In short
- A single contrast-enhanced CT delivers 15 to 45 g of iodine, with free iodide at 30 to 300 times the recommended daily intake1.
- Pooled prevalence of overt post-contrast hyperthyroidism across 30 studies is 0.1 % (95 % CI 0 to 0.6)2. It is rare.
- Relative risk estimates run from 1.16 to 2.50 depending on population and outcome definition345. There is no single number.
- The risk lives almost entirely in pre-existing thyroid disease: autoimmune thyroiditis, autonomous nodules, latent Graves disease.
- Per exposure, the two largest sex-stratified analyses point towards men, not women5. The female-predominant argument is about baseline vulnerability, not per-scan risk.
There is a version of this question that gets asked on every on-call shift, usually in the form of somebody standing in the doorway holding a request form. The patient has a thyroid history. Is the CT still safe?
The honest answer has two halves. The event is rare. And when it happens, it happens almost exclusively in people whose thyroid was already not behaving.
What a CT actually delivers
An iodinated contrast injection is a pharmacological iodine load of an order the thyroid never meets otherwise. A single examination carries roughly 15 to 45 grams of iodine, and the free iodide fraction alone reaches 30 to 300 times the recommended daily intake1. For comparison, the daily requirement is measured in micrograms.
Two mechanisms follow, and they point in opposite directions. In autoimmune thyroiditis, the gland fails to escape from the Wolff-Chaikoff effect and hormone synthesis stays suppressed, producing hypothyroidism. In an autonomous nodule or latent Graves disease, the load feeds unregulated synthesis instead, the Jod-Basedow phenomenon, producing hyperthyroidism. Both are dose-dependent at the level of free iodide1.
How often, in numbers
The most quotable figure is the smallest one. A 2021 meta-analysis of 30 studies put the pooled prevalence of overt post-contrast hyperthyroidism at 0.1 percent, with a confidence interval running from 0 to 0.62. If you scan a hundred patients, you will not see it.
The relative risks look more alarming and mean something different. A Taiwanese cohort reported adjusted hazard ratios of 1.46 for any thyroid dysfunction and 2.00 for new hypothyroidism3. A US case-control study found an adjusted odds ratio of 2.50 for incident overt hyperthyroidism4. A doubling of a very small absolute risk is still a very small absolute risk.
The scatter is itself the finding. These are different populations with different outcome definitions, and anyone quoting a single relative risk for post-contrast thyroid dysfunction is quoting one study rather than the literature.
The sex direction is not the one people assume
Iodinated contrast has several genuinely female-predominant risks. Hypersensitivity reactions run 1.5 to 3-fold higher in women, and extravasation odds are 1.4 to 2.1-fold higher. Thyroid dysfunction is not one of them, at least not per exposure.
The two largest sex-stratified analyses find men at higher per-exposure relative risk. The Veterans Health Administration cohort, roughly 4.25 million patients, reported an odds ratio of 1.42 in men against 1.16 in women5. That cohort is 92.9 percent male, so composition confounding prevents a clean claim in either direction, but it certainly does not support the opposite.
The argument that does survive is different and worth stating precisely. Autoimmune thyroid disease is 7 to 10-fold female-predominant6, and pre-existing thyroid dysfunction is itself the dominant risk factor for the post-contrast event. A given iodine load therefore meets more latent thyroid pathology in women. That is a population-level statement about baseline vulnerability, not a claim that an individual woman faces higher risk from her scan.
What follows for the request in your hand
Three things, and none of them is “cancel the CT”.
The event is rare enough that it does not outweigh a genuine indication. A missed pulmonary embolism or a missed abscess is a present-tense risk. Post-contrast hyperthyroidism is a 0.1 percent event that is usually self-limiting and, when it is not, treatable.
The people who need attention before the injection are a defined group rather than everybody: known or suspected hyperthyroidism, autonomous nodules, multinodular goitre, and patients from iodine-deficient regions. Which patients need a TSH beforehand, and who can go straight to the scanner, is exactly the question I built a decision tool for, based on the ESUR 2025 guidance and the ETA recommendations. It takes under a minute and it beats reconstructing the logic from memory at 22:00.
And the load itself is a variable, not a constant. A fixed contrast volume delivers more iodine per unit of distribution volume to a smaller patient with a higher fat fraction, which is measurable and which we published on7. Reducing unnecessary volume reduces the absolute number of these events proportionally, in both sexes.
Frequently asked
Can you give iodinated contrast to a patient with hyperthyroidism?
Manifest untreated hyperthyroidism is the one setting where elective contrast should be deferred and endocrine advice sought. In a genuine emergency the indication usually wins, with thyroid follow-up arranged afterwards. Treated or euthyroid patients are generally not a contraindication.
How common is thyroid dysfunction after CT contrast?
Pooled prevalence of overt post-contrast hyperthyroidism across 30 studies is 0.1 percent, with a confidence interval from 0 to 0.6 percent. Subclinical changes are more common but often transient.
Does everyone need a TSH before contrast-enhanced CT?
No, and routine screening of all patients is not recommended by current guidance. Screening is targeted at defined risk groups. The decision tool on this site works through which groups those are.
What is the Jod-Basedow phenomenon?
Iodine-induced hyperthyroidism. An autonomous nodule or a latent Graves gland uses a large iodine load to drive unregulated hormone synthesis, because it is not subject to normal regulatory feedback.
What is Wolff-Chaikoff escape failure?
A high iodine load transiently suppresses hormone synthesis in any thyroid. A normal gland escapes this suppression within days. An autoimmune-damaged gland may not, and the result is hypothyroidism.
How long after contrast does thyroid dysfunction appear?
Typically weeks to a few months rather than hours. That delay is why the association was under-recognised for a long time and why it rarely surfaces during the imaging episode itself.
References
- Lee SY, Rhee CM, Leung AM, Braverman LE, Brent GA, Pearce EN. A review: radiographic iodinated contrast media-induced thyroid dysfunction. J Clin Endocrinol Metab 2015;100:376–383. doi:10.1210/jc.2014-3292
- Bervini S, Trelle S, Kopp P, Stettler C, Trummler J, Maeder MT. Prevalence of iodine-induced hyperthyroidism after administration of iodinated contrast media: a systematic review and meta-analysis. Thyroid 2021;31:1020–1029. doi:10.1089/thy.2020.0583
- Kornelius E, Chiou JY, Yang YS, Peng CH, Lai YR, Huang CN. Iodinated contrast media increased the risk of thyroid dysfunction: a 6-year retrospective cohort study. J Clin Endocrinol Metab 2015;100:3372–3379. doi:10.1210/jc.2015-2329
- Rhee CM, Bhan I, Alexander EK, Brunelli SM. Association between iodinated contrast media exposure and incident hyperthyroidism and hypothyroidism. Arch Intern Med 2012;172:153–159. doi:10.1001/archinternmed.2011.677
- Inoue K, Guo R, Lee ML, et al. Iodinated contrast administration and risks of thyroid dysfunction: a retrospective cohort analysis of the U.S. Veterans Health Administration system. Thyroid 2023;33:230–238. doi:10.1089/thy.2022.0393
- Vanderpump MPJ. The epidemiology of thyroid disease. Br Med Bull 2011;99:39–51. doi:10.1093/bmb/ldr030
- Becker J, Huber A, Bette S, et al, Decker JA. Are we systematically overdosing women? Revisiting standardized contrast protocols for thoracoabdominal CT scans. Eur Radiol 2025;35:3729–3738. doi:10.1007/s00330-024-11329-8
Guidance referred to: ESUR Guidelines on Contrast Agents v10.0 (2025); ACR Manual on Contrast Media v2024.
