Abstract
Background:
In 2013, the American Thyroid Association (ATA) issued a “Policy Statement on Thyroid Shielding During Diagnostic Medical and Dental Radiology.” The recently updated National Council on Radiation Protection and Measurement Radiation Protection in Dentistry and Oral and Maxillofacial Imaging (NCRP Report No. 177) prompts this review of progress related to patient thyroid shielding since the ATA statement was published.
Summary:
Relevant publications appearing since the ATA statement were identified by querying PubMed for “thyroid and dental and (collar or shielding)” and substituting specific dental radiographic procedures in the search. The search was expanded by reviewing the cited papers in the PubMed-retrieved papers and by use of the Web of Science to retrieve papers citing the PubMed retrieved publications. Although many quantitative studies have appeared reflective of current dental radiographic instrumentation and practice, much more can be done to foster minimizing radiation to the thyroid.
Conclusions:
We list seven areas that should be pursued. Among them are harmonizing guidelines for the use of thyroid collars based on the recent studies and a comprehensive survey of current dental radiological practice patterns.
Introduction
In the early 1970s, when one of us (A.B.S.) began to give academic lectures on thyroid pathophysiology, the audience was asked whether they knew what a “goiter belt” was (“Prior to the 1920s, endemic iodine deficiency was prevalent in the Great Lakes, Appalachians, and Northwestern regions of the United States, a geographic area known as the ‘goiter belt’”) (1). If today I were to ask what a “thyroid collar” is, it is likely that most everyone in the audience would know that it is a barrier to shield the thyroid/neck that is used in visits to the dentist.
In 2013, after a controversy about the use of thyroid collars during mammography, the American Thyroid Association (ATA) issued a “Policy Statement on Thyroid Shielding During Diagnostic Medical and Dental Radiology” (2). The purpose of this article is to update the analyses and conclusions included in the 2013 Statement focusing on dental radiology. Mammography poses virtually no risk to the thyroid because of the low dose to the thyroid and the age range associated with its use. The rationale for an update at this time is the new report from the National Council on Radiation Protection (NCRP) on dental radiology and the changes in dental radiology reflected in it (3).
Methodology
Publications appearing since the ATA statement were identified by querying PubMed for “thyroid and dental and (collar or shielding)” and substituting specific dental radiographic procedures in the search. The search was expanded by reviewing the cited papers in the PubMed-retrieved papers and by use of the Web of Science to retrieve papers citing the PubMed retrieved publications.
Background
A meta-analysis of studies relating exposure to dental radiology to thyroid cancer was published in 2019 (4). Six case-control and one cohort study were included. The authors found that multiple episodes of dental radiology were associated with the risk of thyroid cancer. They did recognize that potential bias, especially recall bias, and the lack of thyroid dose estimates limited the strength of their conclusions. Most relevant to the current report, the exposures to dental diagnostic radiology occurred decades ago, when the thyroid doses were undoubtedly much higher than now.
Dental radiographic studies are divided into intraoral and extraoral methods, depending on where the image is formed. Intraoral, where the film is situated in the mouth, is the most common. Extraoral includes panoramic, cephalographic, and cone beam computed tomography (CBCT) images. The setup for extraoral imaging including CBCT (also referred to as a 3D) image is shown in Figure 1 (5). Clinical uses for these methods are summarized in Table 1.

Cone beam computer tomographic setup (5).
Categories of Dental Radiology
Adapted from Ref. (16).
An intraoral procedure includes multiple individual views. The dental estimates are for the U.S. population in 2016, as reported in Ref. (7). Effective doses, as defined in the text, not thyroid doses are shown.
As reported in Ref. (18).
CBCT, cone beam computed tomography.
There have been substantial secular changes in dental radiology. For countries such as the United States, the number of dental procedures has remained steady over many years. However, as a result of changes in the guidelines, the number of images per procedure has changed. The introduction and increasing use of CBCT has added to the total thyroid dose for those patients requiring this procedure. For the most frequently used procedure, intraoral imaging, there has been a dramatic decrease in the associated thyroid dose (6). In 2016, there were 296 million intraoral examinations performed in the United States, 21 million panoramic examinations, and less than 1 million cephalometric examinations (7). In addition, in 2014–2015 in the United States there were 3.7 million CBCTs performed in adults and 1.5 million in children. During 2017, among children between the ages of 2 and 17, 84.2% of boys and 85.6% of girls visited a dentist at least once (8). The latter statistics demonstrate that among children dental examinations occurred nearly once a year.
