates lower at high-volume centers (>25 cases/year)SignificantlySignificantly

S5. Special Situations and Surveillance

Pregnancy

Thyroid nodules detected during pregnancy require special considerations. TSH and ultrasound are safe; FNA is safe and preferred over delaying diagnosis. Radioiodine scintigraphy is contraindicated. If FNA reveals papillary carcinoma:

  • Most well-differentiated thyroid cancers grow slowly and surgery can typically be deferred to postpartum without compromising outcomes
  • If surgery is required during pregnancy, the second trimester is safest
  • Levothyroxine suppression therapy is not contraindicated (continue if already on it; initiate at standard pregnancy dose)
  • Postpartum reassessment with ultrasound and decision regarding surgery

Pediatric thyroid nodules

Children with thyroid nodules have a much higher rate of malignancy (~25%) compared to adults (5–10%). The ATA published separate pediatric guidelines (2015) emphasizing earlier and more aggressive workup. Key differences from adult management:

  • FNA threshold lowered: any solid hypoechoic nodule, any nodule with high-risk features regardless of size
  • Use ultrasound-guided FNA (palpation alone less reliable in small necks)
  • Higher rate of multifocality and lymph node metastases
  • Total thyroidectomy with central neck dissection often preferred over lobectomy due to higher recurrence risk
  • RAI ablation in selected cases despite concerns about secondary malignancy in children
  • Long-term surveillance for recurrence and treatment-related complications

Multinodular goiter

For patients with multiple nodules, FNA decisions should be guided by sonographic risk stratification of each nodule, not by overall gland size. Sample the most suspicious nodule(s) regardless of size. Avoid sampling every nodule — this unnecessarily increases anxiety and overdiagnosis. Surveillance is per the highest-risk nodule.

Family history of thyroid cancer

Familial papillary thyroid carcinoma (3+ first-degree relatives) accounts for 3–5% of papillary carcinomas. Other inherited syndromes:

  • MEN2A/2B: medullary thyroid carcinoma; RET mutation; consider prophylactic thyroidectomy in early childhood (before age 5 in MEN2A, before age 1 in MEN2B)
  • FAP (familial adenomatous polyposis): cribriform-morular variant of papillary carcinoma; APC mutation
  • Cowden syndrome: PTEN mutation; multinodular goiter with malignancy risk
  • Carney complex: PRKAR1A mutation; thyroid follicular adenomas/carcinomas
  • DICER1 syndrome: multinodular goiter and rare thyroid cancers

Surveillance schedules for benign nodules

Initial findingSurveillance scheduleWhen to repeat FNA
TR1, TR2 noduleNone requiredOnly if new clinical finding
TR3 nodule <2.5 cmUS at 1, 3, 5 yearsIf growth ≥20% in 2 dimensions or new suspicious feature
TR4 nodule <1.5 cmUS at 1, 2, 3, 5 yearsIf growth or new suspicious feature
TR5 nodule <1 cmAnnual US × 5 yearsIf growth or new suspicious feature
Bethesda II benign FNAPer TI-RADS scheduleIf growth ≥20% in 2 dimensions or new suspicious feature
Bethesda I (non-diagnostic)Repeat FNA in 4–12 weeks; lobectomy if persistently non-diagnostic with suspicious features—

Beware “thyroid cancer epidemic” — overdiagnosis

The dramatic 3-fold rise in thyroid cancer incidence over the past 30 years has been almost entirely driven by detection of small papillary microcarcinomas (≤1 cm) on imaging — without a corresponding rise in mortality. This is a textbook overdiagnosis phenomenon. The implication: be judicious about ultrasound use in low-risk patients, follow TI-RADS-based FNA thresholds rigorously, embrace active surveillance for selected microcarcinomas, and counsel patients honestly about the indolent nature of most thyroid cancers. Not every small thyroid cancer requires aggressive surgical intervention.

Practice Insights: TI-RADS Size Thresholds, Molecular Testing for Indeterminate, Active Surveillance for Microcarcinoma

(The content below synthesizes published clinical patterns and expert consensus rather than first-person case reports.)After managing thyroid nodules across endocrinology clinic, ultrasound-guided biopsy service, and a multidisciplinary thyroid cancer clinic over many years and roughly several hundred biopsied nodules, the central conviction is that in experienced hands the field has shifted away from biopsy-everything-that-looks-suspicious toward TI-RADS size-threshold-based selection, molecular testing for indeterminate cytology, and active surveillance for low-risk papillary microcarcinoma. In experienced practice the patients who do worst are not those with missed cancer; they are those who underwent diagnostic lobectomy for Bethesda III-IV cytology that subsequent molecular testing would have shown was benign — they live with thyroid hormone replacement and surgical scar for an entirely avoidable workup.

