Understanding Thyroid Cancer Imaging: A Patient's Guide to CT and MRI Scans

Table of Contents

• Key Points
• Why This Research Matters
• Types of Thyroid Cancer and Risk Factors
• How CT and MRI Are Performed
• Incidental Thyroid Nodules: Unexpected Findings
• Evaluating Lymph Node Spread
• Preoperative Imaging for Invasive Disease
• Evaluating for Recurrence After Treatment
• What This Means for Patients
• Study Limitations
• Recommendations for Patients
• Frequently Asked Questions
• Source Information

Key Points

• Incidental thyroid nodules are found in up to 1 in 6 neck CT scans, but only 0-9% are malignant.
• In a Japanese study, no deaths occurred over 10 years in patients with non-aggressive small thyroid cancers who received no treatment.
• Iodinated CT contrast can delay radioiodine treatment by 2-6 months, so MRI is a useful alternative.
• Up to 61% of lymph node metastases are smaller than 10 mm, so size alone is not a reliable predictor.
• A negative whole-body iodine scan occurs in 50-80% of patients with recurrent or persistent thyroid cancer.

Why This Research Matters

If you've been told a thyroid abnormality was found on a CT or MRI scan, you're not alone. While ultrasound is the standard first test for a lump you can feel (a palpable thyroid nodule) or a known thyroid cancer, thyroid problems are frequently discovered first on CT and MRI scans done for entirely different reasons. This review article, published in Cancer Imaging in 2013, explains how radiologists approach four common situations:

1. Finding an unexpected (incidental) thyroid nodule
2. Evaluating whether thyroid cancer has spread to lymph nodes
3. Pre-surgical imaging when cancer appears to be invading nearby structures
4. Checking for cancer recurrence after treatment

The authors — radiologists from Duke University, the University of Pittsburgh, Yale School of Medicine, the University of Melbourne, and the University of California San Francisco — emphasize that understanding how thyroid cancer behaves is essential for interpreting imaging correctly. This article translates their findings for patients and families navigating a thyroid cancer diagnosis.

Types of Thyroid Cancer and Risk Factors

Thyroid cancer is far more common than it used to be. The incidence is estimated at 37,000 new cases per year in the United States, and this number has more than doubled over the last 30 years. Much of this increase is attributed to the discovery of incidental nodules on imaging rather than new environmental causes.

There are four main types of thyroid cancer, and their behavior varies dramatically:

• Papillary thyroid carcinoma (88% of cases): The most common type, generally slow-growing and highly treatable.
• Follicular thyroid carcinoma (8% of cases): Includes the Hurthle cell variant.
• Medullary thyroid carcinoma (1% of cases): Arises from neuroendocrine C cells that produce calcitonin.
• Anaplastic thyroid carcinoma (1% of cases): An aggressive, undifferentiated tumor that typically occurs in older adults.

Other primary cancers of the thyroid — such as squamous cell carcinoma, sarcoma, and lymphoma — are extremely rare, together accounting for less than 1% of cases.

Papillary and follicular carcinomas are grouped together as differentiated thyroid carcinomas (DTC) because they retain many features of normal thyroid tissue. Both have excellent prognoses, with 10-year survival rates greater than 95% for papillary and 85% for follicular carcinoma. Medullary thyroid carcinoma also has a favorable outlook, with a 75% survival rate at 10 years. In stark contrast, anaplastic carcinoma has a devastating 5-year survival rate of only 7%.

Small papillary cancers are particularly noteworthy for their indolent (slow-moving) behavior. Epidemiological studies show that despite increased diagnosis of small thyroid cancers, survival rates have not improved — suggesting many of these cancers would never have caused harm. A Japanese study showed no deaths over 10 years in patients with non-aggressive small thyroid carcinomas who received no treatment at all.