The methods of expressing radiation exposure in dental radiology are potentially confusing (9). Organ dose is the amount of radiation absorbed by an organ expressed as Gy (Gray). Equivalent dose modifies organ dose depending on the type of radiation used. For dental radiation, equivalent dose is equal to organ dose, but expressed as Sv (Sievert). Effective dose (Sv) is a protection quantity defined as the sum of the sex-averaged equivalent doses in all specified tissues and organs of a reference person weighted by the tissue weighting factor. The weighting takes the sensitivity of individual organs into account. It is not used to estimate cancer risk for individual exposed people. Effective organ dose (sometimes called radiation-weighted dose) is neither a standard nor a useful measure. (Presumably, it is the organ dose multiplied by the weighting factor for determining the effective dose.) Organ doses and equivalent doses are most often used by epidemiological studies whereas effective doses are used by radiation protection organizations, such as the International Atomic Energy Agency (IAEA). Unless stated otherwise, all doses mentioned in this article are thyroid-specific and we retain the units (Gy or Sv) used by the author(s).
Update
The summary of the ATAs 2013 Policy Statement was as follows: “Although the thyroid doses associated with dental x-rays have not been shown to cause thyroid cancer, it is prudent to reduce thyroidal radiation exposure as much as possible without compromising the clinical goals of dental examinations. The ATA thus endorses the recommendations of NCRP-145. However, the ATA also urges a reconsideration of the less stringent requirement for thyroid shielding in adults as compared to children. Adult risk for radiation-induced thyroid cancer may be less, but still merits efforts to reduce it, given that the use of shielding is safe and readily available. The ATA also recommends that efforts be made to encourage and monitor compliance with the ADA and NCRP guidelines and to reduce, as much as possible, the areas of ambiguity in them.”
When the ATA statement was published in 2013, based on a pooled analysis of multiple studies, it was possible to say that the lowest dose of external radiation to the thyroid administered during childhood definitively shown to confer a thyroid cancer risk was 100 mGy (10). A new pooled analysis including more studies and longer follow-up has demonstrated that the risk is demonstrable at doses as low as 50 mGy (11). In addition, no evidence for a threshold dose and no evidence for rejecting the linear extrapolation to even lower doses was found. This is in keeping with the consensus that the linear no-threshold model is applicable to radiation risks (12). Fifty milligray is in the range that may accumulate with multiple occurrences of certain dental radiographs (see the following four sections). These findings provide further support for following the ALARA (as low as reasonably achievable) principle in all facets of dental radiology (13,14).
In 2016, the average annual non-therapeutic medical effective radiation dose in the United States was 2.16 mSv (15). Non-CBCT radiographs contribute 0.25% to this dose, but the contribution of CBCT is not known. Table 1 shows the estimated effective doses of four categories of dental radiology in the U.S. and compares them with transcontinental airplane travel (7,16 –18). The IAEA reported dose range for CBCT (50–100 μSv) is broad, because it depends on the field of view (FOV). The range is similar to the one reported for CBCT by Abdelkarim (19). The IAEA data resulted from an international survey and, therefore, reflect an unknown combination of digital and analogue image capture methods. In the table, the effective doses are shown as an indication of the relative total doses among the four categories, specifically the high dose from CBCT procedures. Since, by 2014–2015, nearly 90% of dental facilities in the United States used digital imaging (20), for the following summary of thyroid-specific doses we have, where possible, relied on publication from the past 10 years that would reflect digital imaging.
Intraoral radiographs
Granlund et al. (21) estimated that the thyroid dose from a set of 4 intraoral images is 8 μGy and from a set of 18 images it is 53 μGy.
Panoramic radiographs
In five studies of panoramic dental images, the thyroid absorbed dose, although variable, was less than 300 μGy (21 –25).
Cephalometric radiographs
In a report in 2015, the thyroid dose from a lateral cephalometric exam was 30.7 μSv (26). A publication reviewing earlier reports found a range of doses of 5–57 μGy (27).