ACR TI-RADS Size Thresholds Anchor FNA Decision

in experienced hands, the FNA decision rests on ACR TI-RADS classification combined with nodule size thresholds. TR1 (benign): no FNA. TR2 (not suspicious): no FNA. TR3 (mildly suspicious): FNA if ≥2.5 cm; surveillance if 1.5-2.4 cm. TR4 (moderately suspicious): FNA if ≥1.5 cm; surveillance if 1.0-1.4 cm. TR5 (highly suspicious): FNA if ≥1.0 cm; surveillance if 0.5-0.9 cm. Below these thresholds, surveillance with serial ultrasound (12-24 months) rather than biopsy. This single framework avoids the over-biopsy era of “any nodule over 1 cm gets stuck” and reduces unnecessary procedures, with the trade-off of accepting some delayed diagnosis of low-risk microcarcinomas that have favorable natural history regardless.

Molecular Testing Rescues Indeterminate Cytology

For Bethesda III (AUS/FLUS) and Bethesda IV (FN) cytology, the established reflex is molecular testing before recommending diagnostic lobectomy. Afirma GSC (gene sequencing classifier) — high negative predictive value (~96%), benign call reliably rules out malignancy and avoids surgery. ThyroSeq v3 (next-generation sequencing panel) — both rule-in and rule-out capability via gene mutations + fusions + copy number. ThyGeNext + ThyraMIR — combined mutation + microRNA testing. Molecular testing reduces diagnostic surgery rates from roughly 75% (historical) to 30-40% for Bethesda III, and roughly 95% to 50% for Bethesda IV. The cost is offset by avoided unnecessary surgeries and surgical complications.

Active Surveillance for Selected Low-Risk Papillary Microcarcinoma

For papillary thyroid microcarcinoma (≤1 cm) without high-risk features (no extrathyroidal extension, no lymph node metastasis, not posterior subcapsular near recurrent laryngeal nerve, not adjacent to trachea), active surveillance is emerging as an alternative to immediate surgery. Kuma Hospital and Memorial Sloan Kettering data: roughly 90% of these microcarcinomas remain stable or grow <3 mm over 10 years; the small percentage that progress are still curable with delayed surgery without compromising long-term survival. the patients selection: age 60+ (more conservative below this), small tumor (≤1 cm), no high-risk imaging features, patient understanding and acceptance, reliable follow-up. Serial ultrasound every 6 months × 2 years, then annually.

Bethesda Classification Drives Next Steps

The Bethesda cytology classification translates the FNA result into a treatment recommendation: I (non-diagnostic, 5-10%) — repeat FNA within 3 months. II (benign, 60-70%) — surveillance ultrasound 12-24 months. III (AUS/FLUS, 5-15%) — molecular testing first; if benign call → surveillance; if suspicious → repeat FNA or lobectomy. IV (FN, 5-10%) — molecular testing; if benign → surveillance; if suspicious → diagnostic lobectomy. V (suspicious for malignancy, 3-5%) — total thyroidectomy or lobectomy depending on risk and patient preference. VI (malignant, 5%) — total thyroidectomy ± central neck dissection.

Illustrative Scenario

A patient in their forties was found to have a 2.1 cm thyroid nodule on screening neck ultrasound. ACR TI-RADS scoring: solid composition (2), hypoechoic (2), wider-than-tall (0), smooth margin (0), no echogenic foci (0) — total 4 points, TR4 (moderately suspicious). Size 2.1 cm exceeds 1.5 cm FNA threshold for TR4, so FNA performed. Cytology: Bethesda III (AUS/FLUS) — atypical follicular cells with mild architectural and nuclear atypia. Rather than proceeding to diagnostic lobectomy, the treating clinicians sent ThyroSeq v3 molecular testing on the FNA sample — returned benign with no driver mutations. Patient avoided surgery and was placed on surveillance ultrasound at 12 months. Two-year follow-up showed stable nodule with no change in size or sonographic features.

Takeaway: Bethesda III + molecular testing benign call = surveillance, not surgery. The era of “diagnostic lobectomy for all indeterminate cytology” is over. Molecular testing avoids approximately 50-60% of these unnecessary surgeries with high negative predictive value, and the patient avoids permanent thyroid hormone replacement plus surgical scar.