Risk factors differ by cancer type. Papillary carcinoma is associated with ionizing radiation exposure, especially:

• Childhood head and neck radiation therapy
• Total body irradiation for bone marrow transplantation

Family history matters for both differentiated and medullary types. About one-quarter of medullary thyroid carcinoma cases are linked to familial medullary thyroid carcinoma (FMTC), caused by inherited mutations in the RET proto-oncogene. When FMTC occurs with tumors of other endocrine glands, it's called multiple endocrine neoplasia (MEN). Other familial syndromes associated with MTC include Cowden syndrome, familial polyposis, Carney complex, and Werner syndrome. In developing countries, follicular and anaplastic carcinomas have been linked to diets low in iodine.

How CT and MRI Are Performed

If you have known thyroid cancer, your doctors think carefully before ordering a contrast CT scan. This is because the free iodide load in contrast material interferes with iodine uptake in the thyroid for at least 6 to 8 weeks. For patients with differentiated thyroid cancer, this can delay diagnostic thyroid scans and radioiodine ablation therapy for 2 to 6 months, depending on the institution's policies. MRI contrast (gadolinium) does not interfere with iodine uptake, making MRI a valuable alternative in many cases.

The imaging protocols used at the authors' institutions are standardized:

• CT: Multidetector imaging from the skull base to the tracheal bifurcation (the point where the windpipe splits), with or without contrast. Images are reconstructed in 2-mm slices in three planes: axial (top to bottom), coronal (front to back), and sagittal (side to side).
• MRI: Similar coverage from skull base to tracheal bifurcation, including axial and coronal T1-weighted images, fat-suppressed T2-weighted images, and post-contrast axial and coronal T1-weighted images.

Communication between the radiologist and referring clinician is essential before scanning — particularly to determine whether contrast is safe and appropriate for each patient's situation.

Incidental Thyroid Nodules: Unexpected Findings

With the increased use of CT and MRI, incidental thyroid nodules (ITNs) — sometimes called "thyroid incidentalomas" — have become a growing challenge. These are nodules found on scans performed for unrelated reasons. They are common, appearing in up to 1 in 6 CT studies of the neck.

The dilemma for radiologists is real. In the absence of obvious local invasion or a hot spot on FDG-PET (fluorodeoxyglucose positron emission tomography), there are no CT or routine MRI features that reliably identify which nodules are malignant. Some studies have shown value in adding diffusion-weighted imaging to neck MRI because benign nodules have a higher "apparent diffusion coefficient" value, but ultrasound remains the preferred test for further evaluation.

There are strong arguments against automatically working up every small incidental nodule with ultrasound:

• The malignancy rate in incidental nodules is low, ranging from 0% to 9%
• The prognosis for malignancy is excellent — many patients die with their thyroid cancer, not from it

Yet radiologists worry about missing cancer. Many struggle with balancing cost-effectiveness against the fear of overlooking a potential tumor.

Because no official guidelines existed for reporting ITNs on CT and MRI, some authors suggested borrowing from ultrasound criteria and using a size cut-off of 10 mm or 15 mm to decide which nodules need further testing. The most common method is a 10-mm cut-off — but this approach has a major drawback: up to 78% of incidentally detected nodules on CT would meet that threshold and require ultrasound, creating enormous numbers of unnecessary follow-up tests.

Dr. Hoang and colleagues proposed a better strategy — a 3-tiered risk categorization system based on nodule size combined with features of advanced disease and patient age. Here's how it works:

High-risk history includes: thyroid cancer in one or more first-degree relatives; history of external beam radiation as a child; exposure to ionizing radiation in childhood or adolescence; previous removal of one thyroid lobe (hemithyroidectomy) with discovery of thyroid cancer; MEN2/FMTC-associated RET proto-oncogene mutation; or calcitonin level above 100 pg/ml.

This system was tested in a study of 133 CT-detected incidental nodules at a single institution, using the National Cancer Institute's SEER database for comparison. The results were striking: compared with the old 10-mm size cut-off, the 3-tiered system using a 15-mm nodule size threshold identified almost half the number of nodules for workup while capturing the same proportion of cancers — with no difference in the high-mortality cancers missed.