Cone beam computed tomography
Among dental radiographic procedures, CBCT has the potential for the highest thyroid gland exposure to radiation. The “large exposure ranges make CBCT doses difficult to generalize,” as demonstrated in the 2015 report by Ludlow et al. (28). They estimated the thyroid dose ranges from 10 CBCT units and a large number of FOV, the latter divided into large, medium, and small. For 16 large FOVs, the median thyroid dose was 504 μSv (central 80% 171–2548 μSv; minimum 50 μSv; maximum 6333 μSv); for medium FOV, the median was 434 μSv (central 80% 234–1680 μSv; minimum 70 μSv; maximum 2700 μSv); and for small FOV, the median was 143 μSv (central 80% 36–333 μSv; minimum 1 μSv; maximum 403 μSv). The regression between the vertical dimension and thyroid dose was positive, but not statistically significant (not shown). Concern about CBCT radiation has led to additional publications with dose estimates similar to those of Ludlow et al. (22,25,29 –35). There are two lessons: CBCT is of the greatest concern among dental imaging modalities, and a dental practitioner will have difficulty estimating the dose from a CBCT procedure for an individual patient.
For the following summary of thyroid collar effectiveness, wherever possible, we have relied on publication in the past 10 years.
Intraoral radiographs
The use of rectangular, as opposed to circular, collimation reduces the thyroid dose (as well as the effective dose) from intraoral examinations (36,37). Combined with rectilinear collimation, a thyroid collar reduces the thyroid dose by 27% to 59% depending on which of six rectangular collimators was tested (37).
Panoramic radiographs
Data on the effectiveness of thyroid collars in panoramic radiographs are sparse. A study of four units, two manufactured in Finland and two in Germany, found thyroid dose reductions of 9.8%, 9.65%, 19.3%, and 23.7%, with the latter two being statistically significant (38). In a study using an Italian unit, Hafezi et al. tested several collar configurations (39). They concluded that a frontal collar with an area of at least 300 cm2, but not much larger was best, with a thyroid dose reduction of 56%.
Cephalometric radiographs
There are little data for thyroid shielding in cephalometric radiographs. For orthodontic lateral cephalography, Hoogeveen et al. designed and created a thyroid protector (“a cephalographic thyroid protector”) that leaves the cervical vertebrae exposed. It reduced the thyroid dose by 85% compared with no shielding (26).
Cone beam computed tomography
Recently, Pauwels et al. summarized the results of thyroid shielding reported in seven publications, six of them from 2012 and later (40). On average, 45.9% of the thyroid dose was reduced by the shielding. Given the wide spectrum of CBCT exams, the challenge is to know when using shielding will interfere with the exam.
Summary and Perspectives
Since 2013, the year of the ATA policy statement, there continues to be a robust interest in the use of thyroid shielding in all realms of dental radiology. Emblematic of this is the recent 2019 NCRP report, which expands recognition of the value of thyroid shielding to all ages. Specifically, the recent NCRP Report No. 177 (3) states: “Thyroid shielding shall (emphasis in original) be provided for patients when it will not interfere with the examination.” The 2012 American Dental Association guidelines and the Image Gently statement make similar statements (13,41). The position of these three organizations can be translated to the conclusion that the use of thyroid shielding should be the default decision for all dental radiology. This is in accord with the ALARA principle and is in keeping with the ATA statement that thyroid collars should be used at all ages (2). There are circumstances where thyroid shielding is arguably superfluous. An example would be a senior citizen getting routine intraoral imaging as part of a preventive dental evaluation. Still, shielding in such instances would reduce thyroid exposure and serve as an important standard across the entire age span. With regard to dental radiographs during pregnancy, both the American Dental Association (ADA) and the American Congress of Obstetricians and Gynecologists indicate that dental radiographs during pregnancy are safe and should be performed with thyroid shielding and abdominal shielding (with a lead apron) (42). However, some have questioned the use of lead aprons in a recent editorial (43).
Although the default should be using a thyroid collar, the NCRP does not define when collars could interfere with an examination. In addition, not all organizations are in agreement about when to use a collar. Notably, in its 2020 guidance, the British Institute of Radiology states “The use of patient contact shielding is not recommended (emphasis in original) for dental radiography for the majority of imaging situations” (44).