Practice Pattern Refinements vs Standard Algorithm

  1. ACR TI-RADS size thresholds anchor FNA selection. TR3 ≥2.5 cm, TR4 ≥1.5 cm, TR5 ≥1.0 cm trigger biopsy. Below thresholds: surveillance, not biopsy. Avoids over-biopsy era.
  2. Molecular testing reflex for Bethesda III and IV cytology. Afirma GSC, ThyroSeq v3, or ThyGeNext+ThyraMIR. Reduces diagnostic surgery rates from ~75% to ~30-40% (Bethesda III) and from ~95% to ~50% (Bethesda IV). Documented benign call → surveillance.
  3. Active surveillance for selected low-risk papillary microcarcinoma. ≤1 cm, no high-risk features, age ≥60 (preferred), reliable follow-up, patient acceptance. Serial ultrasound q6 mo × 2 years, then annually. Convert to surgery if growth >3 mm, new high-risk feature, or patient preference change.
  4. Total thyroidectomy not always required for low-risk PTC. Lobectomy is acceptable for differentiated thyroid cancer ≤4 cm, intrathyroidal, no extrathyroidal extension, no clinical lymph node involvement (per ATA 2015 guidelines). Avoids lifelong thyroid hormone replacement in this subgroup.
Thyroid Nodule Pathway — TI-RADS → FNA → Bethesda → Molecular Complete pathway from ultrasound classification through FNA decision to cytology-based management with molecular rescue. Thyroid Nodule — TI-RADS → FNA → Bethesda → Molecular Size threshold gates FNA · Molecular testing rescues indeterminate TR1-2 Benign / not suspicious No FNA Surveillance US every 2-3 years until stable TR3 Mildly suspicious FNA if: ≥2.5 cm Surveillance: 1.5-2.4 cm US q1-2 y TR4 Moderately suspicious FNA if: ≥1.5 cm Surveillance: 1.0-1.4 cm US annually TR5 Highly suspicious FNA if: ≥1.0 cm Surveillance: 0.5-0.9 cm US q6-12 mo FNA → Bethesda category I (5-10%): repeat FNA in 3 mo II (60-70%): benign → surveillance III (5-15% AUS): molecular test IV (5-10% FN): molecular test V (3-5% suspicious): surgery VI (5% malignant): total TT Molecular testing options ▸ Afirma GSC (high NPV ~96%) ▸ ThyroSeq v3 (rule in/out) Molecular testing impact (Bethesda III/IV) ▸ Diagnostic surgery rate: ~75% → ~30-40% (Bethesda III) ▸ Diagnostic surgery rate: ~95% → ~50% (Bethesda IV) ▸ Benign call → surveillance instead of lobectomy Active surveillance criteria (low-risk PTC) ▸ Papillary microcarcinoma ≤1 cm, no high-risk features ▸ Age ≥60 preferred, reliable follow-up, patient acceptance ▸ US q6 mo × 2 yr → annually · ~90% stable at 10 years Surgical decision (Bethesda V/VI) ▸ Lobectomy acceptable for differentiated TC ≤4 cm, intrathyroidal, no LN involvement (ATA 2015) ▸ Total thyroidectomy for >4 cm, ETE, LN+, or patient preference for ablation strategy

S6. Pearls, Pitfalls, and FAQ

Pearl 1 — Order TSH first, then ultrasound

Always check TSH before ordering thyroid scintigraphy. If TSH is suppressed, the nodule may be autonomously hyperfunctioning (“hot”), with malignancy risk only 1–4% — FNA is generally not needed. Scintigraphy with euthyroid TSH is wasteful and provides no actionable information beyond what ultrasound already shows. The TSH check costs $10 and saves both unnecessary radiation and unnecessary FNAs.

Pitfall 1 — Biopsying every nodule in a multinodular goiter

The instinct to “FNA everything” in a multinodular goiter is wrong. Apply TI-RADS to each nodule individually and biopsy only those meeting size thresholds for their TR level — typically 1 to 3 nodules out of 10+. Biopsying small benign-appearing nodules increases procedure burden, anxiety, and risk of indeterminate cytology requiring unnecessary surgery. Sample the worst-looking nodule(s), not all of them.

Pearl 2 — Molecular testing rescues many indeterminate cytologies

For Bethesda III (AUS/FLUS) and IV (follicular neoplasm), molecular testing (Afirma GSC, ThyroSeq) can avoid 50–60% of diagnostic lobectomies. The negative predictive value is ~96% — a benign molecular result effectively reclassifies the nodule as benign with high confidence, allowing sonographic surveillance instead of surgery. Counsel patients on the test, cost, and downstream implications. Most insurance covers it for appropriate Bethesda categories.