Evaluating Lymph Node Spread

Papillary carcinoma and medullary thyroid carcinoma are the types most likely to spread to lymph nodes. Nodal metastases are uncommon in follicular carcinoma. Sometimes, an enlarged lymph node is the first sign of thyroid cancer — patients may have no visible thyroid abnormality at all. In this situation, any thyroid nodule seen on CT or MRI should be considered suspicious and evaluated further with ultrasound.

It's also important to know that some thyroid primaries may be completely invisible on CT and MRI because they are small, diffuse, or multifocal (scattered in multiple areas of the gland). One illustrative case involved a 58-year-old man who presented with large cystic neck masses; his thyroid looked entirely normal on CT and even on ultrasound, yet surgery revealed multifocal papillary carcinomas ranging from 0.1 cm to 1.4 cm in the isthmus and both lobes.

Findings that suggest a neck mass originated from the thyroid include:

• Cystic components (fluid-filled areas)
• Calcifications
• Intense enhancement after contrast
• Proteinaceous or hemorrhagic content, appearing hyperdense on CT and bright (hyperintense) on T1-weighted MRI

A critical warning from the authors: cystic neck masses in young adults should not be dismissed as congenital cysts. They should be considered squamous cell carcinoma or thyroid carcinoma until proven otherwise. In one case, a 19-year-old woman had a cyst that looked simple and harmless on CT but was actually a level IV nodal metastasis from a 1-cm papillary carcinoma in her right thyroid lobe.

Where Does Thyroid Cancer Spread?

Thyroid nodal metastases commonly occur in the central compartment (level VI) and lateral nodal groups (levels II–IV). The Delphian node (also called the prelaryngeal lymph node) is the highest lymph node in the central compartment. Involvement of this node in papillary thyroid cancer is a powerful predictor: patients with a Delphian node metastasis are nine times more likely to have lateral nodal involvement.

Other sites that should not be neglected include:

• Lower paratracheal nodes in the superior mediastinum (level VII)
• Retropharyngeal nodes (behind the throat)
• Retroesophageal groups (behind the esophagus)

One case highlighted how subtle retropharyngeal nodes can be: a 52-year-old woman with papillary carcinoma had a metastatic right retropharyngeal node that was nearly invisible on standard CT windowing — only noticeable when the radiologist narrowed the window width to improve contrast.

The AJCC/UICC TNM staging system classifies nodal stage by location:

• N1a: Level VI nodal disease (including pretracheal, paratracheal, and Delphian nodes)
• N1b: Unilateral or bilateral lateral cervical nodes, or superior mediastinal nodes

Superior mediastinal involvement may prevent curative surgery, so CT or MRI may be ordered when predictors of mediastinal disease exist — such as lateral nodes or a primary tumor larger than 1.5 cm.

A crucial point for radiologists: abnormal nodal morphology (shape, internal characteristics) is a better predictor of metastatic disease than size alone, because up to 61% of nodal metastases are smaller than 10 mm — below the traditional size threshold for concern. Additionally, skip metastases (discontinuous nodal spread that bypasses the central compartment) occur in up to 21% of medullary thyroid carcinoma cases.

Preoperative Imaging for Invasive Disease

For most thyroid cancers — with the exception of most anaplastic carcinoma cases — treatment involves total or near-total thyroidectomy, central nodal resection, and possibly radioiodine ablation. Small tumors under 10 mm may be treated with lobectomy alone (removing just one lobe), but if the small tumors are multifocal, patients still need total thyroidectomy and radioiodine ablation.

Here's an important limitation: CT and MRI cannot reliably diagnose multifocal disease or determine the histology (cell type) of thyroid cancer. That's why preoperative evaluation starts with ultrasound to detect multifocal disease and lymphadenopathy. CT and MRI are added when local invasion is suspected.

Why does invasion matter so much? Locally invasive cancer may require:

• More extensive surgery, such as laryngectomy (removal of the voice box)
• Additional surgical specialists, such as thoracic or reconstructive plastic surgeons
• A decision not to operate if the cancer is too extensive

Local invasion is also a key component of the AJCC/UICC tumor (T) staging system. The system focuses on four groups of structures:

• The airway and nerves centrally (trachea, esophagus, larynx, and recurrent laryngeal nerve)
• The carotid arteries laterally
• The prevertebral space posteriorly
• The mediastinum inferiorly

Note: Patients with differentiated thyroid cancer younger than 45 years can only be staged as stage I or II (stage II means metastatic disease). For patients 45 and older with DTC, and for MTC patients of any age, distant metastasis (M1) means stage IVc. All anaplastic carcinomas are by definition T4a or T4b, so all are stage IV.