In a comprehensive literature review of pediatric dental radiology, Van Acker et al. (45) provide some clarity about CBCT images, as follows: “The use of a thyroid shield should be routinely used except in two situations: first, when the CBCT examination is intended to image structures below, or very close to, the axial level of the top of the shield, in which situation artifacts from the shield might affect the quality, and second, when using tube current modulation during the scan, giving real-time feedback from the detector to the exposure control. In case of using automatic exposure control based on scout images, the thyroid shield should be positioned only after the scout images have been taken.” Similarly, Pauwels et al. (40) suggest: “Thyroid shielding should not be used [in CBCT] when visualization of tissue below, or slightly above, the axial level of the top of the shielding is needed.” Further, “Thyroid shielding should be routinely used for children, regardless of the position of the FOV, and can be beneficial for thyroid protection of adults up to the age of 50.” Relying on data from 2004, Van Acker et al. concluded that thyroid collars are “contraindicated” for panoramic images (45). However, specially designed shielding can achieve significant thyroid dose reductions, as cited earlier (39). Sansare et al. showed that a thyroid collar could be used in cephalometric imaging as long as the image was not performed to assess skeletal maturity, that is, an FOV extending below the second cervical vertebra generally not needed for dental imaging (27). Van Acker et al. adopted this recommendation (45).
What is apparent is that both national and international organizations with a professional interest in radiation protection in dental radiology need to harmonize the recommendation about the use of thyroid collars. Discordant and conflicting guidelines should be resolved as clearly as possible.
As a prerequisite to harmonization, there needs to be more information about how practicing dentists are using thyroid shielding. Although there is relatively little information about this, recent publications indicate that this goal is important and achievable. In 2020, Gillies et al. reported the responses from 1332 Canadian dentists from a survey of 9052 Canadian dentists, a response rate of 14.7% (46). For intraoral images, 86.9% respondents reported using thyroid collars. No information was requested for the use of thyroid collars in extraoral imaging. Also in 2020, Campbell et al. reported responses from 1124 pediatric dentists in the United States from a survey of 7087 (response rate 16%). Sixty-eight percent reported that they always used thyroid collars and lead aprons (47). In 2019, Alzamzami et al. reported a survey of U.S. endodontists about their use of CBCT, but they did not ask about radiation exposure precautions in general or thyroid collars specifically (48). In an article “Has CBCT Become Too Much of a Good Thing?” Valanchovic describes a hypothetical, but realistic, situation where the use of CBCT in general practice, rather than in a dental school setting, could result in greater radiation exposure than necessary (49). Although education in CBCT use, especially during dental training, has been emphasized, it appears that no well-structured efforts to comprehensively determine the extent of thyroid shielding with CBCT have been conducted.
Knowing when and how dentists use thyroid shielding would be an important step leading to wider implementation. The ATA and others should continue to advocate for studies measuring the adoption of dental radiology radiation safety methods, especially thyroid shielding, in clinical practice.
Improved methodology of thyroid shielding should be pursued. This should include development and use of age-specific thyroid collars. Determining the thyroid's position before performing a radiologic exam should be evaluated (50). Fakhoury et al. have highlighted variability in thyroid shields, suggesting that standardization would be valuable (51). Sutanto et al. described a novel thyroid shield with the potential of only very small artifacts in CT images (52). Each potential innovation in dental radiology should be evaluated for its effect on thyroid exposure.
As part of its response to the media attention to dental (and mammographic) radiology in 2013, the ATA became a “Collaborating Organization” of the NCRP and a member of the Image Gently Campaign. The ATA should continue to monitor activities as they relate to protecting the thyroid.
Finally, the ATA should participate in public education about safety in dental radiology, as emphasized by an editorial written by the leadership of the Image Gently Campaign (13).
Footnotes
Acknowledgments
The authors thank Byron Kitahara, DDS and David Borrego, PhD for valuable comments during the preparation of this article.
Authors' Contributions
Study conception and initial review of the literature: A.B.S. Interpretation of the data and article preparation and approval: all authors.
Author Disclosure Statement
Funding Information
No funding was received for this article.