Pitfall 2 — Reflexively performing total thyroidectomy for low-risk cancer

The 2015 ATA guidelines significantly liberalized lobectomy for differentiated thyroid cancer ≤4 cm without extrathyroidal extension or lymph node involvement. Total thyroidectomy commits a patient to lifelong levothyroxine, twice the surgical complication risk, and no clear oncologic advantage. For low-risk cases, lobectomy provides equivalent oncologic outcomes with substantially lower morbidity. Refer to a high-volume surgeon and follow guideline-recommended extent of surgery.

Pearl 3 — Active surveillance is reasonable for papillary microcarcinoma

Selected papillary microcarcinomas (≤1 cm, no extrathyroidal extension, no lymph node disease, away from critical structures) can be safely managed with active surveillance based on Japanese (Kuma) and US (MSK) experience. Tumor growth is slow (5–10% in 5 years), new lymph node metastases rare (1–4%), and surgery can be performed later if needed without compromised outcomes. This avoids surgical morbidity in a population with extremely indolent disease. Selection requires careful informed consent and ongoing US surveillance.

Pitfall 3 — Missing a suspicious lymph node

The ultrasound report should always document the central and lateral cervical lymph nodes alongside the thyroid nodule itself. Suspicious lymph nodes (cystic change, microcalcifications, hypervascularity, loss of fatty hilum) warrant FNA at any size — including thyroglobulin washout for highest sensitivity. A “low-risk” appearing thyroid nodule with a suspicious lymph node is a high-risk presentation requiring urgent evaluation and likely total thyroidectomy with neck dissection. Always look at the nodes, not just the nodule.

Frequently Asked Questions

When should a thyroid nodule be biopsied?
FNA biopsy is recommended based on ACR TI-RADS score and nodule size: TR5 (highly suspicious) ≥1.0 cm; TR4 (moderately suspicious) ≥1.5 cm; TR3 (mildly suspicious) ≥2.5 cm; TR1 and TR2 (benign or not suspicious) generally do not require FNA. Suspicious lymph nodes warrant FNA at any size. Smaller nodules (≤1 cm) with very high suspicion features (extrathyroidal extension, suspicious lymphadenopathy) may also be biopsied.
What is the Bethesda classification?
The Bethesda System for Reporting Thyroid Cytopathology classifies FNA results into 6 categories: I (non-diagnostic, ~5–10% repeat FNA), II (benign, 0–3% malignancy risk), III (AUS/FLUS, 10–30% risk), IV (follicular neoplasm, 25–40% risk), V (suspicious for malignancy, 50–75% risk), VI (malignant, 97–99% risk). Categories III and IV often warrant molecular testing (Afirma GSC, ThyroSeq) before deciding between repeat FNA, lobectomy, or surveillance.
What ultrasound features suggest malignancy?
ACR TI-RADS scores 5 sonographic categories: composition (cystic→spongiform→mixed→solid), echogenicity (anechoic→hyperechoic→isoechoic→hypoechoic→very hypoechoic), shape (taller-than-wide), margin (smooth→ill-defined→lobulated→extrathyroidal extension), and echogenic foci (large comet-tail→macrocalcifications→peripheral→punctate). Higher composite scores correlate with malignancy risk. Punctate echogenic foci (microcalcifications), markedly hypoechoic, taller-than-wide shape, and irregular/extrathyroidal margin are the most concerning.
What is the role of TSH and thyroid scintigraphy?
Initial evaluation includes TSH measurement. If TSH is low (suggesting hyperfunction), proceed to radioiodine (I-123) scintigraphy. A ‘hot’ nodule (concentrating tracer more than surrounding parenchyma) is rarely malignant (1–4%) and FNA is not required — manage as autonomously functioning nodule with thionamides, RAI ablation, or surgery. A ‘cold’ nodule (no tracer uptake) follows the standard ultrasound risk stratification and FNA pathway. If TSH is normal or elevated, proceed directly to ultrasound without scintigraphy.
Is molecular testing useful for indeterminate cytology?
Yes. For Bethesda III (AUS/FLUS) and IV (follicular neoplasm) cytology, molecular testing significantly improves diagnostic accuracy. The Afirma Genomic Sequencing Classifier (GSC) has high negative predictive value (~96%) — a benign result allows surveillance instead of diagnostic surgery. ThyroSeq v3 detects mutations and gene fusions associated with malignancy and provides positive predictive value to guide extent of surgery. Insurance coverage and cost ($3000–5000) limit universal use.

Leave a Reply

Your email address will not be published. Required fields are marked *