How Accurate Are CT and MRI for Detecting Invasion?

MRI and CT have similar accuracy for predicting invasion of the esophagus, trachea/larynx, and recurrent laryngeal nerve. The key numbers from several retrospective studies:

The main sign for tracheal and esophageal invasion on both MRI and CT is a mass contacting 180° or more of the organ's circumference. Other findings suggesting tracheal invasion include deformity of the lumen, focal mucosal irregularity or thickening, and an intraluminal mass. The esophagus is harder to evaluate than the trachea because it doesn't naturally contain air to outline its interior. On MRI, the most suspicious finding for esophageal invasion is a focal T2 signal in the outer layer of the esophageal wall. On CT, loss of the normal esophageal wall and lumen is the key clue.

Invasion of the recurrent laryngeal nerve (RLN) — the nerve controlling the vocal cords — can be predicted by effaced fatty tissue in the tracheoesophageal groove where the nerve travels. Other imaging signs include vocal cord dysfunction and 25% or more of the primary tumor's circumference abutting the capsule at the posterior portion of the thyroid (the "sign of posterior extracapsular invasion").

Vascular and prevertebral space invasion are designated T4b disease and generally preclude curative surgery. Dr. Seo and colleagues found that contact of the tumor with 180° or more of the vessel circumference was a highly specific sign for invasion of the common carotid artery and internal jugular vein on CT. However, a different study of head and neck tumors found this sign had only 50% accuracy for arterial invasion due to many false negatives. That study found more accurate CT signs to be arterial compression or deformation, or fat/fascial plane deletion — with 84% accuracy. Increasing the circumferential encasement threshold to 270° improves specificity further; on MRI this sign had 100% sensitivity and 88% specificity.

For prevertebral muscle invasion, MRI can exclude involvement if the retropharyngeal fat is preserved. However, confirming invasion using findings like muscle T2 hyperintensity, enhancement, or contour abnormality is less reliable, with accuracy of only 60%.

CT and MRI also play a secondary but valuable role in detecting anomalous anatomy that could complicate surgery. One example is the non-recurrent inferior laryngeal nerve (NRILN), a variant where the nerve branches directly from the vagus nerve and enters the larynx instead of looping under the right subclavian artery. Radiologists can suspect a right NRILN when they see an aberrant right subclavian artery — a warning that helps surgeons avoid injuring the nerve.

Evaluating for Recurrence After Treatment

After treatment for differentiated thyroid cancer, serum thyroglobulin level is used as a tumor marker. When thyroglobulin rises, the standard next steps are a neck ultrasound and a whole-body iodine scan (using 131I or 123I).

Here's the challenge: the whole-body iodine scan is negative in 50–80% of patients who actually have recurrent or persistent disease. This situation represents progression to dedifferentiated thyroid cancer — the tumor cells have lost their ability to take up iodine, making them invisible on iodine scanning but potentially visible on other imaging.

In these cases, MRI or PET/CT can help locate the recurrence. MRI has two advantages:

• It can be performed safely without interfering with future iodine-based treatments, since MRI contrast is not iodinated
• It can detect nodal disease with high protein content from colloid, thyroglobulin, and blood products

The role of retropharyngeal nodal evaluation becomes even more critical after lateral and central neck dissections have been performed, as anatomy is altered and normal drainage pathways have been disrupted.

PET/CT has 81–82% sensitivity and 64–89% specificity for detecting recurrent tumor in patients with elevated thyroglobulin and a negative whole-body iodine scan. As tumors dedifferentiate, their tendency to take up FDG increases — but intense FDG uptake, while correlated with more aggressive disease, doesn't always mean a worse prognosis. A notable pitfall is that most thyroid nodal metastases are small and may fall below the resolution limits of PET/CT, producing false negatives.

Beyond regional lymph nodes, PET/CT can also detect unrecognized distant metastases in the lungs and bones — information that can change treatment from a curative approach to a palliative one if disease is too extensive.

For medullary thyroid carcinoma, the tumor markers are serum calcitonin and carcinoembryonic antigen (CEA). FDG-PET uptake can be helpful when positive, but FDG avidity in MTC is variable, making negative scans difficult to interpret.

What This Means for Patients

Several practical messages emerge from this review:

First, an incidental thyroid nodule is common and usually benign. The malignancy rate is only 0–9%, and even when cancer is present, survival rates for the common types exceed 85–95% at 10 years. The new 3-tiered reporting system helps avoid unnecessary procedures while still catching dangerous cancers.

Second, if you've been scheduled for a contrast CT and have known thyroid cancer, ask about timing. The iodine in CT contrast can block your thyroid's iodine uptake for 6–8 weeks, potentially delaying radioiodine treatment by 2–6 months. MRI with gadolinium contrast does not have this problem.

Third, if you present with a neck mass that turns out to be a cystic lymph node, thyroid cancer must be considered — even in young adults, even if the thyroid looks normal on imaging. Surgical specimens frequently reveal small multifocal cancers that were invisible on scans.

Fourth, CT and MRI are powerful tools for surgical planning. They determine whether cancer has invaded the trachea, esophagus, nerve, blood vessels, or prevertebral space — information that guides whether surgery is feasible and which surgical team should be involved.

Fifth, after treatment, a rising thyroglobulin level with a negative iodine scan requires further evaluation with MRI or PET/CT. A negative iodine scan does not mean you're cancer-free.

Study Limitations

This is a review article, meaning the authors analyzed and summarized existing research rather than conducting a new clinical trial. Several limitations of the underlying research should be acknowledged:

• Most accuracy data come from retrospective studies, which are more prone to bias than prospective trials.
• CT and MRI cannot reliably determine tumor histology (cell type) or diagnose multifocal disease — small cancers, diffuse growth patterns, and multifocal tumors are frequently invisible on cross-sectional imaging.
• Size criteria are imperfect. Up to 61% of nodal metastases are smaller than 10 mm, meaning normal-sized lymph nodes can still harbor cancer.
• Vascular invasion assessment is challenging. Depending on the criteria used, accuracy for arterial invasion ranged from as low as 50% to 84–88% on MRI with 270° encasement threshold.
• The 3-tiered reporting system was validated in a single institution's cohort of 133 nodules plus a national database; wider validation in diverse practice settings would strengthen confidence in its generalizability.
• The review was published in 2013; imaging technology and guidelines have continued to evolve since that time.

Recommendations for Patients

Based on this research, here are actionable steps patients can take:

1. Ask about the contrast agent. If you have known or suspected thyroid cancer and are scheduled for a CT scan, discuss with your doctor whether iodinated contrast is necessary or whether MRI (gadolinium-based contrast) could be used instead — particularly if radioiodine therapy is planned in the near future.
2. Don't panic about an incidental nodule. Most incidental thyroid nodules are benign, and even malignant ones are often low-risk. Ask your radiologist or doctor which risk category your nodule falls into based on the 3-tiered system described above.
3. Follow through on recommended ultrasound. If your report suggests ultrasound for further characterization, schedule it. Ultrasound provides the most detailed information about nodule features that predict cancer.
4. If you have a cystic neck mass — especially if you're a young adult — insist that thyroid cancer be ruled out before assuming it's a congenital cyst.
5. For patients with known thyroid cancer: understand that CT and MRI play a complementary role to ultrasound. They help detect invasion and nodal spread in areas ultrasound can't fully evaluate, such as the mediastinum (area behind the breastbone) and retropharyngeal space.
6. After treatment, keep up with thyroglobulin monitoring. If your thyroglobulin is rising but your iodine scan is negative, talk to your specialist about MRI or PET/CT to locate potential recurrence or metastasis.

These steps reflect the central message of this review: thoughtful, informed use of CT and MRI can spare patients unnecessary procedures while catching the cancers that truly matter.

Frequently Asked Questions

What is an incidental thyroid nodule found on a CT or MRI scan?

An incidental thyroid nodule is a growth in your thyroid found on a scan done for another reason. These are common, appearing in up to 1 in 6 neck CT scans. Most are benign, with cancer found in only 0% to 9% of cases, so finding one does not mean you have cancer.

Should I worry if a thyroid nodule is found by chance on a CT scan?

Most incidental thyroid nodules are benign, and even cancers are often slow-growing with excellent survival, exceeding 95% at 10 years for the common type. Your doctor may use a risk system based on size, your age, and scan features to decide if ultrasound or biopsy is needed, avoiding unnecessary tests.

Why might my doctor choose MRI instead of a contrast CT scan for thyroid cancer?

Iodinated CT contrast can block your thyroid's iodine uptake for 6 to 8 weeks, delaying radioiodine scans or treatment by 2 to 6 months. MRI contrast, called gadolinium, does not interfere with iodine uptake, so MRI is often a valuable alternative when iodine-based treatment is planned soon.

I have a cystic neck mass. Could it be thyroid cancer even if my thyroid looks normal?

Yes. Cystic neck masses in young adults should not be dismissed as harmless congenital cysts. In one case, a 19-year-old woman's simple-looking cyst was actually a lymph node metastasis from a 1-cm papillary thyroid carcinoma. Sometimes small or multifocal thyroid cancers are invisible on CT, MRI, or even ultrasound.

After thyroid cancer treatment, my thyroglobulin is rising but my iodine scan is negative. What does that mean?

A negative whole-body iodine scan occurs in 50% to 80% of patients who actually have recurrent or persistent disease. This can mean the cancer has lost its ability to take up iodine. In this situation, MRI or PET/CT can help locate the recurrence, so a negative iodine scan does not mean you are cancer-free.

How accurate are CT and MRI at detecting if thyroid cancer has spread to nearby structures?

CT and MRI have similar accuracy for predicting invasion of the esophagus, trachea/larynx, and recurrent laryngeal nerve. For example, MRI sensitivity for tracheal invasion was 100% and CT was 59% in one study, while specificity was 84% and 91%, respectively. Both use a 180-degree contact sign as a key indicator.

What is the chance an enlarged lymph node is thyroid cancer if my thyroid looks normal on scans?

Sometimes an enlarged lymph node is the first sign of thyroid cancer. In this situation, any thyroid nodule seen on CT or MRI should be treated as suspicious and checked with ultrasound. One patient had a normal thyroid on CT and ultrasound, yet surgery found multifocal papillary cancers from 0.1 cm to 1.4 cm.

My thyroid nodule was found on a CT scan. Should I get a second opinion on the imaging before deciding on biopsy or surgery?

A second opinion can help you interpret incidental thyroid nodule findings. CT and MRI cannot reliably determine if a nodule is cancerous or detect multifocal disease. Most incidental nodules are benign; malignancy rates are only 0–9%. A second opinion may clarify whether ultrasound-guided biopsy is needed or whether observation is safe. If surgery is planned for invasive disease, expert imaging review can confirm whether the trachea, esophagus, or nerves are involved, affecting surgical approach. Also, MRI may be preferable to contrast CT if radioiodine treatment is anticipated. Diagnostic Detectives Network provides independent expert second opinions.

Source Information

Original article title: Imaging of thyroid carcinoma with CT and MRI 2013

Authors: Jenny K. Hoang, Barton F. Branstetter IV, Andreia R. Gafton, Wai K. Lee, Christine M. Glastonbury

Journal: Cancer Imaging (2013) 13(1), 128–139

DOI: 10.1102/1470-7330.2013.0013

Publication date: Accepted for publication 18 January 2013

Note: This patient-friendly article is based on peer-reviewed research. It is intended for educational purposes and does not replace professional medical advice. Always discuss imaging findings and treatment options with your healthcare team.